117
A fault of default Local and global default in Hebrew feminine verbs Outi Bat-El Tel-Aviv University [email protected] www.outibatel.com Allomorphy The Hebrew University June 2014

A fault of default Local and global default in Hebrew feminine verbs

  • Upload
    norah

  • View
    64

  • Download
    0

Embed Size (px)

DESCRIPTION

Allomorphy  The Hebrew University  June 2014. A fault of default Local and global default in Hebrew feminine verbs. Outi Bat-El Tel-Aviv University [email protected]  www.outibatel.com. Default. A llomorphy. Given two or more surface allomorphs, which one is the default? - PowerPoint PPT Presentation

Citation preview

Page 1: A fault of default Local and global default in Hebrew feminine verbs

A fault of defaultLocal and global default

in Hebrew feminine verbs

Outi Bat-ElTel-Aviv Universityobatelposttauacil wwwoutibatelcom

Allomorphy The Hebrew University June 2014

2

DefaultAllomorphy

Given two or more surface allomorphs which one is the default The answer is often trivial

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Two types of allomorphy - two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic)

3

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

Two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic) Non-phonologically-conditioned allomorphy ndash the default is NOT

irregular amp NOT limited in distribution ()

4

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Information to the contrary specific [X]sg [X-en]pl X = ox hellip No information default [X]sg [X-z]pl

The default allomorph appears whenever the specific one fails ndash it is the elsewhere case

English Plural

Reiter R 1980 A logic for default reasoning Artificial Intelligence 1381-132Kiparsky Paul 1973 lsquoElsewherersquo in phonology In SR Anderson and P Kiparsky (eds) A Festchrift for Morris Halle 93-106 Hold Rinehart and Winston 5

Default X In the absence of any information to the contrary assume X

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 2: A fault of default Local and global default in Hebrew feminine verbs

2

DefaultAllomorphy

Given two or more surface allomorphs which one is the default The answer is often trivial

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Two types of allomorphy - two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic)

3

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

Two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic) Non-phonologically-conditioned allomorphy ndash the default is NOT

irregular amp NOT limited in distribution ()

4

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Information to the contrary specific [X]sg [X-en]pl X = ox hellip No information default [X]sg [X-z]pl

The default allomorph appears whenever the specific one fails ndash it is the elsewhere case

English Plural

Reiter R 1980 A logic for default reasoning Artificial Intelligence 1381-132Kiparsky Paul 1973 lsquoElsewherersquo in phonology In SR Anderson and P Kiparsky (eds) A Festchrift for Morris Halle 93-106 Hold Rinehart and Winston 5

Default X In the absence of any information to the contrary assume X

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 3: A fault of default Local and global default in Hebrew feminine verbs

Two types of allomorphy - two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic)

3

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

Two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic) Non-phonologically-conditioned allomorphy ndash the default is NOT

irregular amp NOT limited in distribution ()

4

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Information to the contrary specific [X]sg [X-en]pl X = ox hellip No information default [X]sg [X-z]pl

The default allomorph appears whenever the specific one fails ndash it is the elsewhere case

English Plural

Reiter R 1980 A logic for default reasoning Artificial Intelligence 1381-132Kiparsky Paul 1973 lsquoElsewherersquo in phonology In SR Anderson and P Kiparsky (eds) A Festchrift for Morris Halle 93-106 Hold Rinehart and Winston 5

Default X In the absence of any information to the contrary assume X

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 4: A fault of default Local and global default in Hebrew feminine verbs

Two types of default Phonologically-conditioned allomorphy ndash the default is identical to

the underlying representation (basic) Non-phonologically-conditioned allomorphy ndash the default is NOT

irregular amp NOT limited in distribution ()

4

English Plural

-en Identity -oslash -z

-ɨz -s -z

children sheep buses cats dogs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Information to the contrary specific [X]sg [X-en]pl X = ox hellip No information default [X]sg [X-z]pl

The default allomorph appears whenever the specific one fails ndash it is the elsewhere case

English Plural

Reiter R 1980 A logic for default reasoning Artificial Intelligence 1381-132Kiparsky Paul 1973 lsquoElsewherersquo in phonology In SR Anderson and P Kiparsky (eds) A Festchrift for Morris Halle 93-106 Hold Rinehart and Winston 5

Default X In the absence of any information to the contrary assume X

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 5: A fault of default Local and global default in Hebrew feminine verbs

Information to the contrary specific [X]sg [X-en]pl X = ox hellip No information default [X]sg [X-z]pl

The default allomorph appears whenever the specific one fails ndash it is the elsewhere case

English Plural

Reiter R 1980 A logic for default reasoning Artificial Intelligence 1381-132Kiparsky Paul 1973 lsquoElsewherersquo in phonology In SR Anderson and P Kiparsky (eds) A Festchrift for Morris Halle 93-106 Hold Rinehart and Winston 5

Default X In the absence of any information to the contrary assume X

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 6: A fault of default Local and global default in Hebrew feminine verbs

Question addressedWhich of the two FMSG suffixes appearing in present tense verbs in Hebrew is the default -et or -a

6

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 7: A fault of default Local and global default in Hebrew feminine verbs

On the menu Data The distribution of the FMSG suffixes Take I Default -et Take II Default -a Claim Both default ndash local and global default OT analysis V final stems ndash free variation Experimental results More questions

7

Part ALife is good

Part BLife is not perfect

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 8: A fault of default Local and global default in Hebrew feminine verbs

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

8

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 9: A fault of default Local and global default in Hebrew feminine verbs

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

9

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

PresentNo person contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 10: A fault of default Local and global default in Hebrew feminine verbs

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

10

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Plural Past and FutureNo gender contrast

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 11: A fault of default Local and global default in Hebrew feminine verbs

