[slides] Prosodic Rhythm in Asian American English

Prosodic rhythm in Asian American English Carina Bauman New York University 23 October 2015 NWAV 44

Prosodic rhythm • Languages can be characterized as relatively “stress-timed” or “syllable-timed” • Stress-timed languages, like English or German, are often described as having a “Morse code” rhythm • Stressed syllables are much longer than unstressed syllables

• Syllable-timed languages, like Spanish or Cantonese, are often described as having a “staccato” or “machine gun” rhythm • Adjacent syllables are roughly the same duration

2

Prosodic rhythm • Impressionistically, different dialects of English also have different rhythmic properties • Hall-Lew 2009 (p. 75): “There was a friend of mine ... we used to work together. He spoke with an accent ... as though he was an immigrant. But he was born and raised I think pretty much in Chinatown. ... So when you start talking, what, staccato and choppy? He kind of fit that.”– Sal (male, age 45, Chinese American) • Lim 2010 (p. 148): “where an RP [Received Pronunciation] pronunciation gives a single firm stress, as in a word like specification, ESM [English of Singapore and Malaysia] will tend to give the stressed syllable somewhat less stress and an unstressed syllable somewhat more – specification- or even approximately equal stress in a two-syllable word like cab-bage” (Tongue 1979: 31) • Shousterman 2015 (p. 158): While describing a particularly tense situation, Butchie [a Puerto Rican English speaker] imitates his brother’s reaction to being told to “quiet down”, saying (in an angry tone), “I don’t give a hoot and pa pa pla, you know?”. The pa pa pla in this instance (used to denote meaningless chatter, like the more mainstream blah blah blah) is produced with an evenly timed, staccato-like execution. 3

Prosody & ethnicity • Rhythmic variation in English is often tied to substrate effects • E.g., Singapore English is more syllable-timed than British English because of influence from Chinese and Malay

• Rhythm can also become a marker of ethnicity separate from a substrate effect • This is what Coggshall (2008) found for younger Lumbee Indians in North Carolina. The Lumbee are monolingual English speakers, yet the younger group showed a stress-timed pattern similar to Cherokee English speakers.

Pairwise Variability Index (PVI) • Originally developed by Low, Grabe, & Nolan (2000) to capture the difference between British English & Singapore English • PVI is calculated by taking the difference in duration between two adjacent syllables, divided by the mean duration of the two

5

Pairwise Variability Index (PVI) • Modified by Thomas & Carter (2006) for spontaneous speech data 𝐵−𝐴 𝑃𝑉𝐼 = (𝐴 + 𝐵)/2

• Where A is the duration of the first syllable and B is the duration of the second syllable • Do this for 200 syllable pairs, then take the median value • Lower PVI = more syllable-timed; higher PVI = more stress-timed

6

PVI of English varieties: Hispanic English Thomas & Carter 2006:

Shousterman 2015

7

PVI of English varieties: AIE, Singapore Coggshall 2008

Lim 2010

Asian American English • Ethnicity has been an important variable in the study of linguistic variation, yet until recently Asian Americans have been relatively understudied. • This is changing! See: Hall-Lew (2009), Wong (2014), Nguyen (2015)

• There’s evidence that, at least sometimes, listeners can racially identify Asian Americans from hearing them speak (Hanna 1997, Newman & Wu 2011) • But what are the cues? This has been much harder to pin down. • Perhaps prosody, which has been less well studied than segmental variables, plays a part. 9

The present study • Data collection: ethnographic observations & sociolinguistic interviews w/ 31 college students • The university: a large public school in New Jersey (NJU)* • The sorority: a chapter of Alpha Zeta Nu (AZN), an international Asianinterest sorority with ~50 chapters in the U.S. and Canada

*All names of organizations and individuals are pseudonyms 10

Sorority members Speaker Peggy Kitty Wanda Molly Crystal Pepper Jean Gertrude Maria Shanna Nico Lady Felicia Carol Selene Emma Jubilee Jeansun Laura

Age 21 22 21 21 22 22 22 21 21 21 21 23 19 22 20 21 21 22 24

Ethnicity Chinese Korean Chinese Chinese Chinese Chinese Korean Chinese Vietnamese Chinese Vietnamese Chinese Chinese Chinese Filipina Vietnamese Vietnamese Korean Chinese

Hometown Manchester, NH St. Paul, MN NYC NYC NYC NYC Fort Lee, NJ Fairfield, NJ Piscataway, NJ Marlboro Township, NJ Bridgewater ,NJ Parsippany, NJ Holmdel, NJ Holmdel, NJ Tom's River, NJ Egg Harbor Township, NJ Egg Harbor Township, NJ Egg Harbor Township, NJ Hockessin, DE 11

