English Vowel Hiatus and Consonant Epenthesis

Abstract

Vowel hiatus arises when two vowels are locally adjacent but heterosyllabified within words or across word boundaries. In English, as well-described, vowel clash is resolved by two strategies; glide insertion and glottal stop insertion. In fact, these sounds are not underlyingly present but added for ease of articulation in casual or fast speech. Following sonority-driven prominence scale in V1-V2 sequences, the least marked glides are the most favored to fix vowel hiatus. Which glide is adopted is closely related to the feature of the first vowel; for a palatal /j/-glide, V1 is a high front vowel, for a labio-velar /w/-glide, a high back vowel and for a central liquid /r/-glide, also called the intrusive r, a non-high vowel, i.e. homorganic. Furthermore, provided that V2 gets stressed, a laryngeal plosive also fills an empty onset even though it is the most marked at the point of vowel hiatus. These hiatus resolution strategies are well-couched into Optimality Theory(OT) (Prince and Smolensky, 1993/2004; McCarthy and Prince, 1995) where Dep[F] type constraints ranked over the sonority-driven markedness constraints determine the glide j/w-epenthesis the best and the intrusive r the second-best if the former is banned. Glottal stop addition is also employed as a rescue strategy to remove vowel clash when glides are all blocked, i.e. before V2 bearing stress.