Embed Size (px)
Citation preview
The Intonational Phonology ofYes-No Questions in
African American Vernacular English
Cybelle SmithNWAV
October 29th, 2011
How would you askthis question:
“Do you hear the sleigh bells ringing?”
TIMIT Acoustic-‐Phonetic Continuous Speech Corpus 1
Do you hear the sleighbells ringing?
Pitch
Time
Pitch Contours on the word “ringing”
TIMIT Acoustic-‐Phonetic Continuous Speech Corpus 2
Do you hear the sleighbells ringing?
Pitch
Time
Pitch Contours on the word “ringing”
TIMIT Acoustic-‐Phonetic Continuous Speech Corpus 2
Sources of Intonational Variation:
• Token Level:Semantic intentEmotional stateSyntactic constraints
• Speaker/Societal Level:Individual differencesLanguage differencesSocio-‐linguistic factors
3
Dialect Variation
Grabe (2004)4
African American Vernacular English
• Tarone (1973)– falling [inal pitch in yes-‐no questions more commonin “formal, threatening situations”
– level and rising intonation in informal, familiarsituations
• Jun and Foreman (1996)– more variable boundary tones in AAVE than SEduring role-‐playing
• Green (2002)– [lat [inal syllable (SW Louisiana teens and adults)
5
Current Study Objectives:
• Describe Quantitatively– How are AAVE speakers marking their yes-‐noquestions (in cases with a [lat [inal syllable)?
– Flat syllable across the country?– Con[irm in spontaneous speech
• Assess impact of:– syntactic inversion– semantic constraints
6
Methods
• Corpus analysis:– CALLFRIEND American English (Southernand Non-‐Southern Dialect)
– East Palo Alto Sociolinguistic Interview withFoxy Boston
7
CALLFRIEND• 30 min. phone conversations• “Southern” or “non-‐southern” dialectbased on vowel phonology
• All AAVE speakers = “southern dialect”,regardless of geography
• No ethnic data on individual speakers -‐-‐How do I know this is AAVE?
8
AAVE Identi[ication• First: picked speakers from northern cities in theSouthern Dialect corpus
• Next: veri[ied speakers were using other AAVEfeatures:– zero copula– doesn’t -‐> don’t– consonant cluster simpli[ication (e.g. talked -‐> talk, slipped -‐>slip)
– other sound changes (e.g ð. -‐> d)• Additional evidence:
– Lexical items (brother, sister)– Use of falsetto by one of the male speakers
9
Data Collection: Speakers
10
Speaker
1
2
3
4
5
6
7
8
9
10
11
12
Corpus
Cal
lfrien
dSoci
olin
g
Inte
rvie
w
Gender
Male(4)
Female(8)
Place
East Palo Alto(4)
New York(2)
Michigan(2)
Pennsylvania(4)
Variety
AAVE-influenced HE (1)
SAE (1)
AAVE (1)
SAE (1)
AAVE(2)
SAE(2)
AAVE(4)
7 AAVE4 SAE1 HE
Data Collection: Tokens
• Questions identi[ied and screened forpragmatic intent– Information-‐Seeking (IS)– Information-‐Con[irming (IC)
11
This will be important later!
• Tokens with bad sound quality dropped
End Result: 125 yes-no questions
publica
publications
Quantitative Metrics
12
Getting any publications?Whole QuestionFinal Word
Final Syllable
Final Stressed Syllable (FSS)
Syllable Before FSS
First Syllable of Final Word
tions
Example Question from CALLFRIEND:
Pitch slope from minimumto maximum
Pitch slope fromstart to end
13
Motivation for Features
14
Main Findings:• Level and falling terminal pitch
– in AAVE AND SE– higher rates in AAVE– prosodic impact
• Inversion correlates strongly with pitch slope of the[inal syllable for male AAVE speakers
• “Seeking” vs. “Con[irming” Questions: AAVE Males andSE Females have different phonetic correlates
15
AAVE Contour SimilarityOver Age, Distance, GenderFoxy Boston, Female, Age 13, 1986, East Palo Alto
CFM40PA, Male, Age 40, c.1996, Grew up in PA
Was it fun?
Going to school? 16
Final Syllable Pitch Slope• AAVE speakers more likely to have level or falling [inalsyllable pitch slope
The Pennsylvania Males:
p = .035p = .038 17
Prosodic In[luence• Dialectal difference greater when last twosyllables “unstressed stressed”
The Pennsylvania Males
n = 40, p = .035 n = 15, p = .03918
Question Pitch Slope• Pitch slope over the question as a whole is surprisinglysimilar
The Pennsylvania Males
19
Question Riseby Different Means
19-‐year-‐old Female SE Speaker from NY
40-‐year-‐oldMale AAVESpeakerfrom PA
20
Implications
• Liu and Xu (2007)– Yes-‐no questions diverge from declarativesat stressed syllable of [irst content word
• Could AAVE speakers be exaggeratingthis early rise?