Suffixes in the verb paradigmPast Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 1

2 fm -t ms 2 fm -i

ms -ta ms

3 fm -a 3 fm

ms ms

Pl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u

3 -u 3 -u

11

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FuturePrefixes provide further contrast

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 12: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

12

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 13: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

13

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-t and -et can serve as the only surface cue for tense contrast

lsquoenterrsquo Ms Sg Fm Sg

Past nixnaacutes nixnaacutes-t lsquo2nd pr

Present nixnaacutes nixneacutes-et

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 14: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

14

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 15: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

15

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 16: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

16

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 17: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

17

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

FMSG suffixes in the verb paradigm

-a appears in both Past and Present

Ms Sg Fm Sg

Past hixniacutes hixniacutes-a lsquoput itrsquo

Present maxniacutes maxnis-aacute

Past kam kaacutem-a lsquoget uprsquo

Present kam kaacutem-a

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 18: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

18

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 19: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

19

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG suffixes in the verb paradigm

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 20: A fault of default Local and global default in Hebrew feminine verbs

Why

20

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 21: A fault of default Local and global default in Hebrew feminine verbs

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH present tense paradigm was not a verbal paradigm in BH

21

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

268-249ו-זהעברית ואחיותיה עדית תשסו-תשסז רוזן על הסמנטיקה של מערכת זמני הפועל בעברית דורון

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 22: A fault of default Local and global default in Hebrew feminine verbs

Why A historical note Modern Hebrew (MH) drew the verb paradigm from Biblical Hebrew (BH) MH Present tense paradigm was not a verbal paradigm in BH Noun morphology is much more chaotic than verb morphology

also with regard to the feminine suffixes The paradigm changed its status but the chaos hasnrsquot yet disappeared

22

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Schwarzwald R Ora 1991 Lexical weight in Hebrew inflectional feminine formation In Alan S Kaye (ed) Semitic Studies In honor of Wolf Leslau Wiesbaden Otto Harrassowitz 1409-1425

-t -et -i -a

Gender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

Present FMSGis the only case where one bundle of morphosyntactic features corresponds to two exponents

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 23: A fault of default Local and global default in Hebrew feminine verbs

Past Present Future

Num Pr Gen Num Gen Num Pr Gen

Sg 1 -ti Sg fm -a -et Sg 12 fm -t ms 2 fm -i

ms -ta ms3 fm -a 3 fm

ms msPl 1 -nu Pl fm -ot Pl 1

2 -tem ms -im 2 -u3 -u 3 -u

23

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-t -et -i -aGender Fm Fm Fm FmNumber Sg Sg Sg SgPerson 2nd 2nd (3rd)Tense Past Present Future Past amp Present

FMSG Present tense suffixes

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 24: A fault of default Local and global default in Hebrew feminine verbs

24

If -a and -et do not share an UR Then the default cannot be implied from the UR

Note No commitment as to whether the two suffixes share an UR

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 25: A fault of default Local and global default in Hebrew feminine verbs

25

If -a and -et share an UR Then the UR cannot serve as default since it differs from both

allomorphs

Faust Noam 2013 Decomposing the feminine suffixes of Modern Hebrew Morphology 23409-440

Phonological representation UR a t a t at CVCV

C V C V C V C V [-a] [-et]

Note No commitment as to whether the two suffixes share an UR

Default ndash assuming we give ldquobadrdquo markedness points to one-to-many association floating segment

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 26: A fault of default Local and global default in Hebrew feminine verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

26

Take I default -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 27: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

27 ספר קמרט ע תשלא גישה חדשה לתיאור הפועל ולהוראתו והדגמתה בנטיית הבינוני בתוך ש קודש )עורך( אורנן

44-32ירושלים המועצה להנחלת הלשון עמ

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 28: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

28

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 29: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

29

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 30: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final stems)

30

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 31: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

31

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 32: A fault of default Local and global default in Hebrew feminine verbs

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

Quantitative distribution

32

Zadok Gila 2012 Similarity Variation and Change In stability in Hebrew weak verbs PhD dissertation Tel-Aviv UniversityComplied and kindly provided by Shmuel Bolozky

Type Token

-a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 33: A fault of default Local and global default in Hebrew feminine verbs

33

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 34: A fault of default Local and global default in Hebrew feminine verbs

Summary of Take IThe data suggest that -a is the specific allomorph thus -et is the default based on three types of evidence

34

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 35: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

-et elsewherePolysyllabic verbs

bull with a non-high vowel in the final syllable nixnaacutesnofeacutel

nixneacutesetnofeacutelet

lsquoenterrsquolsquofallrsquo

bull with a non-low vowel in the first syllable mevakeacuteš mevakeacutešet lsquoaskrsquo

The distribution of -a and -et in the present tense (C-final verbs)

Specific

Default

35

Qualitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 36: A fault of default Local and global default in Hebrew feminine verbs

Present all verbs 41 12292964Present FmSg Present 8 1011229 -a 44 44101

-et 56 57101

36

Type Token -a 33 161498 43 119664281055

-et 68 337498 57 161391281055

498 most frequent verbs in written material

Natural speech tokens (2 hours recoding)

Quantitative distribution

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 37: A fault of default Local and global default in Hebrew feminine verbs

37

-a shows higher percentage of tokens than types -a appears with irregular verbs bull The 10 most frequent verbs are irregular (in different ways)bull 8 out of the 10 take -ayexolaacute lsquocanrsquo osaacute lsquodorsquo magiaacute lsquoarriversquo baacutea lsquocomersquo roaacute lsquoseersquo ʦrexaacute lsquoneedrsquo roʦaacute lsquowantrsquoomeacuteret lsquosayrsquo yodaacuteat lsquoknowrsquo

