Han-Zangyu tongyuanci yanjiu, vol. 3: Han-Zangyu yanjiu de fangfalun tansuo [Cognate words in Sino-Tibetan languages, vol. 3: Methodological issues in Sino-Tibetan research] (review)

Methodological issues in Sino-Tibetan research, the third volume in the series ‘Cognate words in Sino-Tibetan languages’, discusses recently advanced methods for related word analysis (Chs. 2 and 3), presents the database of cognate words in Sino-Tibetan (ST) (Chs. 4 and 5), and offers the results of statistical analysis performed on its basis (Ch. 1).

Chen Baoya and He Fang’s ‘Relativistic rank analysis of kernel consistent corresponding words between Chinese and Kam-Tai’ gives an account of the namesake method, based on a study of Swadesh word lists. The authors distinguish between the 100-word list of more basic and the 100-word list of less basic words and argue that genetically related languages have a greater number of related words in the former than in the latter. Conversely, the number of related words in the former list is less than that in the latter if the relationship is due to language contact. This is apparently an independent duplication of Sergej E. Jaxontov’s analysis of the Swadesh 100-word list as consisting of thirty-five more basic and sixty-five less basic words, and of his similar idea of the relationship between the ratio of related and unrelated words within the more and less basic words respectively and the genetic relatedness of the compared languages. Using this method, the authors argue that similar words between Kam-Tai and Chinese are loans. The chapter also addresses the points of criticism that have arisen since the proposal of the method.

Xing Kai’s ‘Semantic comparative method’ summarizes the method of deep semantic correspondence advanced by Xing Gongwan in the 1990s. This method is based on the assumption that sets of internally related words (i.e. word families) in one language cannot be borrowed in their entirety into another language. If such sets are shared by two languages, while preserving parallel differences in sound and meaning, the languages are argued to be genetically related. The article summarizes sixteen types of deep correspondences between Chinese and Tai and demonstrates the cognate relationship between these groups. The list of 305 sets of Chinese-Tai deep semantic correspondences totaling 595 cognates is appended to the article.

Jiang Di’s ‘A development report on the Sino-Tibetan cognate database retrieval software’ and ‘The Sino-Tibetan cognate database project: A manual for data retrieval’ present the database of ST cognate words, which presently includes data for 130 ST languages, each represented by 1,332 lexical items. Besides a general introduction to the database, the chapters contain information on technical aspects, such as different methods of retrieval, input, help, and system maintenance.

Finally, Huang Xing’s ‘Statistical analysis of relations between ST languages’ illustrates the use of the database in linguistic research, testing with its help the reliability of Chen’s and Xing’s methods.