Other interviewees Speaker Gwen Janet Doreen Anna Marie Mary Jane Barbara Natasha Maya Elektra Karima Lucy Karla

Age 18 18 18 20 19 18 20 18 21 19 18 18

Ethnicity White Korean White Chinese/White Vietnamese Chinese Chinese African American/White White Indian Vietnamese Chinese

Hometown Vernon, NJ Palisades Park, NJ Fair Lawn, NJ Denville, NJ Dunellen, NJ Hillsborough, NJ Morganville, NJ Long Branch, NJ Mt Holly, NJ Galloway, NJ Egg Harbor Township, NJ Orange County, CA

12

Possible substrate influences • The speakers' heritage languages include several varieties of Chinese (Mandarin, Cantonese, Shanghainese, Fujianese, Wenzhounese), Vietnamese, Korean, Tagalog, and Gujarati. • Chinese and Vietnamese are both consistently described as syllable-timed (e.g., Nguyen 2004, Lim 2010, Mok 2008). • The status of stress in Korean is rather controversial (see Arvaniti 2012 for discussion), but recent studies have tended to conclude that Korean is more likely syllable- or mora-timed than stress-timed (Jeon 2006, Kim et al. 2008, Mok & Lee 2008, Moon-Hwan 2004). • Comparatively little has been written about Tagalog prosody, but it has also been argued to be syllable-timed (Montanari 2006:44, Santos & Guevara 2011). • Gujarati has also been described as syllable-timed (Joshi 2014).

PVI results Sorority members Speaker N PVI (median) SD Carol 286 0.332 0.291 Emma 206 0.368 0.343 Felicia 211 0.371 0.333 Wanda 216 0.373 0.294 Kitty 209 0.383 0.356 Selene 214 0.385 0.314 Pepper 211 0.415 0.357 Jubilee 213 0.417 0.341 Shanna 219 0.428 0.331 Crystal 221 0.436 0.372 Jean 207 0.440 0.342 Molly 323 0.444 0.352 Maria 204 0.451 0.360 Jeansun 215 0.462 0.354 Peggy 227 0.467 0.340 Gertrude 217 0.500 0.377 Laura 214 0.507 0.361 Lady 222 0.519 0.424 Nico 210 0.528 0.355 0.433 0.347

Non-sorority members (non-Asian) SpeakerN PVI (median) SD Gwen 208 0.382 0.324 Elektra 203 0.391 0.355 Doreen 226 0.448 0.343 Maya 218 0.483 0.345 0.426 0.342

Non-sorority members (Asian) N PVI (median) SD Speaker Janet 213 0.384 0.318 Natasha 216 0.394 0.336 Barbara 214 0.4 0.318 Karima 206 0.4 0.325 Mary Jane 209 0.414 0.359 Lucy 201 0.416 0.336 Anna Marie 247 0.429 0.323 Karla 215 0.431 0.3316 0.409 0.331 14

PVI results PVI (median) 0.600

0.500

0.400

0.300

0.200

0.100

0.000

Stylistic effects? • Lim (2010), Arvaniti (2012) have noted that different speech styles may show different rhythmic patterns • Because of my ethnographic work with the sorority members, they were already somewhat familiar with me by the time of the interview. • The non-sorority members, in contrast, were recruited through social media and did not know me before the interview. If they were speaking more formally, with less vowel reduction, this might have led to lower PVI scores.

• There is also the possibility of intraspeaker stylistic variation in PVI, which I did not control for in my interviews.

PVI of English varieties revisited Minimum, Mean, & Maximum PVI 0.7 0.65 0.6 0.55 0.5 0.45 0.4 0.35 0.3 0.25 0.2

PVI x speech rate Thomas & Carter (2006), Coggshall (2008), and Lim (2010) all considered a possible effect of speech rate on PVI • Thomas & Carter (2006) found that for Spanish and Hispanic English speakers, PVI increased with increased syllable duration, but for European and African American speakers there was no correlation median PVI x median syllable length 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.05

0.055

0.06

0.065

0.07

0.075

0.08

0.085

PVI x Age of Acquisition (sorority members) 0.55 0.53 0.51

Laura Gertrude

0.49

Median PVI

Nico

Lady

0.47

Peggy Jeansun

0.45

Maria

Molly

Jean Shanna

0.43

Crystal Jubilee Pepper

0.41 0.39

Selene Kitty Wanda Emma

Felicia

0.37 0.35 0

1

2

3

4 5 6 Age of English Acquisition

7

8

9

10

The red horizontal line represents the mean PVI (0.43), while the red vertical line represents the median age of acquisition (4.5 years) 19

Conclusions • On average, Asian American speakers do show lower PVI values than what has been reported for European and African American English, indicating more syllable-timed speech. • On the other hand, some of the non-Asian speakers in this study also showed quite low PVIs. Could this be an effect of style? Region? Age?

