Introduction •  /s/-aspiration in Spanish: estanco ‘kiosk’ [ehˈtaŋko]

•  Sound change from pre- to post-aspiration (Ruch & Harrington 2014)

•  Indexical function of phonetic variation (e.g. Clopper & Pisoni 2004, Foulkes et al. 2011, Mack & Munson 2012)

Results

•  Perception experiment18 stimuli pairs A-B from 17 speakers (11 SE, 6 GR; 11 < 55y, 6 > 55y),19 filler pairs (e.g. secaron – secaron) from 19 other speakers; randomized

•  Andalusian listeners deduced speaker‘s age and geographic origin from fine-phonetic detail (pre-/post-aspirated stops)

•  Participants mentioned stop aspiration as a typical feature of Seville Spanish, but not of younger speakers⇨ explicit knowledge of geographical variation⇨ implicit knowledge of change

•  Speakers who were perceived as being from Seville were at the same time judged as being younger, consistent with the production data (Ruch 2013).⇨ no knowledge of sound change in Granada Spanish

•  Although Eastern Andalusian varieties tend to converge towards the Castilian standard (Hernández-Campoy & Villena Ponsoda 2009), younger Granada speakers are adopting a typically Western feature

•  Granada: independent process or contact-induced change?

3rd Workshop on Sound Change, University of California, Berkeley, May 28-31 2014 The indexical function of pre- and post-aspiration

in a sound change in progress

H1: Post-aspirated stops are associated with speakers from Seville, pre-aspirated stops with speakers from Granada.

H2: Post-aspirated stops are associated with younger, pre-aspirated stops with older speakers.

H3: The effect of the phonetic variant on the hearer’s response is more marked in words with /st/ than in words with /sp, sk/.

Hanna Ruch, URPP Language and Space, University of Zurich hanna.ruch@uzh.ch

⇨ Do listeners of Andalusian Spanish have (explicit or implicit) knowledge of this variation and change?

•  31 isolated words(e.g. estado, escapa, España) spoken by Andalusians

•  Pre-test: AXB, 16 subj. Are differences perceptible? Método%

14%

•  50%par8cipantes:%22%granadinos,%28%sevillanos%(16%E%60%años)%•  preguntas%personales%sobre%movilidad,%contactos,%etc.%%•  datos%cualita8vos%/%comentarios%

Where do you think is this speaker from?How old do you think is he/she?

Production of pre- and post-aspiration (48 speakers; Ruch 2013)

[ph, th, kh] estanco [eˈthaŋko] post-aspiration

[hp, ht, hk] estanco [ehˈtaŋko] pre-aspiration

/sp, st, sk/ Andalusian Spanish

→

Hypotheses

Method

Qualitative results: How did you distinguish the speakers?

•  59 listeners: 28 from Seville, 31 from Granada; 50 < 55 y., 9 > 55 y.

Discussion

H1: Perceived speaker‘s origin

H2: Perceived speaker age

H3: Place of articulation

Listener‘s response vs. actual speaker‘s origin / speaker age

Mixed models in R

Fixed factors:Stimulus, listener’s origin

Random factors:Listener, word, speaker

Stimulus (pre/post) significant: 𝞆2[1] = 80.2, p < 0.001

No effect of listener’s origin

Fixed factors:Stimulus, listener’s origin

Random factors:Listener, word, speaker

Stimulus (pre/post) significant: 𝞆2[1] = 45.0, p < 0.001

No effect of listener’s origin

No effect of place of articulation (p, t, k) on perceived speaker age(p = 0.12) or perceived speaker’s origin(p = 0.99)

Mixed Models:Speaker’s origin: 𝞆2[1] = 4.4, p < 0.05Stimulus type: 𝞆2[1] = 85.9, p < 0.001

Speakers from

Speaker‘s origin Number of mentions

Pronunciación de la s 24 (39%) Entonación / tono / musicalidad 20 (32%) Las vocales 19 (31%) Aspiración de la s / fonemas finales 8 (13%) En las st (sp, sk) 8 (13%)