-et surfaces as -at after the historical pharyngeals

Type Token

-a 33 161498 43 119664281055

-et 68 338498 57 161391281055

Regularity -a associats with irregular verbs (more than -et)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -aAffiliation with irregular verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 38: A fault of default Local and global default in Hebrew feminine verbs

The FMSG suffixes -et and -a in the present tense

Which is the default

38

Take II default -a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 39: A fault of default Local and global default in Hebrew feminine verbs

39

The fault of the default Default X In the absence of any information to the contrary assume X The default allomorph appears whenever the specific one fails Assuming with Take I that -et is the default

we expect -et to appear whenever -a failsHowever hellip -a appears whenever -et fails (blocked) Therefore -a must be the default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 40: A fault of default Local and global default in Hebrew feminine verbs

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

The suffix -et is hosted by the weak syllable of a binary trochaic foot

40

-a emerges when the preceding vowel is not mid and it cannot

change to mid

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 41: A fault of default Local and global default in Hebrew feminine verbs

oxeacutelet lsquoeatrsquo oxeacutel PRSFMSG -et FMSG -a FMSG

oxeacutel-et oxel-aacute

šaacutera lsquosingrsquorsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Monosyllabic verbs take -a in Presentšaacutera lsquosingrsquokaacutema lsquoget uprsquoraacuteʦa lsquorunrsquo

-et has priority over -a in Presentoxeacutelet lsquoeatrsquomevašeacutelet lsquocookrsquonixneacuteset lsquoenterrsquo

41

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 42: A fault of default Local and global default in Hebrew feminine verbs

Phonological constraints on -et

[ C V C V C ]Ft

-lw -hi e t

The suffix -et is preceded by a mid vowel in an open syllable

42

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 43: A fault of default Local and global default in Hebrew feminine verbs

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

43

šaacutera lsquosingrsquo šaacuter PRSFMSG -et FMSG -a FMSG

šaacuter-et

šaacuter-a

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 44: A fault of default Local and global default in Hebrew feminine verbs

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

44

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 45: A fault of default Local and global default in Hebrew feminine verbs

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et šaacuter-a

šeacuter-et

FAITHVMONO

A vowel in a monosyllabic input has an identical correspondent in the output

45

šaacutera lsquosingrsquo šaacuter PRSFMSG MIDV-et -et FMSG -a FMSG

šaacuter-et

šar-a

šeacuter-et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 46: A fault of default Local and global default in Hebrew feminine verbs

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

46

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Cf nixnaacutes ndash nixneacuteset lsquohe ndash she entersrsquo

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 47: A fault of default Local and global default in Hebrew feminine verbs

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et maxneacutes-et

maxnis-aacute

FaithV[high]A high vowel in the input has an identical correspondent In the output

47

maxnisaacute lsquoput inrsquo maxniacutes PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 48: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

48

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 49: A fault of default Local and global default in Hebrew feminine verbs

šaacutera lsquosingrsquo šaacuter PRSFMSG FAITHVM MIDV-et -et FMSG -a FMSG

šaacuter-et

šaacuter-a

šeacuter-et

FAITHVMONO (FAITHVM)A vowel in a monosyllabic input has an identical correspondent in the output

49

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 50: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a and -et in the present tense (C-final verbs)

50

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 51: A fault of default Local and global default in Hebrew feminine verbs

maxnisaacute lsquoput inrsquo maxnis PRSFMSG FAITHVH MIDV-et -et FMSG -a FMSG

maxniacutes-et

maxneacutes-et

maxnis-aacute

FAITHV[high] (FAITHVH)A high vowel in the input has an identical correspondent In the output

51

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 52: A fault of default Local and global default in Hebrew feminine verbs

-aMonosyllabic verbs šaacuter šaacutera lsquosingrsquo

Polysyllabic verbs

bull with a high vowel in the final syllable makšiacutev makšivaacute lsquolistenrsquo

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

52

No synchronic story There is nothing bad with jaxoacutel ndash jaxoacutelet and kateacuten ndash kateacutenet

Consolation This group is gradually shrinking

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 53: A fault of default Local and global default in Hebrew feminine verbs

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

53

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 54: A fault of default Local and global default in Hebrew feminine verbs

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

54

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 55: A fault of default Local and global default in Hebrew feminine verbs

-a

bull with a low vowel in the initial open syllable jaxoacutelkateacuten

jexolaacutektenaacute

lsquocanrsquolsquoreducersquo

The distribution of -a in the present tense (C-final verbs)

55

Consolation This group is gradually shrinking Used only as adjectives kaveacuted ndash kvedaacute lsquoheavyrsquo

zakeacuten ndash zkenaacute lsquooldrsquo

New verb pattern and suffix gadeacutel ndash gdelaacute lsquogrowrsquogodeacutel ndash godeacutelet

jašeacuten ndash ješenaacute lsquosleeprsquo

jošeacuten ndash jošeacutenet

Gone for many speakers kateacuten ndash ktenaacute lsquoreducersquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 56: A fault of default Local and global default in Hebrew feminine verbs

When the default becomes specific

56

FMSG -aFMSG -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Present

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 57: A fault of default Local and global default in Hebrew feminine verbs

57

FMSG -aFMSG -et -a-a

Local default

Globaldefault

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

When the default becomes specific

Present

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 58: A fault of default Local and global default in Hebrew feminine verbs

Local and global default

58

Specific-en

Default -z

-a

English plural Hebrew FMSG present

Default -et

Default

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 59: A fault of default Local and global default in Hebrew feminine verbs

Local and global default

59

Specific-en

Default -d

-a

English past

Local default -et

Global default

Hebrew FMSG present

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 60: A fault of default Local and global default in Hebrew feminine verbs