• Asian American speakers who acquired English later tend to be more syllable-timed, supporting a hypothesized substrate effect. • Open questions: How well does PVI correlate with perceived rhythm? Besides vowel duration, what other prosodic variables may contribute to the perception of rhythm?

Acknowledgements • This work was supported by an NSF Doctoral Dissertation Grant (#1349147). • Thank you to my committee: Greg Guy, John Singler, Renee Blake, Sonia Das, and John Baugh. • Thank you to Cara Shousterman, Libby Coggshall, and Sean Martin for invaluable help with PVI!

References •

Arvaniti, Amalia. 2012. The usefulness of metrics in the quantification of speech rhythm. Journal of Phonetics 40 (3): 351-73.

•

Coggshall, Elizabeth L. 2008. The Prosodic Rhythm of Two Varieties of Native American English. University of Pennsylvania Working Papers in Linguistics 14(2).

•

Hall-Lew, Lauren. 2009. Ethnicity and phonetic variation in a San Francisco neighborhood. PhD dissertation, Stanford University.

•

Hanna, David B. 1997. Do I sound "Asian" to you?: Linguistic markers of Asian American identity. University of Pennsylvania Working Papers in Linguistics 4(2). 141-53.

•

Jeon, H-S. (2006). Acoustic measure of speech rhythm: Korean learners of English. Unpublished M.Sc. dissertation, University of Edinburgh.

•

Joshi, Paresh. 2014. Phonological Contrastive Analysis of Supra-Segmental Features RP and GIE with Special Focus on Gujarati Phonology. International Journal for Teachers of English 4(3):112-8.

•

Kim, J., Davis, C., & Cutler, A. (2008). Perceptual tests of rhythmic similarity: II. Syllable rhythm. Language and Speech, 51(4), 343-359.

•

Lim, Laureen T. C. 2010. The Limits of Linguistic Variation: Stylistic Diversity in Singapore English. PhD dissertation, New York University.

•

Low, Ee L., Esther Grabe & Francis Nolan. 2000. Quantitative characterizations of speech rhythm: Syllable-timing in Singapore English. Language and Speech 43. 377-401.

•

Mok, P. P. K., & Lee, S. I. (2008). Korean speech rhythm using rhythmic measures. Paper presented at the 18th International Congress of Linguists (CIL18). Seoul, Korea.

•

Mok, P. (2008) On the syllable-timing of Cantonese and Beijing Mandarin. In Proceedings of the 8th Phonetics Conference of China (PCC 2008) and the International Symposium on Phonetic Frontiers (ISPF 2008). Beijing.

•

Montanari, Simona (2006). Language differentiation in early trilingual development: evidence from a case study. PhD dissertation, University of Southern California.

•

Moon-Hwan, C. (2004). Rhythm typology of Korean speech. Cognitive Processing, 5, 249-253.

•

Newman, Michael & Angela Wu. 2011. “Do you sound Asian when you speak English?” Racial identification and voice in Chinese and Korean Americans’ English. American Speech 86(2). 152-78.

•

Nguyen, Emily. 2015. Ethnic identity and /æ/-raising among Vietnamese Americans. Paper presented at LSA 2015, Portland, OR.

•

Nguyen, Thi Anh Thu (2004). Prosodic transfer: the tonal constraints of Vietnamese acquisition of English stress & rhythm. PhD dissertation, University of Queensland.

•

Santos, Timothy Israel D. and Rowena Cristina L. Guevara (2011). Classification of Filipino speech rhythm using computational and perceptual approach. Proceedings of the 25th Pacific Asia Conference on Language, Information and Computation, 450-8.

•

Thomas, Erik R. & Phillip M. Carter. 2006. Prosodic rhythm and African American English. English World Wide 27(3). 331-55.

•

Wong, Amy Wing-mei. 2014. Diverse Linguistic Resources and Multidimensional Identities: A Study of the Linguistic and Identity Repertoires of Second Generation Chinese Americans in New York City. PhD dissertation, New York University.