In their landmark sociolinguistic study Indian English, Sahgal and Agnihotri (1988) were the first to show that loss of rhoticity was increasing in prestige, with younger, English-medium educated, female speakers preferring null to a trill or tap in post-vocalic position. Sahgal and Agnihotri follow the classic descriptions of Indian English (Bansal 1990, Kachru 1994) on the trilled quality of postvocalic /r/ and do not differentiate between any postvocalic /r/s in their study. Despite their prediction that non-rhoticity would spread even further in Indian English, more recent studies of English-medium educated speakers in India show diverse results. Wiltshire and Harnsberger (2006) found that the speakers of their study from Tamil and Gujarati L1 backgrounds were only 16% rhotic, but that the nature of this /r/ differed for the two groups. Chand (2010) however shows more than 50% approximant /r/ in postvocalic position among Delhi-based Hindi-English bilinguals alongside null and trill. This is a striking finding because the approximant /r/ may need to be treated as a recent import i.e. from American English. The present study investigates /r/ among Hindi speakers from Dehradun, and Marathi speakers from Pune, and finds that speakers are indeed predominantly non-rhotic, but where there is postvocalic /r/ this is consists of trills, taps and derhotacized taps; Approximant /r/ appears predominantly before retroflex consonants, and thus it is debatable whether approximant postvocalic /r/ exists in Indian English or whether it is an artefact of the following retroflex consonant. Issues such as the quality of postvocalic /r/, the influence of the L1, and the linguistic environment for postvocalic /r/ all issues that have to be considered carefully in the study of rhoticity in Asian Englishes more broadly (Sharbawi and Deterding 2010, Tan 2012) if we are to properly differentiate between local and global influences on rhoticity.
|Effective start/end date||1/07/16 → …|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.