60

Why is -a the global default-a seems to associate with FMSG more than other feminine suffixes All a-final nouns are FMSG (except laacutejla lsquonightrsquo and šulijaacute lsquoapprenticersquo) There are Vt-final nouns which are not FMSG

bull Final (unstressed) et seacuteret lsquofilmrsquo keacutelet lsquoinputrsquo peacutelet lsquooutputrsquobull Final it tafriacutet lsquomenursquo šarviacutet lsquoscepterrsquo

Loan nounsbull Final a ndash always feminine televiacutezja piʤaacutema maacuteskara dieacutetabull Otherwise ndash masculine flirt ʤoacuteint diskeacutet pakeacutet

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Experimental support is required

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 61: A fault of default Local and global default in Hebrew feminine verbs

B1 poneacute ndash ponaacute lsquoturnrsquo soneacute ndash soneacutet lsquohatersquo

B2 nivnaacute ndash nivneacutet lsquois builtrsquo nirpaacute ndash nirpeacutet lsquois healedrsquo

B3 mafneacute ndash mafnaacute lsquopoint outrsquo maxbiacute ndash maxbiaacute lsquohidersquo

B4 mefaneacute ndash mefanaacute lsquoclearrsquo mexateacute ndash mexateacutet lsquopurifyrsquo

B5 mitpaneacute ndash mitpanaacute lsquois vacatedrsquo mitkaneacute ndash mitkaneacutet lsquojealousrsquo

61

Vowel final verbs (the ldquolife is not perfectrdquo part)

The distribution of -a and -et is unpredictable As expected there is inter- and intra-speaker variation

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 62: A fault of default Local and global default in Hebrew feminine verbs

62

Vowel final verbs ndash historical reason for the chaos

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה)

Historically final Ɂ (א)

B1 poneacute ndash ponaacute soneacute ndash soneacutet

B2 nivnaacute ndash nivneacutet nirpaacute ndash nirpeacutet

B3 mafneacute ndash mafnaacute maxbiacute ndash maxbiaacute

B4 mefaneacute ndash mefanaacute mexateacute ndash mexateacutet

B5 mitpaneacute ndash mitpanaacute mitkaneacute ndash mitkaneacutet

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 63: A fault of default Local and global default in Hebrew feminine verbs

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

63

Vowel final verbs ndash distribution of suffixes (type)

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Introduction On default

The distribution of Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

ChaosV-final stems

Final remarksTake II

default -a

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 64: A fault of default Local and global default in Hebrew feminine verbs

Question addressedWhich of the two FMSG suffixes appearing in V-final present tense verbs in Hebrew is the default -et or -a

64

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 65: A fault of default Local and global default in Hebrew feminine verbs

65

Experiment (pilot)

Participants20 monolingual native speakers of Hebrew (mean age 225)

Task Simple sentences with present tense MSSG verbs Same verb in FMSG

Eg aacuteladin metateacute lsquoAladdin sweepsrsquo jasmiacuten hellip lsquoJasmin helliprsquo

Material (presented in the following order)bull Nonce verbs ndash 12 4 C-final

8 V-finalbull Actual verbs ndash 15 V-final 5 -a verbs

10 -et verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 66: A fault of default Local and global default in Hebrew feminine verbs

66

Results Actual verbs -a verbs (n=5) 100 accuracy -et verbs (n=10) -et ndash 46

-a ndash 54 (108200) ldquoerrorrdquo

kore mevate mitxabe mitpale mevade metate memale mexate mekane merape

lsquoreadrsquo lsquoexpressrsquo lsquohidersquo lsquosurprisersquo lsquoverifyrsquo lsquosweeprsquo lsquofillrsquo lsquopurifyrsquo lsquoenvyrsquo lsquohealsrsquo1920 1720 1220 1120 820 720 620 520 420 32095 85 60 55 40 35 30 25 20 15

33 282 --- --- 369 --- 163 --- --- ---

Correct responses of -et verbs

Position in Bolozkyrsquos frequency list

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 67: A fault of default Local and global default in Hebrew feminine verbs

Binyan Historically final V (ה) Historically final Ɂ

(א)

B1 poneacute ndash ponaacute 99 soneacute ndash soneacutet 18

B2 nivnaacute ndash nivneacutet 24 nirpaacute ndash nirpeacutet 9

B3 mafneacute ndash mafnaacute 43 maxbiacute ndash maxbiaacute 18

B4 mefaneacute ndash mefanaacute 52 mexateacute ndash mexateacutet 12

B5 mitpaneacute ndash mitpanaacute 43 mitkaneacute ndash mitkaneacutet 8

78 (255327 ) -a 22 (72327 ) -(e)t

67

Vowel final verbs ndash distribution of suffixes (type)

Type frequency could play a role in the experiment

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 68: A fault of default Local and global default in Hebrew feminine verbs

68

-a verbs (n=5) 100 accuracy Not in reality

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 69: A fault of default Local and global default in Hebrew feminine verbs

69

mitkašeacutet instead of mitkašaacute lsquofind it difficult becomes hardrsquo

-a verbs (n=5) 100 accuracy

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Not in reality

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 70: A fault of default Local and global default in Hebrew feminine verbs

70

merameacutet instead of meramaacute lsquocheatsrsquo

koneacutet instead of konaacute lsquobuyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

roʦeacutet instead of roʦaacute lsquowantrsquo

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 71: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsC-final ndash 4 verbs As expected mostly -et

71

Syncretic with past 3rd FMSG - always -a

-a -etB1 0 020 100 2020B2 45 920 55 1120B4 0 020 100 2020B5 5 120 95 1920

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 72: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

72Expectation ndash equal distribution Not significant (p=01548)Expectation ndash 78 -a and 22 -et (following type distribution) Significant (plt00001)

