Vowel epenthesis in Bangla: An Optimality Theory analysis

This paper examines the occurrence of epenthetic vowels before and between the initial consonant clusters in Bengali speakers of English, and provides an Optimality Theory (OT) analysis to account for this phenomenon. Native Bengali words disallow initial consonant clusters, and many word-initial consonant clusters in loan words are simplified according to these phonotactics. The maximum syllabic structure is CVC in Bengali and speakers often carry this restriction over to loan words. For example, geram (CV.CVC) instead of gram (CCVC) for the Sanskrit loan word village, or iskul (VC.CVC) instead of skul (CCVC) for the English word school (Kar, 2009). I argue that in rising sonority clusters, a vowel is inserted between the two consonants and in falling sonority clusters (i.e., [s]-stop clusters) the vowel is inserted before the consonant cluster. I also explain that the sonority sequencing constraint SYLLABLE CONTACT treats [s]-stop clusters differently from obstruent-sonorant clusters, and the differing epenthesis pattern can be explained properly if it is considered an effect of SYLLABLE CONTACT – the preference of sonority to fall across a syllable boundary, which was proposed by Murray and Venneman (1983) and also supported by Gouskova (2001). With tableaux, I demonstrate that the epenthesis in consonant clusters is caused by the prohibition on consonant clusters in Bengali and the site of epenthesis is determined by SYLLABLE CONTACT (Gouskova, 2001). I also demonstrate that the constraint that prefers epenthesis before the [s]stop cluster is CONTIG-IO (Kager, 1999). Furthermore, I propose that apart from SYLLABLE CONTACT, two other constraints *OO and *OR can also account for the vowel epenthesis in Bengali.