An autosegmental/metrical model of Chickasaw intonation

This chapter presents a formal model for transcribing the intonational properties of Chickasaw, a Western Muskogean language spoken in south-central Oklahoma by perhaps a few hundred speakers. The proposed model adopts many elements and assumptions of the autosegmental/metrical (AM) framework originally developed to analyse English intonational structure (Pierrehumbert 1980, Silverman et al. 1992, Beckman and Hirschberg 1994, Pitrelli et al. 1994) and subsequently extended to other languages, such as Japanese (Beckman and Pierrehumbert 1986, Pierrehumbert and Beckman 1988, Venditti 1995, this volume, Campbell and Venditti 1995), Korean (Jun 1993, this volume), French (Jun and Fougeron 1995), and other languages described in this volume. Because its prosodic system differs rather substantially from most other languages analysed within the AM paradigm, Chickasaw is an interesting test case for establishing the ability of the AM framework to model a broad range of intonation systems (see also Bishop and Fletcher this volume for analysis of the intonation of Bininj Gun-wok, another language prosodically quite different from others modelled within the AM framework). Although the Chickasaw data require modifications of certain aspects of the AM systems employed by investigators of other languages, this chapter will show that the AM framework is ultimately well suited to the analysis of Chickasaw intonation.