Where has 32 gone

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etB1 50 2040 20 840B2 0 040 65 2640B4 23 940 38 1540B5 43 1740 35 1440

29 46160 39 63160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 73: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

73

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 74: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

74

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Dictionary distributionBinyan Historically final ltעgt Type

B1 šomeacutea ndash šomaacuteat lsquohearrsquo 38

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo 23

B3 mašpiacutea ndash mašpiaacute lsquoaffectrsquo 36

B4 mevaʦeacutea ndash mevaʦaacuteat lsquoperformrsquo 27

B5 mitbaʦeacutea ndash mitbaʦaacuteat lsquois executedrsquo 22

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 75: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

75

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Binyan ltאgt ltהgt ltעgt B1 99 -a 18 -et 38 -at -a 62B2 25 -et 9 -et 23 -at -et 15B3 43 -a 18 -a 36 -a -at 23B4 52 -a 12 -et 27 -atB5 43 -a 8 -et 22 -atTotal 473 262 55 65 14 146 31

Dictionary distribution

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 76: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

76

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -atB1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160 Still missing 24

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 77: A fault of default Local and global default in Hebrew feminine verbs

Results Nonce verbsV-final ndash 8 verbs (2 for each binyan)

77

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -et -at -eCetB1 50 2040 20 840 5 240 25 1040B2 0 040 65 2640 18 740 18 740B4 23 940 38 1540 3 140 38 1540B5 43 1740 35 1440 8 340 15 640

29 46160 39 63160 8 13160 24 38160

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 78: A fault of default Local and global default in Hebrew feminine verbs

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

78

-eacuteCet (24) ndash experiment data How desperate one can get

There is no C-copying within the inflectional paradigm There are no structural relations of this sort in the inflectional paradigm

We want -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

>
>
>
>
>

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 79: A fault of default Local and global default in Hebrew feminine verbs

Input melage mezate tfʃoke goze mitgaze mitgade nirxa nirʒa

add hs melageacutehet tfʃokeacuteset

add t tfʃokeacutetet gozeacutetet nirxeacutetet nirʒeacutetet

add n melageacutenet mezateacutenet tfʃokeacutenet gozeacutenet mitgazeacutenet nirxeacutenet nirʒeacutenet

add r melageacuteret tfʃokeacuteret mitgadeacuteret

C copy melageacuteget mezateacutetet tfʃokeacuteket gozeacutezet mitgazeacutezet mitgadeacutedet nirʒeacutezet

79

-eacuteCet (24) ndash experiment data

The strategies used to add a consonant are familiar from derivation C-copyingbull Actual words xam lsquohotrsquo ndash ximeacutem lsquoto heatrsquobull Experiment (coining novel verbs) kiseacute lsquochairrsquo ndash mekaseacutes

Adding a consonant ndash coronal Experiment (coining novel verbs)kiseacute lsquochairrsquo ndash mekaseacutet mekaseacuten

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Berman Ruth 1989 Childrenrsquos knowledge of verb structure Data from Hebrew A paper presented in the 14th Annual Boston University conference on Language Development

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 80: A fault of default Local and global default in Hebrew feminine verbs

) ־5ה ) או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה בכל סיומת שלרוב אלא שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני בניין

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

The feminine suffix in the participle (present tense) can be either -Vt or -a

80

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 81: A fault of default Local and global default in Hebrew feminine verbs

roʦeacutet lsquowantrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

81

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 82: A fault of default Local and global default in Hebrew feminine verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

82

Two forms for almost every FMSG one with -a and one with -et

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 83: A fault of default Local and global default in Hebrew feminine verbs

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

83

Two forms for almost every FMSG one with -a and one with -et

Not only for V-final stems but also for C-final stems

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 84: A fault of default Local and global default in Hebrew feminine verbs

84

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת־ת היא השגורה הסיומת הבניינים ברוב כך ־)ה הסיומת שגורה הפעיל בבניין ואילו

סורקה יושבה מוצאים בספרות אבל סורקת יושבת לומר רגילים קל בבניין למשל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם מותר יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת ממולאת ) ( ) ( ושאר פיעל בבניין 0את מ1קנ במקום 10א)ה מ1קנ קל בבניין קו2ראת במקום קו2ר1א)ה לומר

בהן כיוצא הצורות

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In most verb classes (binyanim) the common suffix is -Vt (this is our default)

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 85: A fault of default Local and global default in Hebrew feminine verbs

85

) בניין ) בכל שלרוב אלא ־)ה או ־ת להיות יכולה ההווה הבינוני צורות של הנקבה סיומת שאינה או האחרת הסיומת ואילו השגורה לסיומת אחת סיומת נעשתה הפועל מבנייני

כבולים בצירופים או ספרותית בלשון רק אותה שמוצאים או כלל משמשת כך ־)ה הסיומת שגורה הפעיל בבניין ואילו ־ת היא השגורה הסיומת הבניינים ברוב

סורקת יושבת לומר רגילים קל בבניין למשל סורקה יושבה מוצאים בספרות אבל אומרים הפעיל בבניין זאת לעומת שערה וסורקה לחלון יושבה היא ביאליק של כבשירו ידו אין אבל משיגה אומרים כלל בדרך רגשת מ יש ההלכה בספרות אבל מרגישה

משגת ( ד שיו משורשים להבדיל ף באל המסתיימים משורשים בינונית לצורות הדין הוא ndash מדכאת( קוראת ־ת סיומת הבניינים ובשאר ממציאה מקריאה בהפעיל בסופם

יושבת במקום יושבה לומר שמותר כפי אבל מתבטאת מומצאת מותר ממולאת ) ( HIא5ה מHקJנ קל בבניין קוMרLאת במקום א5ה HרMקו ( לומר ( פיעל בבניין 0את מ1קנ במקום

