The Naturalness of Palatalization

Abstract

Section headingPalatalization is commonly attested in the languages of the world, and has received considerable attention in the literature over the years.However, it is by no means clear which processes count as palatalization phonologically or if the term refers to one or several phenomena. Palatalization processes exhibit great variation within and across languages, regarding the triggers of the process, the targets, and the output of the process. Most studies focus on a certain palatalization process in a specific language. Nevertheless, typological surveys have revealed some intriguing tendencies. Among these are apparently universal implicational relations that have been shown to hold among palatalization triggers and targets. For instance, survey results indicate that if labials undergo palatalization, so do coronals and dorsals (Bhat 1978, Chen 1973, Bateman 2007. If lower front vocoids trigger palatalization, so do higher front vocoids, and if non-front high vocoids act as triggers, so do front high vocoids (Bateman 2007, Kochetov 2011. In addition, coronals tend to be palatalized by high vocoids, while dorsals tend to be palatalized by front vocoids (Bhat 1978, Kochetov 2011.Outputs of palatalization show a range of variation. The targeted segment either acquires a secondary palatal place of articulation or shifts the primary place of articulation closer to the palatal region. Interestingly, changes in the primary place of articulation are often accompanied by spirantization. In addition, the change of manner without the change of place is also attested (Bhat 1978, Bateman 2007.Theoretical approaches to palatalization face a considerable challenge of characterizing languagespecific patterns in addition to capturing cross-linguistic tendencies and variability. Palatalization processes have been crucial in developing and testing representational and computational phonological models (Clements