Optimality Theory and language death

This paper exposes a gap in the literature regarding studies that use an optimality-theoretic framework to examine phonological variation in endangered language varieties. It shows that dying languages, characterized by inter- and intra-speaker variation and multiple acceptable surface forms, provide crucial testing ground for extensions of classic Optimality Theory (Prince and Smolensky, 1993; McCarthy and Prince, 1994) such as crucially non-ranked constraints (Anttila and Cho, 1998; Anttila, 2002), floating constraints (Reynolds, 1994; Nagy and Reynolds, 1997), Linear Optimality Theory (Keller, 2000, 2006; Sorace and Keller, 2005), Stochastic Optimality Theory (Boersma, 1998; Boersma and Hayes, 2001), and the learning algorithms associated with them. However, little work has been done mapping raw frequencies from dying languages into optimality-theoretic grammars. Optimality-theoretic analyses of schwa epenthesis in Vimeu Picard (Auger and Steele, 1999; Auger, 2001) and in Louisiana Regional French (LRF) (Blainey, 2009) suggest that speakers of dying languages may possess individual grammars that are related, but not identical, to a more general grammar. At the same time, the preliminary research in LRF offers Optimality Theory highly variable and therefore difficult phenomena to describe and predict. More studies combining language death, phonology, and Optimality Theory are critical to a better understanding of the processes involved in attrition, language shift, and obsolescence. The varieties of French spoken around the world, especially those in a linguistic minority situation such as LRF, constitute ideal starting points for expanding the literature of attritional phonological changes and variation in dying languages seen through the lens of Optimality Theory