Date of Award

Spring 2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Linguistics

First Advisor

Shaw, Jason

Abstract

This dissertation examines how lexical tone can be represented with articulatory gestures, and the ways a gestural perspective can inform synchronic and diachronic analysis of the phonology and phonetics of a language. Tibetan is chosen an example of a language with interacting laryngeal and tonal phonology, a history of tonogenesis and dialect diversification, and recent contact-induced realignment of the tonal and consonantal systems. Despite variation in voice onset time (VOT) and presence/absence of the lexical tone contrast, speakers retain a consistent relative timing of consonant and vowel gestures. Recent research has attempted to integrate tone into the framework of Articulatory Phonology through the addition of tone gestures. Unlike other theories of phonetics-phonology, Articulatory Phonology uniquely incorporates relative timing as a key parameter. This allows the system to represent contrasts instantiated not just in the presence or absence of gestures, but also in how gestures are timed with each other. Building on the different predictions of various timing relations, along with the historical developments in the language, hypotheses are generated and tested with acoustic and articulatory experiments. Following an overview of relevant theory, the second chapter surveys past literature on the history of sound change and present phonological diversity of Tibetic dialects. Whereas Old Tibetan lacked lexical tone, contrasted voiced and voiceless obstruents, and exhibited complex clusters, a series of overlapping sound changes have led to some modern varieties that are tone, lack clusters, and vary in the expression of voicing and aspiration. Furthermore, speakers in the Tibetan diaspora use a variety that has grown out of the contact between diverse Tibetic dialects. The state of the language and the dynamics of diaspora have created a situation ripe for sound change, including the recombination of elements from different dialects and, potentially, the loss of tone contrasts. The nature of the diaspora Tibetan is investigated through an acoustic corpus study. Recordings made in Kathmandu, Nepal, are being transcribed and forced-aligned into a useful audio corpus. Speakers in the corpus come from diverse backgrounds across and outside traditional Tibetan-speaking regions, but the analysis presented here focuses on speakers who grew up in diaspora, with a mixed input of Standard Tibetan (spyi skad) and other Tibetan varieties. Especially notable among these speakers is the high variability of voice onset time (VOT) and its interaction with tone. An analysis of this data in terms of the relative timing of oral, laryngeal, and tone gestures leads to the generation of hypotheses for testing using articulatory data. The articulatory study is conducted using electromagnetic articulography (EMA), and six Tibetan-speaking participants. The key finding is that the relative timing of consonant and vowel gestures is consistent across phonological categories and across speakers who do and do not contrast tone. This result leads to the conclusion that the relative timing of speech gestures is conserved and acquired independently. Speakers acquire and generalize a limited inventory of timing patterns, and can use timing patterns even when the conditioning environment for the development of those patterns, namely tone, has been lost.

COinS