Word-final Consonant and Cluster Acquisition in Indian English(es)

In the acquisition of a second language, numerous factors interact simultaneously, including the patterns of the first language (L1), the patterns of the target language (L2), universals of language acquisition, and the amount and type of exposure to the L2. English is learned as a second language in India, where the first languages of the English learners differ widely in the phonotactics of consonants and consonant clusters. Because of the diversity of L1 phonotactics and the relative unity of the L2 target, variations in Indian English can be used to test the effect of the L1 phonotactics on the acquisition of the L2, i.e., the role of transfer. OT accounts have proposed a means for modeling two other important factors in L2 acquisition, in addition to transfer: markedness/universals, and input frequency (e.g., Hancin-Bhatt and Bhatt 1997, Broselow, Chen, and Wang 1998, Hancin-Bhatt 2000, Broselow and Xu 2004, Hansen 2004, Peng and Ann 2004, Barlow 2005). The variations in L2 Indian English word-final consonant and cluster acquisition let us test the success of the OT approach. In this study, data from speakers of five different Indian L1s were recorded speaking Indian English, and their production of word-final consonants and consonant clusters was examined. Their productions of the voicing of final obstruents depended heavily on transfer from their L1 or the emergence of the unmarked, with voiceless obstruents prevailing for speakers of the L1 that did not allow any word-final obstruents. Their production of final consonant clusters also revealed the effects of markedness and transfer, but, in addition, suggests that L2 speakers of English treat final clusters ending in /-s/ as special, just like L1 speakers. Applying the Graduate Learning Algorithm (Boersma 1997, Boersma and Hayes 2001) helps to illustrate that the special treatment of /s/ in clusters cannot result from frequency alone, supporting the claim that Cs clusters should be treated as special in L2 as well as L1 phonology (Yildiz 2005).