בהן כיוצא הצורות ושאר

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

According to the Hebrew Language Academy

In literary style one can use the other suffix -a instead of -et and vice versa

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 86: A fault of default Local and global default in Hebrew feminine verbs

86

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 87: A fault of default Local and global default in Hebrew feminine verbs

87

This is what our great poets used to do

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

יא הה ב 13לוןיוש ח ל13הו ק 13השור ר ע 13 ש

יא ם ה יכ ינ ע 13בה רוצ 13פ

ה ר יא ב י ה ינ ע 13וב

Hayim Nahman Bialik

jošvaacute ndash jošeacutevetsorkaacute ndash soreacutevet

Miriam Yalan Shtekelis

נורית קמה לפתע אךהיא אבא קוראהאבא

החושך גרש מהר בוארע ילד מפריע הוא

korʔaacute ndash koreacutet

Not native speakers of Hebrew

Immigrated to Israel - age 51

Immigrated to Israel - age 20

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 88: A fault of default Local and global default in Hebrew feminine verbs

88

Whether or not we are ldquoallowedrdquo to use both suffixes our language knowledge assumes a system Indeed the system suffers from a certain degree of inconsistency

when it comes to V-final stems However grammatical systems tend to fix themselves over time and

the interesting questions arebull How does the system work now (I gave a partial answer only with

regard to C-final stems)bull In what way the system is going to fix itself (I need more

experimental work here)

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 89: A fault of default Local and global default in Hebrew feminine verbs

89

Summary hellip Two allomorphs for the FMSG present tense Both have some characteristics of the default allomorph though -a

has also characteristics of the specific allomorph My claim

Both are default ndash but in different domains local default in the present tense and global default beyond the present tense

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 90: A fault of default Local and global default in Hebrew feminine verbs

ThankYou

Thank you Betsy Ritter and Noam Faust for brain storming

Thank you Daniel Asherov and Stav Klein for the experiment

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 91: A fault of default Local and global default in Hebrew feminine verbs

91

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 92: A fault of default Local and global default in Hebrew feminine verbs

Why OT Reason I Rule-based approachbull FMSG -a monosyllabic stems etc bull FMSG -et

This analysis fails to reflect the observation that -a appears whenever -et is blocked ie also -a behaves like a default

OT analysis reflects this observation allowingbull -et to be the local default of FMSG present tense andbull -a to be the global default to appear whenever the local default -et

fails

92

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 93: A fault of default Local and global default in Hebrew feminine verbs

Why OT Reason II Underspecificationbull -et FMSG Verb Presentbull -a FMSG Verb -a FMSG

This analysis fails to account for the following facts

93

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 94: A fault of default Local and global default in Hebrew feminine verbs

94

Present tense verbs (participles) can also serve as nouns

Verb Noun

moreacute lsquopoints at MSSGrsquo lsquoteacherrsquo

ʦoleacutelet lsquodives FMSGrsquo lsquosubmarinersquo

menaheacutel lsquomanages MSSGrsquo lsquomanagerrsquo

kam lsquoget up MSSGrsquo lsquoenemyrsquo

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 95: A fault of default Local and global default in Hebrew feminine verbs

-et and -a appear with other lexical categories

-a MsSg FmSg

Noun jeacuteled jaldaacute lsquoboy-girlrsquo

Verb jalaacuted jaldaacute lsquogave birthrsquo

Adjective xašuacutev xašuvaacute lsquoimportantrsquo

-et MsSg FmSg

Noun dajaacutel dajeacutelet lsquostewardrsquo

Verb oleacutex oleacutexet lsquogoesrsquo

Adjective mejutaacuter mejuteacuteret lsquoredundantrsquo

95

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 96: A fault of default Local and global default in Hebrew feminine verbs

-a -etmištaacuter lsquoregimersquo mištaraacute lsquopolicersquo

guf lsquobodyrsquo gufaacute lsquocorpsersquo

meduraacute lsquobonfirersquo

kivuacuten lsquodirectionrsquo kavanaacute lsquointentrsquo kaveacutenet lsquo(gun) sightsrsquo

seacutevel lsquosufferingrsquo siboacutelet lsquostaminarsquo

mirpeacuteset lsquobalconyrsquo

96

-et and -a serve as derivational suffixes as well

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Both -a and -et are specified for two features only ndash Feminine and Singular

OT can handle the distribution of the feminine singular suffixes with co-phonologies

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 97: A fault of default Local and global default in Hebrew feminine verbs

ponaacute lsquoturnrsquo poneacute PRSFMSG VV -et FMSG -a FMSG MAX

poneacutea

ponaacute

poneacuteet

poneacutet

97

koret lsquoreadrsquo koreacute PRSFMSG VV -et FMSG -a FMSG MAX

koreacutea

koraacute

koreacuteet

koreacutet

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 98: A fault of default Local and global default in Hebrew feminine verbs

98

koreacutet lsquoreadrsquo koreacute PRSFMSG VV MAX Stem MAX Suff -et -a

koraacute

koreacutet

ponaacute lsquoturnrsquo poneacute PRSFMSG VV MAX Suff MAX Stem -et -a ponaacute

poneacutet

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 99: A fault of default Local and global default in Hebrew feminine verbs

koreacute-et koreacutet lsquoreadrsquo suffix vowel e is deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e is deleted

99

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

koreacutet lsquoreadrsquo koreacute -et -a MAXVSuff MAXVStm

koreacutet

koraacute

kore-et

pone-a

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 100: A fault of default Local and global default in Hebrew feminine verbs

Binyan Historically final ltעgt Historically final ltחgt

B1 šomeacutea ndash šomaacuteat lsquohearrsquo šoxeacuteax ndash šoxaacutexat lsquoforgetrsquo

