Korean Umlaut Revisited

Abstract

This paper examines what conditions Korean Umlaut. Since Umlaut involves the place assimilation of a back vowel to a following high front vocoid across a consonant or cluster. we might raise a question of whether the process is conditioned by the place of articulation of an intervening consonant or cluster. as usually assumed in the literature. Or we might ask ourselves whether there is something else other than phonological accounts for the occurrence of Korean Umlaut. In this paper. a survey of Korean Umlaut was made for the questions raised in the above. The results of the survey show that the place of articulation of an intervening consonant or cluster is not consistent with conditioning Umlaut. and that it is by non- phonological variables -- lexical. dialectal and individual differences -- that Korean Umlaut may or may not be triggered. Based on the results of the survey, it is proposed that Korean Umlaut is phonologically represented as the spread of the secondary vocalic feature [coronal] to a preceding vowel across an intervening consonant or cluster, regardless of its place of articulation. and that it is a sound change called lexical diffusion.