21
Inversion
22
Auxiliary-‐Subject Inversion
• Inverted:– Is it a book?– Do you like pizza?
• Non-‐Inverted:– It’s a book?– You like pizza?
23
Inversion and Intonation
• Haan (2001) -‐ Dutch• Grabe (2004) -‐ English in British Isles
Inverted Questions
Non-‐Inverted Questions
24
Inversion and Pitch Slope
• Huge effect for AAVE males, slight effectfor everyone else
p < .01 25
Tarone’s Claim Revisited• Tarone (1973) claims AAVE speakers more likely tohave falling terminal pitch in “formal and threateningsituations”
• Tarone’s examples:– Are you the teacher?– Is the man here?
• Maybe her results re[lect greater use of inversion informal speech
26
FormalSituation
InvertedQuestion
FallingTerminal Pitch
Formal Situation
InvertedQuestion
Falling Terminal Pitch
Model 1
Model 2
(threatening)
(more standardgrammar)
(AAVE phonology)
(more standardgrammar)
27
Semantic In[luence
28
Information Seeking vs.Con[irming Questions
• Information seeking:A question to which you don’t know the answer:– Are your parents around?
• Con[irming:– Foxy: Well-‐ I don't know hardly about nothing go on cause I'mnew there. I just started.
– Interviewer: That's right. In September huh?– Foxy: Um mm, I started in October.– Interviewer: Oh in October. So you didn't start at thebeginning of the year?
29
Semantic In[luence• Con[irming Questions:
– greater pitch range on the [inal syllable– higher maximum pitch on the [inal syllable
• Dialectal differences:– AAVE males raise minimum pitch on last syllable, pitch rangeincrease not signi[icant
– SE females lower minimum pitch on last syllable, pitch rangesigni[icantly larger
30
Final Syllable Pitch RangeFemales
p = .618 p = .015*
31
Males
Information Con[irming
Information Seeking
p = .901 p = .351
Final Syllable Maximum PitchFemales Males
Information Con[irming
Information Seeking
p = .718 p = .059 p = .010** p = .783
32
Final Syllable Minimum PitchFemales Males
Information Con[irming
Information Seeking
p = .308 p = .597 p = .005** p = .723
33
Why this is interesting• AAVE and SE speakers use different contours toexpress interrogativity
BUT the function:– ConJirming information -> exaggerate intonational cues tointerrogativity
remains the same
• From this, would expect in different language varietieswith different contours, this correlation should hold.
• The question is -‐-‐ how universal is this correlation or isit speci[ic to languages similar to English?
34
Summary• Intonation contours of yes-‐no questions variable inAAVE and SE speakers
• But some constraints:– Higher rate of [lat and falling [inal pitch in AAVE– Male AAVE speakers show signi[icant pitch contour correlationwith changes in inversion
– AAVE males and SE females both exaggerate interrogativemarkers of their language variety for con[irming questions
35
Thank you!
• Meghan Sumner, advisor• John Rickford, lots of helpful advice, alloweduse of EPA Neighborhood Survey
• Additional people I should thank:Kathryn Potts, Arto Antilla, Christopher Potts,Robert Podesva, Tyler Kendall, Sun-‐Ah Jun, LisaGreen, Marisa Tice, Marie-‐Catherine deMarneffe, Jason Grafmiller, David Clausen, MarkLinsey
36
ReferencesGrabe, E. (2004). Intonational variation in urban dialects of English spoken in the
British Isles. In Gilles, P. and Peters, J. (eds.), Regional Variation in Intonation.Linguistische Arbeiten, Tuebingen, Niemeyer, 9-‐31.
Green, Lisa J. (2002). African American English: A Linguistic Introduction, Cambridge:Cambridge University Press.
Haan, Judith (2001). Speaking of Questions. An Exploration of Dutch QuestionIntonation. Dissertation, Netherlands Graduate School of Linguistics (LOT).
Jun, Sun-‐Ah and Christina Foreman. (1996). Boundary tones and focus realization inAfrican American English Intonation. Abstract. Third Joint Meeting of ASA and ASJ.Hawaii. Fall.
Liu, F. and Xu, Y. (2007). Question intonation as affected by word stress and focus inEnglish. In Proceedings of The 16th International Congress of Phonetic Sciences,Saarbrucken. August, 2007. 1189-‐1192.
Tarone, Elaine E. (1973). Aspects of intonation in Black English. American Speech. DukeUniversity Press. 48-‐1/2, 29-‐36.
37
38
