東漢 鄭玄의 注釋에 반영된 章組ㆍ以母ㆍ邪母의 구개음화 여부

Abstract

This study is designed to discuss whether zhang series[章組] words, yi initial[以母] words, and xie initial[邪母] words have been palatalized into affricates or fricatives in Zheng Xuan’s sound glosses of the Eastern Han. South Coblin in A Handbook of Eastern Han Sound Glosses, initially asserted that Zheng Xuan’s sound glosses reflected the palatalization of the initials in question according to their frequent contacts with jing series which were believed to be affricates[*TS-type] in Eastern Han. However, there is convincing evidence that shows these initials were not yet palatalized in his sound glosses as they also had a number of contacts with plosives, such as *T-type or *Kl-type words. The main point of this study lies in how we expound the two discrepant phenomena. On the basis of Zheng-Zhang Shangfang‘s Old Chinese Phonology, this study made a survey on the sound glosses shown in Zheng Xuan’s comments in Liji[≪禮記≫], Yili[≪儀禮≫], Zhouli[≪周禮≫], and Shijing[≪詩經≫], along with some of his scattered annotations in other works. The result shows that the jing series which had contacts with the initials in question had the segment /l/, the place of articulation of which was also shared by zhang series, yi initial, and xie initial. It consequently made it possible for zhang series and other initials to have had contacts with jing series. Therefore it is reasonable to assume that the words of zhang series, yi initial, and xie initial were not yet palatalized in the Zheng Xuan’s phonological system. While the initials in question had frequent contacts with the jing series in the glosses of Zheng Xuan and some other scholars such as Liu Xi and Ying Shao, they hardly had contacts with the jing series in the other party of contemporary scholars’ glosses. There could be two possible reasons why the methods of annotating were different between these two parties. First, the l’s in jing series had been lost in Standard Chinese of the Eastern Han, and while it was reflected in the other party of scholars’ glosses, Zheng Xuan’s glosses might have reflected or preserved rather conservative phonological system in which the l’s in jing series were preserved. Second, the l’s in the jing series had been preserved in Standard Chinese of the Eastern Han. Phonologically speaking, the l’s in the jing series were placed in the marginal position of any single initial as against another segment which took the main position in the same initial, such as -s- in *(ʔ)sl-. In this case, the l’s functioned like post-initials while the t’s, d’s, and l’s in the all initials but in the jing series were placed in the main slot of the initials. Zheng Xuan might have tolerated the alternations between the main and the marginal elements while the other party seemed to have a tendency to only tolerate the alternations among or between the main ones. To assure which analysis is more convincing, further research is required.