Abstract
The purpose of this dissertation is to reveal the process and mechanism of the emergence of General Taiwanese under the mutual competition of Minnan dialects Chang-chou and Ch‘üan-chou. This is verified in three steps: 1) measuring the age differences of some critical items by geographic and social dialectal investigations. 2) comparing two sets of data from Chung (1967) and our previous investigation of Taiwanese variant identification of university students (1999). 3) giving analyses and explanations for these phenomena. The real-time comparisons with the data available show that the synchronic variations are not merely stable age-grading, but are ‘changes in progress.’ It is revealed that all the directions of changes in all variables are consistent, whether Chang-chou or Chiuan-chou accent. All the prevailing variants are consistent to the result of the investigation of Taiwanese phonetic identification of university students. There are some changes in progress in Taiwanese, such as the consonant j→l, the collapse of central vowels into front vowels, tone leveling and mid-toning. All changes are given a proper universal explanation. We try to verify that the New Taiwanese is not merely a blending or mixture of the two dialects, but a fusion or syncretism. We will show that, as a result of this fusion, New Taiwanese has become the center of convergence, and all variants not conforming to this trend are in the process of decay. This dissertation verifies the emergence of the General Taiwanese, and clarifies the mechanism of the fusion of the two dialects. The results of this research support the ‘drift’ theory of Sapir (1921) and the theories of ‘variable as a structural unit’ suggested by structural dialectologists. We also propose the hypotheses of ‘categorial diffusion’ and ‘structural diffusion’.