It’s often thought that as they grow up, the children of immigrants begin to sound like their locally-born friends rather than their parents. Devyani Sharma and Lavanya Sankaran, though, found that things are more complex than this – language change between different generations is more gradual than might be expected, and it’s also more complex.
Sharma and Sankaran worked in the Punjabi community in Southall, London, where, over the course of the last 60 years, South Asians have shifted from being a minority group to a majority one which now makes up more than 60 per cent of the local population. The researchers analysed the English of three groups of South Asians, totalling 42 individuals. One group consisted of first generation immigrants who had migrated from India as adults, and the two other groups were locally-born second generation South Asians, one older (aged between 35 and 60) and one younger (aged between 18 and 35). The older second generation group had grown up in Southall at a time when South Asians were still a minority group there and when race relations in the area were hostile. By the time the second, younger, group (aged 18-35) was growing up, South Asians were no longer such a minority in Southall and, perhaps as a result, race relations had shifted to a cooperative coexistence.
The researchers focussed on the pronunciation of /t/, which has a distinctive local pronunciation as well as a South Asian pronunciation. The local London pronunciation of /t/ is glottalised (with the pronunciation of words like water or feet sometimes represented in popular writing as wa’er and fee’). As you might expect, the first generation South Asian speakers had almost no glottalised pronunciations of /t/. By contrast, both second generation groups used glottalised /t/; furthermore, they followed the same pattern, using this pronunciation more often at the end of a word than the middle of a word (so, more often in feet than water). In their use of glottalised /t/, then, the second generation were speaking more like locally-born people of their age than their parents – just as we might expect.
However, the South Asian speakers sometimes pronounced /t/ as a retracted or retroflex consonant, as in Punjabi, the Indian language that they also spoke. Here the tip of the tongue is curled back to touch the ridge just behind the top teeth (or close to the ridge). You can hear this pronunciation in the stereotyped English of Apu, the Indian immigrant in The Simpsons. The first generation immigrant group used retroflex /t/ 35 per cent of the time. The second generation groups also used this pronunciation, albeit less often: 16 per cent of the /t/’s in the English of the older second generation were retroflex, and 8 per cent in the English of the younger speakers. The second generation, then, had not altogether abandoned the pronunciation of their parents: although language change was taking place across the generations in these immigrant families, it was a more gradual process than is often supposed.
The change was also more complex than expected. Unlike both their parents and the older second generation group, the younger speakers used retroflex /t/ more often at the beginning of a word, where it is more noticeable (for example, in tea or toffee). They also pronounced it with a “fortis” (more energetic) phonetic quality.
In interviews with the researchers younger second generation male speakers used retroflex /t/ more often than younger female speakers Even here, though, the picture is more complicated than this gender difference suggests. Female speakers used a surprisingly high number of pronunciation features influenced by Punjabi, including retroflex /t/, when they were speaking English at home. For female speakers, then, there seems to be a sharper compartmentalisation of styles across their repertoire.
Sharma and Sankaran point out that other pronunciation features pattern in a similar way in the English of these three groups of speakers. They explain that for the older second generation group, surviving at school and in public meant they had to downplay Indianness and pass as British, so they acquired local pronunciations and weakened their use of South Asian ones. Many individuals in this group then went into their fathers’ businesses and had continuing ties with India. Depending on where they were and who they were talking to, they needed to signal that they belonged either to a British or an Indian group. As a result, they were able to control two distinct pronunciations of English. The younger generation not only had less regular contact with India, but by the time they were growing up race relations in the area were less hostile, so they did not need to try to pass as British. Instead, using a focused, Punjabi-inflected speech style allows them to signal their allegiance to the now sizeable local British Asian community.
Sharma and Sankaran note that in immigrant communities elsewhere – in North America, for example – there may be more rapid assimilation to local patterns of pronunciation since, as they have shown, linguistic assimilation depends in part on social factors such as community relations and the size of the migrant community.
Devyani Sharma and Lavanya Sankaran (2011) Cognitive and social forces in dialect shift: Gradual change in London South Asian speech. Language Variation and Change 23: 399-428.
This summary was written by Jenny Cheshire