Speaker age Number of mentions

Tono de voz 36 (58%) Timbre de voz 19 (31%) Velocidad 10 (16%) Elementos de pronunciación

7 (12%)

Are „Sevillian“ speakers at the same time perceived as being younger? mm = age ~ origin + (1|listener) + (1|speaker) + (1|word)

Pre-aspirated stimuli: no effect of perceived age on dialect judgements (𝞆2[1] = 0.4, p = 0.53)

Post-aspirated stimuli: significant effect of perceived age on dialect judgements:(𝞆2[1] = 14.6, p = 0.001)

Speaker age:𝞆2[1] = 12.4, p < 0.001Stimulus type:𝞆2[1] = 42.8, p < 0.001

Wahrgenommenes Sprecheralter

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

s1s2

young old

Perceived speaker age

estanco

Listeners from

Stimulus: präasp. postasp.

Wahrgenommene Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

s1s2

Wahrgenommene Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

s1s2

Granada Seville

Perceived speaker‘s origin

Listeners from

Stimulus: präasp. postasp.

Wahrgenommene Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

s1s2

Perceived speaker's origin

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

postpre

Wahrgen. Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

postpre Perceived age

Density

0.0000.0050.0100.0150.020

0 50 100

Older

0 50 100

Younger

postpre

Wahrgen. Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

postpre

Seville&(SE)& Granada&(GR)&

Karten:&d3maps.com&

Post3aspira;on&

Pre3aspira;on&

Seville&(SE)& Granada&(GR)&

!"#$%&'()"*

Karten:&d3maps.com&

Post3aspira;on&

Pre3aspira;on&

Pre3aspira;on& Post3aspira;on&

Wahrgen. Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

postpre

Wahrgen. Sprecherherkunft

Density

0.000

0.005

0.010

0.015

0 50 100

Granada

0 50 100

Sevilla

postpre

Manipulation of duration in Praat

Time (s)0.07593 0.3736

0

5000

Freq

uenc

y (H

z)

0.0768044186CBD0026pestana3_roh

Time (s)0.07593 0.3736

-0.06644

0.05463

0

0.0768044186CBD0026pestana3_roh

Time (s)0.07477 0.356

0

5000

Freq

uenc

y (H

z)

0.0761369131CBD0026pestana3_ht

Time (s)0.07477 0.356

-0.06635

0.05478

0

0.0761369131CBD0026pestana3_ht

pre-aspiration A Time (s)

0.07896 0.36150

5000

Freq

uenc

y (H

z)

0.0797886241CBD0026pestana3_th

Time (s)0.07896 0.3615

-0.06644

0.05472

0

0.0797886241CBD0026pestana3_th

post-aspiration B

Lengthening of pre-asp., shortening of post-asp.

Lengthening of post-asp.,

shortening of pre-asp.

References Clopper, C. & D. Pisoni (2004): Some acoustic cues for the perceptual categorization of American English regional dialects. JPhon 32, 111-140. / Foulkes, P. et al. (2011): Sound judgements: Perception of indexical features in children’s speech. In D. Preston & N. Niedzielsky (eds.): A reader in Sociophonetics. Berlin: de Gruyter, 327-356. / Villena Ponsoda, J. & J. Hernández-Campoy (2009): Standardness and nonstandardness in Spain: dialect attrition and revitalization of regional dialects in Spanish. Int. J. Soc. Lang 196/197, 181-214. / Mack, S. & B. Munson (2012): The influence of /s/ quality on ratings of men’s sexual orientation: explicit and implicit measures of the ‘gay lisp’ stereotype. JPhon 40, 198-212. / Ruch, H. (2013): Lautvariation und Lautwandel im andalusischen Spanisch: Prä- und Postaspiration bei /s/ vor stimmlosen Plosiven. Ph.D. dissertation, LMU Munich. / Ruch, H. & J. Harrington (2014): Synchronic and diachronic factors in the change from pre-aspiration to post-aspiration in Andalusian Spanish. JPhon 45, 12-25.

Younger

VTT Closure VOT

Older

Younger

Older

/sp/

/st/

/sk/

/sp/

/st/

/sk/

Eastern Andalusians Western Andalusians

40 ms