Optimality Theory and African Language Phonology

ACAL, held at the University of Illinois in 1989. At that time, the dominant research paradigm was autosegmental phonology, a theory which is concerned with issues in the representation of distinctive features (Goldsmith 1976; Leben 2006). Work on African language phonology was so central to the development of autosegmental phonology, that Goldsmith (1990) in his presentation felt no need to defend or explain the mutual influence of theory and ‘descriptive’ work in African language phonology in the 13-14 years since his dissertation (Goldsmith 1976) had crystallized the basic autosegmental formalism and research program. Instead, it evaluated historical precedents to the autosegmental approach. I think no one listening to that talk would have guessed that some five years later, research on autosegmental formalism would have come virtually to a halt. We know the reason for this. The rise of Optimality Theory (OT) shifted attention from theories of representations to a constraints-based theory of phonological processes and their interactions. It is now roughly 13-14 years since the influential founding work in Optimality Theory (Prince & Smolensky 1993, published 2004; McCarthy & Prince 1993a,b) made OT a leading paradigm for phonological research. This is a good moment to evaluate how this new theory has given us a new perspective on some classic problems in African language phonology: tone, vowel harmony, and reduplication. In this paper, I first briefly introduce some central principles of OT that will be important to the analyses which follow. Then, for the sake of concreteness, I present one case study each in tone, vowel harmony and reduplication, to illustrate how work in OT has shifted our perspective on these topics. I conclude with some brief remarks on challenges that work on African language phonology poses for OT. The paper comes with the following caveat: it is not intended to be a survey of interesting work in OT on African languages. My apologies to authors of all the good work I do not have space to refer to here. And naturally, the choice of languages for the case studies is limited by my expertise.