Vowel devoicing as gestural alignment


On November 4, 2022 at 13:30, Jacob Aziz will present Vowel devoicing as gestural alignment.

The Zoom link for the meeting is here. The password will be distributed through the mailing list.

Abstract

Vowel devoicing has been variably described as a purely phonetic (e.g., Jun et al., 1997) or phonological (e.g,, Tanner et al., 2019) process. In this work, I look at devoicing data from a dialect of Malagasy (Austronesian, Madagascar) and find that it is accurately modelled using a version of Articulatory Phonology (Browman & Goldstein, 1986) couched in Optimality Theory (Gafos, 2002; Hall 2003). By considering phonological alignment constraints as well as the inherent phonetic properties of Malagasy segments, I am able to account for the variation in devoicing observed in Malagasy. This result shows that vowel devoicing is not necessarily purely phonetic or phonological, but can be a result of multiple factors from within and beyond the phonological grammar.