UC San Diego

This dissertation presents both a descriptive and a formal account of palatalization patterns as identified in a balanced sample of 117 languages. I distinguish between two palatalization types, one involving a primary place of articulation change (full palatalization, e.g. t--> tS), the other involving the acquisition of a secondary palatal articulation (secondary palatalization, e.g. t--> tj). The focus is on similarities/differences in palatalization patterns due to the place of articulation of target consonants, and on palatalization triggers. I develop a formal analysis which uses Articulatory Phonology (AP) and Optimality Theory (OT), making crucial reference to the oral articulators (lips, tongue) that produce the sounds involved in palatalization and their interaction during speech production.

Two main patterns are identified regarding palatalization triggers: (i) if lower front vowels are triggers, so are higher front vowels; (ii) if high back vowels are triggers, so are high front vowels. Regarding palatalization targets, I identify a striking dependency of labial palatalization on the palatalization of coronal and dorsal consonants: while coronal and dorsal palatalization can be independent or co-occurring in a given language, labial palatalization is always dependent on the palatalization of coronals and dorsals. Furthermore, labials do not undergo full palatalization. The few cases where this appears attested are explained via diachronic changes which did not involve palatalization of the labial itself. Historical evidence indicates that a palatal glide following the labial hardened to a palatal consonant, and that the labial ultimately deleted.

The proposed account explains the occurrence of palatalization, as well as the general palatalization patterns and labial palatalization. Paltalization is viewed as the result of temporal overlap of articulatory gestures produced with the two major articulators, tongue and lips. Full palatalization results from great overlap of tongue gestures, and secondary palatalization results from minimal overlap of tongue/tongue or lips/tongue gestures. The formal OT implementation relies on constraints that have an articulatory motivation and also capture the dependency of labial palatalization on the palatalization of coronal and dorsal consonants. The results of the crosslinguistic study and the formal analysis demonstrate that phonetic articulation must be incorporated in the explanation of phonological patterns.