Проблемы китайского и общего языкознания. К 90-летию С. Е. Яхонтова

 450    Christopher I. Beckwith revise our understanding of the Pai-lang Songs, and with it the reconstruction of the Pai-lang language, taking into account the results of recent scholarship on Old Chinese and Middle Chinese, but focusing primarily on some issues of fundamental importance that seem to have been overlooked. One of the most important points made by Coblin, citing an observation of Tung Tso-pin, is that the songs rhyme in the Chinese translation — while by contrast the Pai-lang texts do not rhyme 1 — showing, in their view, that the songs must have been composed in Chinese and translated into Pai-lang 2 . This point is presumably based on the a priori assumption or belief that the Pai-lang poet or poets followed the rhyme systems that have been recon- structed for Middle and Old Chinese on the basis, ultimately, of the Sung Dynasty rhyme tables. However, the texts are explicitly said to have been transcribed by a writer from a far southwestern province of the Later (‘Eas- tern’) Han Dynasty, so they must reflect a far southwestern dialect, not the Central dialect of the capital. Moreover, the Pai-lang phonological system is by definition different from the Chinese phonological system or systems. To represent the Pai-lang language, the Pai-lang and Chinese author or authors would presumably have picked Chinese transcriptional characters accord- ing to their pronunciation by Pai-lang speakers, not their pronunciation by Chinese speakers. In addition to these methodological questions, there are two specific objec- tions to be raised to Tung and Coblin’s view, based on the texts themselves. Firstly, the fact that the songs rhyme in Chinese does not mean that they must have been composed in Chinese, because Classical Chinese is perhaps the easiest language in the world in which to rhyme 3 . And although the songs segmental transcriptions in Central Eurasian alphabetic scripts, which have been fully examined (and compared to the usual reconstruction based on the Ch’ieh-y n 切韻 ) by [Takata 1988]. 1 Coblin does not not clearly state this in so many words, but his article does not mention any Pai-lang rhymes, nor does it even mention the possibility that the Pai-lang songs might rhyme [Coblin 1974: 196]. 2 With Coblin’s strong interest in Late Old Chinese, it is understandable that he devotes considerable space in his article to a discussion of the rhymes of the Chinese translations according to his reconstruction of Eastern Han (i. e., Later Han) period Chinese. The reconstruction of the latter given in [Coblin 1974] is revised by him in his subsequent work [Coblin 1983], q. v. 3 Modern Mandarin has only some four hundred possible syllables, while at the same time there are tens of thousands of characters available to write that limited number of syllables. Although Middle Chinese and Old Chinese certainly had more rhymes than Mandarin, the possibilities were nevertheless far more limited than with a language such as English. The earliest layer of Old Chinese reconstructed by [Starostin 1989] has only fifty-odd theoretical rhymes; the number of actual rhymes is even smaller. Moreover,