Exemplar-based models of phonology predict that frequent words should lead sound change. This study tests that prediction with data on the LIKE homonym set, which undergoes /ay/-raising in Philadelphia. The different LIKE homonyms (verb, discourse marker, conjunction, etc.), despite large frequency differences, are shown to undergo /ay/-raising not only simultaneously but with consistently near-identical values. The exception is the lexical verb, which starts out with a lower nucleus than the other homonyms and then catches up. This difference is argued to be the result of function word reduction in speech production rather than a reflection of phonetic differences in the lexicon. The inability of pure exemplar theory to constrain frequent effects from arising is a weakness; hybrid models with an intervening level of phonological representation are promising in this respect.
Proceedings of the 31st West Coast Conference on Formal Linguistics
edited by Robert E. Santana-LaBarge
Table of contents