| Abstract: | I have two goals in this paper. One is to deconstruct the account of nasal place neutralization processes in Spanish given by Harris (1984ab), and to demonstrate that the typological predictions that it appears to make are falsified by neutralization patterns in languages other than Spanish. The spirit (if not the letter) of Harris's account is at the heart of a great deal of work on the autosegmental analysis of neutralization, and I follow Lombardi (1999) in the abandonment of this approach in favor of a typologically more accurate one involving Optimality Theoretic constraint interaction. My second goal is to provide a novel account of the distinct results of final nasal neutralization in different varieties of Spanish, one that satisfactorily explains the observed variation via the interaction of independently motivated constraints. |