B2 nišmaacute ndash nišmaacuteat lsquois heardrsquo niftaacutex ndash niftaacutexat lsquois openedrsquo

B3 mašpiacutea ndash mašpiaacute lsquoinfluencersquo mašbiacuteax ndash mašbixaacute lsquoimproversquo

B4 mešageacutea ndash mešagaacuteat lsquomaddenrsquo mešabeacuteax ndash mašabaacutexat

B5 mitpareacutea ndash mitparaacuteat lsquogo wildrsquo mitpateacuteax ndash mitpataacutexat

78 (255327 ) -a 22 (72327 ) -(e)t

Distribution ndash type

100

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 101: A fault of default Local and global default in Hebrew feminine verbs

101

Further thoughts hellip Since the same forms are used for both nounsadjectives and verbs

there might be a reorganization in the systembull -a for nounsadjectivesbull -et for verbs

Partial results the of -a lowered with we added ldquonowrdquo or ldquothis momentrdquo to the sentencesbull -a with ldquonowthis momentrdquo ndash 32 (1856)

aacuteladin metateacute axšav mitaacutexat lašatiacuteax lsquoAladdin sweeps now under the rugrsquobull -a without ldquonowthis momentrdquo ndash 27 (28104)

aacuteladin metateacute mitaacutexat lašatiacuteax lsquoAladdin sweeps under the rugrsquo Further studies are required also with reference to FMSG in nouns

and adjectives

Introduction On default

Hebrew FMSG suffixes in the verb paradigm

Take I default -et

OT analysis

Local amp global default

Chaos V-final stems

Final remarksTake II

default -a

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 102: A fault of default Local and global default in Hebrew feminine verbs

Final ltעgt Final ltאgt Final ltהgt

nišmaacute ndash nišmaacuteat lsquois heardrsquo nimʦaacute ndash nimʦeacutet lsquois foundrsquo nifnaacute ndash nifneacutet lsquoturnrsquo

24 24 9

102

Vowel final verbs ndash distribution of suffixes

On defaul

t

Hebrew verb

inflectional

suffixes

Take I default -

et

Take II default -

a

Local amp

global defaul

t

OT analysi

s

Chaos V-final verbs

Final remark

s

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

MSSG ndash CiCCaacute for all vowel final stems MSSG ndash unique for final ltעgt

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 103: A fault of default Local and global default in Hebrew feminine verbs

koreacute-et koreacutet lsquoreadrsquo suffix vowel e ndash deleted

poneacute-a ponaacute lsquoturnrsquo stem vowel e ndash deleted

103

bull Strength relation stem gt suffixbull Realize Morph The feature bundle PRESFMSG has an exponentRealizeMorph gtgt MaxVStem gt MaxVSuff

maxnisaacute lsquoput inrsquo koreacute RM MaxVEdge -et -a

koreacutet

koraacute

koreacute

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 104: A fault of default Local and global default in Hebrew feminine verbs

Selecting the optimal formponaacute lsquoturnsrsquo boleacutea MAXV2 VV MAX[a] MAX[e] -et -a

boleacuteaet

boleacuteaet

boleacuteat boleacuteaet

boleacuteaet

bolaacuteet boleaacuteet

boleacuteaa

boleacuteaaacute

boleacuteaa

boleacuteaacutea

104

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 105: A fault of default Local and global default in Hebrew feminine verbs

None of these restrictions is weird in Hebrew morphology

MIDV-etThe suffix -et is preceded by a mid vowel in an open syllable

At the core of Hebrew morphology stand the morphological classes (binyanmishkal) which are defined in terms of bull prosodic structure (number of syllables and syllable structure)bull vocalic patterns andbull affixes

Bat-El O 2002 Semitic verb structure within a universal perspective Language Processing and Acquisition in Languages of Semitic Root-based Morphology J Shimron (ed) Amsterdam John Benjamins 29-59

105

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 106: A fault of default Local and global default in Hebrew feminine verbs

Vowel final stems ndash hellipCe]Stem

106

MSSG FMSG MSSG FMSG

poneacute ponaacute lsquoturnsrsquo soneacute soneacutet lsquohatersquo

mexaseacute mexasaacute lsquocoversrsquo medakeacute medakeacutet lsquodepressingrsquo

mitbaleacute mitbalaacute lsquowears outrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ה) 80 (final א) 20

Distribution in the dictionary

Tarmon Asher and Ezri Uval 1998 Hebrew Verb Tables Jerusalem Tamir Publisher

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 107: A fault of default Local and global default in Hebrew feminine verbs

Vowel final stems ndash hellipCeacutea]Stem

107

MSSG FMSG MSSG FMSG

jodeacutea jodaacuteat lsquoknowrsquo soneacute soneacutet lsquohatersquo

mesajeacutea mesajaacuteat lsquohelprsquo medakeacute medakeacutet lsquodepressingrsquo

mitpareacutea mitparaacuteat lsquogo wildrsquo mitnaseacute mitnaseacutet lsquostuck uprsquo

(final ע) (final א) 20

Distribution in the dictionary

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 108: A fault of default Local and global default in Hebrew feminine verbs

108

Default vs basicDefault of category vs formDefault and underspecification

Nominative is the maximally unmarked case category in the language with no restrictions on its occurrence It is not

assigned when the other cases fail to be assigned rather it is the lack of assignment of other cases

McFadden Thomas 2007 Default case and the status of compound categories in Distributed morphology In Tatjana Scheffler Joshua Tauberer Aviad Eilam and Laia Mayol (eds) Penn Working Papers in Linguistics 13(1) Proceedings of the 30th Annual Penn Linguistics Colloquium 225-238

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 109: A fault of default Local and global default in Hebrew feminine verbs

