Doctoral Dissertation Research: Investigating Sound Change in an Understudied Language: A Sociophonetic Study of Age and Locality Effects

Trustees Of Boston University, Boston

Investigators

Abstract

Language change is a fact of life, and among the various types of sound change, vowel merger has featured particularly prominently in linguistic research. Most research on vowel merger has focused on well-studied Indo-European languages such as English, addressing questions related to the production and/or perception of once-distinct vowels that are becoming less distinct (for example, many speakers of US English no longer produce a distinction between the 'cot' and 'caught' vowels, but some of these speakers can nevertheless perceive the distinction). However, investigating the dynamics of vowel merger in less-studied, and non-Indo-European, languages is important to understanding vowel merger because they often contain vowel features that are absent from well-studied languages and, thus, provide new insights into how vowels can change over time. Using a mixed-methods approach, this project examines a case of vowel merger observed in an under-studied language, which contains a vowel feature absent from Indo-European languages. The findings of the project contribute to scientific knowledge of language variation, contact, and change in a multilingual, understudied society, of the relationship between patterns of language variation and the social and attitudinal associations of speech, and of the basis for identifying vowel mergers in languages with unique vowel features. By collecting speech production and perception data across a range of speaker ages and localities, the project also has implications for speech-language pathology, language pedagogy, speech recognition technologies, and forensic linguistics. This project explores patterns of variation that have been implicated in previous cases of vowel merger but focusing on a rarer case of vowel merger. In particular, the project examines whether demographic and social factors, such as age, gender, and locality, help predict the extent of merger in this case; whether a vowel merger in production corresponds to one in perception; and whether a vowel merger that is apparent according to the traditional vowel height and vowel backness dimensions also appears consistently in other, less frequently examined acoustic measures. The project addresses these questions by collecting production and perception data from two different communities (urban versus rural), including adolescents (ages 10-18 years), younger adults (19-39 years), and older adults (40+ years). Production data come from conversational interviews and a picture-naming task; perception data come from a listening task. Audio recordings from the production tasks undergo acoustic analyses, which help determine whether there is a vowel merger in production and support exploration of additional diagnostics for vowel merger. Study participants also complete a detailed questionnaire that includes questions about language/dialect contact and language attitudes. The project results form the basis of the first open-access corpus of the language being studied, a tool for language researchers and a source of information about various speech norms in an under-studied language. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.