Multilingualism as an identity signal: a focus on phonetic variation in Asian communities in English speaking countries

(Submitted on May 2020 for UCL PLIN0032 Coursework)

Research in sociolinguistics has demonstrated that speakers use Fine Phonetic Detail (FPD) to establish group boundaries and personal identities. This can also be seen in multilingual communities where issues of identity and culture are major as being multilingual adds on to the groups and potential identities that one can belong to. Switching between identities can impact the use of language. Moreover, when languages are in contact, linguistic variations are likely to happen. In this essay, I will explore the phonetic and phonological variation in multilingual communities and show that these variations are affected by social factors. I would like to focus on two Asian communities in English speaking countries, British Indians and American Chinese. First, I will analyse the variation of retroflex stops in Punjabi-English bilinguals in London. I will also look at Indian accents in Glasgow Asians. Then, I will explore the raising of /ɔ/ vowel by American Chinese in New York City. Crucially, I argue that the use of phonetic and phonological variation in multilingual communities are affected by their motivation to signal a social or ethnic identity.

To start with, I will look at South Asian community that mainly originated from Pakistan and India which is the largest ethnic minority in the United Kingdom. A huge population of South Asians is found in Southhall, West London and most of them are Punjabi (one of the languages in India) speakers. A distinct feature found in the English of Punjabi-English bilinguals is the use of retroflex /t/. As alveolar stops /t, d/ are not present in Punjabi, they are commonly replaced by the retroflex stops /ʈ, ɖ/ in Indian Englishes (Bhatia, 2003; Sailaja, 2009). This has made retroflexion a highly salient phonetic feature in the Indian community (Sharma, 2011). Sharma & Sankaran (2011) compared the use of retroflex /t/ by extracting acoustic data from interviews in three groups of Punjabi-English speakers in Southall: non-native Indian-born group (Gen 1), older native British-born group (older Gen 2) and younger native British-born group (younger Gen 2). They found that the rate of use of retroflex /t/ has been declining over time (Figure 1), but it resembles a development of new variant in the community’sspeech. Their results show both Gen 2 groups have successfully acquired the British English glottal stop (Figure 2), with older Gen 2 producing a lower rate of glottalisation as older Gen 2 constantly switch between the Indian (/ʈ/) and British variant (/ʔ/). Figure 3 summarises the distribution of retroflex /t/ and glottal stop (British English variant) in the three groups.




Figure 1. Percentage use of retroflex /t/ in the three groups. (Taken from Sharma & Sankaran, 2011)

Figure 2. Percentage of /t/ glottaling and their distribution by group. Both Gen 2 groups produced significantly more glottal stops than Gen 1 and showed similar pattern of distribution to native monolingual British English speakers. (Taken from Sharma & Sankaran, 2011)


Figure 3. Acquisition of Indian English and British English traits over time. (Taken from Sharma & Sankaran, 2011)


By analysing the content of interview, the authors found that the older Gen 2 group have experienced racism at school which caused them to ‘hide’ their Indian features andtune in to the British variants. Indian variants were only used within their family. In contrast, when the younger Gen 2 group were born, there were less racism and a strong sense of belonging to the local community. This encouraged the younger generation to integrate Punjabi features into British English to signal a new form of social identity, resulting in the relocation of retroflex /t/ in their English. They patterned with British English speakers in most of their speech but Punjabi variants were scattered saliently. The finding that social identification affects the use of Asian accents in Punjabi-English bilinguals is consistent with Hirson & Sohail (2007)’s finding where participants who self-identified as ‘Asian British’ produced more rhotic forms of /r/ (a distinct feature inIndian languages) than those who self-identified as ‘British’. In short, although there is acorrelation between age and the use of Indian accent features, social network and identity signalling are the crucial factors that affect the use of phonetic variation in English of British Indians.

In addition, Lambert et al. (2007) have investigated British Asian accents in Glasgow by comparing acoustic analysis in Glaswegian of Glaswegian-Punjabi/Urdu bilinguals and Glaswegian monolinguals. He found distinct phonetic features present in the speech of Glasgow Asians that are absent in monolingual speakers, which confirms the presence of Glasgow Asian accents. Crucially, the accent variation serves as a social indexical function for Asian high school girls. Postalveolar variants which are common in Asian accent features were favoured when talking about topics related to home or cultural practices. There is also a link between social practices and accent variation in which girls who participate in more westernise practices use the postalveolar variants the least. This is another evidence that phonetic variations are used to form social identities and boundaries.

