A Dispersion-Theoretic Account of Taiwanese CV phonotactics

Abstract

In Taiwanese, oral voice consonants and nasal stops are in complementary distribution in the onset position: oral voiced consonants only precede phonemically oral vowels, and nasal stops only precede nasal vowels. Similar restriction in the distribution of these segments is not found in other languages with phonemic nasal vowels, such as French and Portuguese. Following studies showing that Taiwanese nasal vowels are fully nasalized, while French and Portuguese ones have delay in nasality (Chang et al., 2011; Delvaux et al., 2008; Parkinson, 1983), this study proposes a Dispersion-Theoretic account to connect phonetic observation about vocalic nasality to the phonotactic restrictions on voicing and nasality. Using a three-stage analysis (Flemming 2006, 2008), where Phonetic Realization is a distinct component of the grammar, the analysis is show how a cross-linguistic different in phonetic implementation of nasality is able to derive differences in surface phonotactics. The analysis also makes explicitly testable predictions about the perception and the typology of the distributions of voiced and nasal segments.