Difficulties and Directions (On the Continuous Tonal Changes of Tianjin Dialect)

Author: Huang Liangxi
Publisher:
Publish Date: 2005-12-01
Features: Tone sandhi is a major feature of Chinese dialects, and it poses two questions for linguistic theory: (1) What is the relationship between tone sandhi and the original tone? (2) Can tone sandhi in multi-syllable combinations be derived from the basic tone sandhi of two-syllable combinations? If so, which combination occurs first? From both aspects, Tianjin dialect is highly worth in-depth research. Briefly, the tone sandhi in Tianjin dialect depends on the interaction of two constraints: no adjacent identical tone shapes and no adjacent identical tone heights. In this book, L represents the low level tone shape of yinping, H represents the high level tone shape of yangping, LH represents the rising tone shape of shangsheng (or simply R), and HL represents the falling tone shape of qushi (or simply F). When two yinping tones LL are adjacent, due to the shape constraint (no identical shapes), the preceding yinping tone must change; at the same time, due to the second constraint (no identical heights), Tianjin dialect inserts an H after the preceding yinping tone, turning it into LH, i.e., shangsheng. Thus, we can derive a simple rule: LL—LHL (i.e., RL), which reflects the correspondence between tone sandhi and the original tone. When the number of adjacent syllables increases to more than two, such as LLL, RRR, RFL, LLLL, RFFFL, etc., we face the issue of multi-syllable tone sandhi. It is not difficult to imagine how complex the tone sandhi situation would be. Therefore, the study of tone sandhi in Tianjin dialect is fascinating—it involves both the concept of rule ordering in traditional generative phonology and the interactive constraints in the recently popular optimal theory. Additionally, Chen Yuanquan, Yan Xiuhong, and Huang Liangxi (2004) encountered even more challenging difficulties when dealing with tone sandhi in Hakka dialect. Interested readers can refer to their work.
To ensure that the corpus belongs to a well-structured, rigorous, and consistent linguistic system and to avoid data interference caused by multiple speakers' phonetic systems, we chose to focus on a single speaker. Since tone sandhi in Tianjin dialect fully follows the modern four tones and is unaffected by the Middle Chinese tone origins, we collected the corpus based on combinatorial mathematical principles, eliminating the issue of omissions. The investigation ranges from two-syllable combinations to five-syllable combinations, with preliminary research on six-syllable combinations, which showed no significant differences from four- and five-syllable combinations. Our analysis of the corpus covers various phenomena of tone sandhi, the underlying patterns, principles, and constraints. Despite this, we believe that this book is still far from a comprehensive understanding of tone sandhi in Tianjin dialect. To facilitate further in-depth research by interested experts, we systematically report the corpus in the appendix, ordered by tone category and syllable number. The investigation tables are also provided for reference when studying tone sandhi in the North China Mandarin dialect area.