Caroline R. Wiltshire

Pitch accent in Indian-English teaching discourse

The phonetic correlates of accent/stress distinguish Indian English (IE) from American dialects (Spencer, 1957; Kachru, 1983). We examine the realization of accent in IE compared to American English (AE) produced by teaching assistants in similar contexts. In teaching discourse, we find that a lexically accented syllable is often realized in IE with a relative drop in frequency and without a reliable increase in amplitude. In similar contexts, lexically accented syllables in AE reliably increase in both frequency and amplitude. Following the distinction made in Beckman (1986), we conclude that IE acts as a pitch-accent language rather than as a stress-accent language like AE. We also suggest a source for the distinct prosodics of IE: some Indian languages use a low pitch on accented syllables (Mohanan, 1986; Hayes and Lahiri, 1991; Harnsberger, 1999). We investigate the effect of different first languages on the production of IE using three Indian teaching assistants with different L1 (Bengali, Tamil, Hindi-Urdu), and compare their IE discourse to L1 sentences. The similarity of the results for three different L1 suggests that the phonetic correlates of accent in IE are common to Indian languages.

Phonetic stress in Indian English vs. American English

A recent article in World Englishes (Pickering and Wiltshire, 2000) suggests that one difference between Indian English (IE) and American English (AE) varieties is the phonetic realization of prominence. In IE, the syllable considered prominent by Pickering and Wiltshire falls in pitch, while the amplitude changes little. In this study, we show that the results of Pickering and Wiltshire (2000) may be due to a misinterpretation of the positioning of stress in IE. We show that by considering the louder syllable to be the stressed one, IE stress correlates differ in magnitude but not in direction from AE. However, our results confirm Pickering and Wiltshires in that the phonetic correlates of AE and IE stressed syllables do differ; furthermore, the position of stress may differ, and hence the potential for communication problems between speakers of different varieties remains.

Extending ALIGN constraints to new domains

Alignment theory proposes that constraints on the edges of morphological and prosodic categories interact with each other and with phonological constraints (McCarthy and Prince 1993a). Ito and Mester (1994) extend the family of alignment constraints to relate prosodic categories (syllables) and subprosodic categories (segments and features). This paper further investigates the sensitivity of phonological constraints to edges and argues for extending alignment constraints to prosodic categories larger than the syllable and word (Nespor and Vogel 1986; Selkirk 1986, 1995 ; Inkelas 1989), in order to constrain phrase edges and their interaction with other domain edges and subprosodic units. I argue for such alignment constraints based on syllabification that crosses word boundaries, where words cannot be correctly syllabified in isolation from their phrasal context. Data from the phonologies of Cairene Arabic and Tamil show that the constraints on syllables, words, and phrases are evaluated in parallel, accounting for the locations of epenthetic, deleted, and assimilated segments. Syllabification must be sensitive to the simultaneous evaluation of constraints on all these domains, in order to provide the context for alternations within domains and at their edges. Parallel evaluation in multiple domains dispenses with stages of syllabification and eliminates derivationalism from the distinction between lexical and post-lexical levels of analysis. The analysis is developed within the optimality theory of Prince and Smolensky (1993) and the alignment theory of McCarthy and Prince (1993a)

A reconsideration of universal properties of tonal coarticulation

Although many languages show similar properties in tonal coarticulation, some languages examined recently exhibit properties against previous findings. This study performed statistical analysis including one-way repeated measures ANOVA with Greenhouse-Geisser adjusted values and linear mixed effects models to explore properties of tonal coarticulation in Nanjing Chinese. The results show similar magnitude in carryover and anticipatory effects, inconsistent with most languages examined, and the results for the H/L asymmetry in Nanjing Chinese are not consistent with the literature either. The findings in this study and other recent studies contrary to previous beliefs call for further studies to enhance our understanding of the universality of tonal coarticulation.