109

Default and related notions

Default and underspecification

xxxx

bull Default category is not specified for morphological features

Farkas Donca 1990 Two Cases of Underspecification in Morphology Linguistic Inquiry 21539-50

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 110: A fault of default Local and global default in Hebrew feminine verbs

110

Is it a case of contrast neutralizationbull Due to phonological restrictions on -et the contrast between 3rd

FMSF past and FMSG present forms is neutralizedbull Why -a lsquo3rd FMSF past rsquo and not other feminine suffixes Eg -t lsquo2nd

FMSF past rsquo -i lsquo2rd FMSF futurersquo

Arregi Karlos and Andrew Nevins 2012 Contextual neutralization and the elsewhere principle Ms University of Chicago and University College London

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 111: A fault of default Local and global default in Hebrew feminine verbs

On the notion ldquodefaultrdquo

The Elsewhere Condition (Kiparsky 1982)The more specific process applies before the more general default one such that the specific blocks the default

Common non-theoretical characteristics of the default morphemebull more frequent bull used in loan wordsbull first to appear in childrenrsquos speech

Theoretical characteristics of the default morphemebull less specified bull less restricted contextually

111

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 112: A fault of default Local and global default in Hebrew feminine verbs

112

Outline of the talk

Hebrew verb paradigms undergo inter-paradigm leveling

Hebrew verb paradigms ndash classes and sub-classes

Change-oriented variation ndash inter-paradigm leveling

Similarity

There are two types of directionality

Frequency

Concluding remarks

Introduction Hebrew verb paradigms

Variation DirectionalityConclusion

Similarity Frequency

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 113: A fault of default Local and global default in Hebrew feminine verbs

Default X In the absence of any information to the contrary assume X

Booij G 1995 The Phonology of Dutch Oxford Oxford University Press

113

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 114: A fault of default Local and global default in Hebrew feminine verbs

Binyan Total ltאgt ltהgt ltעgt

B1 719 99 -a 18 -et 38 -at

B2 328 25 -et 9 -et 23 -at

B3 527 43 -a 18 -a 36 -a

B4 861 52 -a 12 -et 27 -at

B5 515 43 -a 8 -et 22 -at

Total 2950 262 9 65 2 146 5

114

Vowel final verbs ndash distribution of suffixes

Results Nonce verbs-a -et -at

B1 50 2040 20 840 5 240B2 0 040 65 2640 18 740B4 23 940 38 1540 3 140B5 43 1740 35 1440 8 340

29 46160 39 63160 8 13160

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 115: A fault of default Local and global default in Hebrew feminine verbs

Default X In the absence of any information to the contrary assume X

Dutch Diminutive

Information to the contrary - specific 1 -jə after stem-final obstruents

- specific 2 -ətjə after sonorant Cs preceded by a short stressed V

No information to the contrary - default -tjə (subject to place assimilation)

115

Phonological information

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 116: A fault of default Local and global default in Hebrew feminine verbs

116

Default vs basic

Zwicky M Arnold 1986 The general case Basic form versus Default form Proceedings of the 12th Annual Meeting of the Berkeley Linguistics Society 305-314

bull Basic ndash a structural notion referring to the underlying representation thus basic is the default (eg Dutch)

bull Default ndash an organizational notion referring to order of rule application - is not always basic non-phonologically conditioned allomorphy)

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117
Page 117: A fault of default Local and global default in Hebrew feminine verbs

QuestionWhich of these two suffixes is the default

Empirical focusThe FM SG suffixes in Hebrew present tense (participle) ndash -et and -a

Theoretical interestExpansion of the notion default ndash local and global default

117

  • Slide 1
  • Slide 2
  • Slide 3
  • Slide 4
  • Slide 5
  • Slide 6
  • Slide 7
  • Slide 8
  • Slide 9
  • Slide 10
  • Slide 11
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • Slide 20
  • Slide 21
  • Slide 22
  • Slide 23
  • Slide 24
  • Slide 25
  • Slide 26
  • Slide 27
  • Slide 28
  • Slide 29
  • Slide 30
  • Slide 31
  • Slide 32
  • Slide 33
  • Slide 34
  • Slide 35
  • Slide 36
  • Slide 37
  • Slide 38
  • Slide 39
  • Slide 40
  • Slide 41
  • Slide 42
  • Slide 43
  • Slide 44
  • Slide 45
  • Slide 46
  • Slide 47
  • Slide 48
  • Slide 49
  • Slide 50
  • Slide 51
  • Slide 52
  • Slide 53
  • Slide 54
  • Slide 55
  • Slide 56
  • Slide 57
  • Slide 58
  • Slide 59
  • Slide 60
  • Slide 61
  • Slide 62
  • Slide 63
  • Slide 64
  • Slide 65
  • Slide 66
  • Slide 67
  • Slide 68
  • Slide 69
  • Slide 70
  • Slide 71
  • Slide 72
  • Slide 73
  • Slide 74
  • Slide 75
  • Slide 76
  • Slide 77
  • Slide 78
  • Slide 79
  • Slide 80
  • Slide 81
  • Slide 82
  • Slide 83
  • Slide 84
  • Slide 85
  • Slide 86
  • Slide 87
  • Slide 88
  • Slide 89
  • Slide 90
  • Slide 91
  • Slide 92
  • Slide 93
  • Slide 94
  • Slide 95
  • Slide 96
  • Slide 97
  • Slide 98
  • Slide 99
  • Slide 100
  • Slide 101
  • Slide 102
  • Slide 103
  • Slide 104
  • Slide 105
  • Slide 106
  • Slide 107
  • Slide 108
  • Slide 109
  • Slide 110
  • Slide 111
  • Slide 112
  • Slide 113
  • Slide 114
  • Slide 115
  • Slide 116
  • Slide 117