Moving on, I am going to look at Chinese Americans, which is the largest ethnic group of Asian Americans in the United States who mostly migrated from Guangdong province of China since 1840s. The New York metropolitan area is home to the largest Chinese American population. Wong (2015) has studied the use of New York City English (NYCE) features by second generation Chinese Americans in depth. One key phonological feature in NYCE is the raising of /ɔ/ vowel (as in thought, caught) to maintain the distinction between the low back vowels /ɔ/ and /ɑ/ (as in lot, cot) which has emerged in the mid-twentieth century (Berger, 1968; Labov, 2006[1966]; Labov et al., 2006). By analysing acoustic data from interviews and reading tasks, Wong found that the raising of /ɔ/ is modulated by age (Figure 4). Older Chinese Americans tend to produce more raised /ɔ/ and patterned with age-controlled NYCE speakers. In addition to age, she found a main effect of ethnic orientation on the use of raised/ɔ/. Ethnic orientation score is calculated by taking into account the use of heritage language, familiarity with heritage culture and social network circle of the speaker. Speakers that have a strong ethnic orientation have a tendency to produce non-raised /ɔ/ (Figure 5). This finding is consistent with her previous study (Wong, 2010) that showed the variation of raised /ɔ/is connected to the speakers’ social network and lifestyle orientation. Speakers with lessChinese ties in their social network and an American-oriented lifestyle tend to produce more raised/ɔ/ than those who have a Chinese-oriented social circle and lifestyle. This suggests that the NYCE feature of raised /ɔ/ is an index of local non-Chinese identity. In contrast, Chinese Americans with stronger ethnic orientation are likely to stay within their ethnic community which results in less exposure to the raised/ɔ/ produced by NYCE speakers. Therefore, the variation in raised/ɔ/ seen in the older generation of Chinese American in New York City can be a marker for forming identities.

Figure 4. Mean normalized height of the vowel /ɔ/ according to year of birth. Higher value of F1 indicates a more raised /ɔ/. Blue circles indicate Chinese American speakers (N=32). Grey diamonds indicate monolingual NYCE speakers. Red dotted line indicates the cut-off point for raised /ɔ/.


Figure 5. Height of the vowel /ɔ/ by ethnic orientation among the older speakers (born before 1980). Higher value of F1 indicates a more raised /ɔ/. Red line indicates the cut-off point for raised /ɔ/.

However, this distinction is not found in younger Chinese Americans. Recall in Figure 4, we see a change towards non-raised/ɔ/ in apparent time. While older Chinese Americans use regional dialect features to form social identities, younger generation do not produce raised/ɔ/ regardless of their ethnic orientation. In fact, recent work has shown that there is a reversal in raised /ɔ/ in NYCE (Becker, 2014). This is mainly due to the social meaning of raised /ɔ/. Wong & Hall-Lew (2014) found that Chinese Americans are aware that the use of raised /ɔ/ is tied to a stereotype of old, white ethnic, workingclass ‘New Yorker’. The negative attributes linked to raised /ɔ/ might have impacted the younger generation to stray away from raising /ɔ/. This has shown that younger generation of immigrants are also actively participating in the regional dialect change.

To sum up, by looking at two multilingual communities, we have seen that socialfactors can affect the use of phonetic and phonological variations in the speaker’slanguage. The language used by different generations of immigrant can have significant difference. Apart from age, social network and identity signalling are crucial factors that caused the variation in multilingual communities. In this essay, we first looked at how ethnic phonological features can blend into the local language to signal a unique multilingual identity. We then see that immigrants may adopt regional dialects in order to blend in to the local speech community. As we are moving towards a global community where more languages and cultures are in contact with each other, linguistic variation is inevitable. More research has to be done to understand the language change found in younger generations that might be signalling a new form of social identity which does not form a clear distinction between cultures. Studying these variations can tell us more about how language change diffuses through a community and allow us to improve on existing theories of linguistics.

References

Becker, K. (2014). The social motivations of reversal: Raised BOUGHT in New York City English. Language in Society,43(4), 395-420. doi:10.1017/s0047404514000372

Berger, M. D. (1968). The Internal Dynamics of a Metropolitan New York Vocalic Paradigm. American Speech,43(1), 33. doi:10.2307/455374

Bhatia, T. K. (2003). Punjabi: A cognitive-descriptive grammar. London: Routledge. Hirson, A., & Sohail, N. (2007). Variability of rhotics in Punjabi-English bilinguals. In 16th International Congress of Phonetic Sciences (ICPhS).

Labov, W. (2006[1996]). The social stratification of English in New York City(2nd ed.). Cambridge: Cambridge University Press.

Labov, W., Ash, S., & Boberg, C. (2006). Atlas of North American English: Phonetics, phonology and sound change. Berlin: Mouton de Gruyter.

Lambert, K., Alam, F., & Stuart-Smith, J. (2007). Investigating British Asian accents: Studies from Glasgow. In 16th International Congress of Phonetic Sciences (ICPhS).

Sailaja, P. (2009). Indian English. Edinburgh: Edinburgh University Press.

Sharma, D. (2011). Style repertoire and social change in British Asian English. Journal of Sociolinguistics,15(4), 464-492. doi:10.1111/j.1467-9841.2011.00503.x

Sharma, D., & Sankaran, L. (2011). Cognitive and social forces in dialect shift: Gradual change in London Asian speech. Language Variation and Change,23(3), 399-428. doi:10.1017/s0954394511000159

Wong, A. W. (2010). New York City English and second generation Chinese Americans. English Today,26(3), 3-11. doi:10.1017/s0266078410000167

Wong, A. W. M. (2015). Diverse Linguistic resources and multidimensional identities: A study of the Linguistic and identity repertoires of second generation Chinese Americans in New York City (Doctoral dissertation, New York University).

Wong, A. W., & Hall-Lew, L. (2014). Regional variability and ethnic identity: Chinese Americans in New York City and San Francisco. Language & Communication,35, 27- 42. doi: 10.1016/j.langcom.2013.11.003