The Unary Feature Hypothesis Of Segmental Structure

This thesis aims at arguing for the feasibility and validity of the unary feature hypothesis by comparing the rhymes of different Chinese dialects as well as the rhymes between Middle Chinese and modern Chinese dialects.The pattern to be attested against the facts drawn from Chinese dialects is shown below,which expresses the structural relation existing between rhymal components as well as between rhymes of different types (α,βhere refer to the unary features). The correspondence between the rhymes will be described and analyzed with this pattern.The dialects in the research mainly concern Xiang Dialect Group and Min dialect Group, which include Changsha Dialect, Ningxiang Dialect, Xiamen Dialect and Jian'ou Dialect. Based on preliminary observation, it can be found that there are correspondences between the rhymes in these dialects. From the perspective of unary feature hypothesis, these kinds of correspondences in terms of segmental structure are expected in Chinese dialects. The correspondences are listed as follows:(1) Correspondence between diphthongs and monophthongs. For a cognate morpheme, Changsha Dialect shows [ai] and [au] while Ningxiang Dialect shows [(?)] and [(?)] respectively. Similarly, Xiamen Dialect shows [ui] while Jian'ou Dialect shows [y]. Binary feature theory cannot directly describe this kind of correspondence while unary feature hypothesis does. Actually, the diphthong can be considered as the linear ordering of the features in the monophthong.(2) Correspondence between rhymes with obstruent coda and those with zero coda. The evolution of obstruent coda in Chinese is exemplified as: p, t, k→(?)→φ. For example, the rhyme [ap] in Middle Chinese remains in Xiamen Dialect but changes to [o] in Changsha...