In The Sound Pattern of English (SPE), Chomsky and Halle (1968) propose that the form of dissimilation rules is, in general, as in (1). (1) [aP] -> [-aP] ,_ [aP] (p. 178) Since dissimilation rules change the feature value [a F] to [-a F] in the context of [a F], it has been proposed that they are a result of the Obligatory Contour Principle (McCarthy 1985, Odden 1987, Yip 1988). McCarthy (1986) proposes that the OCP, in addition to applying to tonal sequences, also applies to melodic sequences as stated in (2). (2) Obligatory Contour Principle: At the melodic level, adjacent identical elements are prohibited (p. 208) In this paper I examine two rules of dissimilation in Takelma: Coronal Dissimilation and Nasal Dissimilation. These rules, I claim, are the result of the Obligatory Contour Principle (henceforth OCP). With respect to these two dissimilation rules the question arises as to whether the OCP functions only in cases of dissimilation or whether it plays a more general role in the grammar of Takelma. In addressing the question of the role of the OCP in Takeima, I focus on the non-symmetrical behavior of coronal sonorants and coronal obstruents. The coronal obstruents neither trigger nor undergo the dissimilation rule. One might suggest that this asymmetrical behavior can be characterized by the underspecification of the coronal feature in the case of obstruents. However, I will show that underspecification cannot account for the fact that Coronal Dissimilation targets only coronal sonorants. Rather, I argue that the OCP in Takelma is conditioned by a segment's value for sonorancy. I show that the dissimilation of coronal sonorants is part of a more general prohibition on adjacent coronal specifications with like values for the feature [sonorant] at both the underlying level and during phonological derivations.

This paper is copyrighted, and is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) - see https://creativecommons.org/licenses/by-nc-nd/4.0/ This paper was presented at the 65th annual meeting of the Linguistic Society of America. I would like to thank the members of that audience for useful comments. I would also like to thank Abby Cohn and John McCarthy for beneficial discussion.