Abstract
The study of the Ch'ieb Yün system should be conducted entirely independent of the Sung dynasty rhyme tables. As to the upper ch'ieh characters of the Kuang Yün, Luh Chi-wei divided 51 groups in his “51 Groups of initials of the Kuang Yün, a Formal Proof.”(YJCS 25. 1939. 1-58), and applying the same cross reference method to the T'ang dynasty rhyme books, Ling Ta-t'ing proved that there were similar tendencies in these rhyme books and the Kuang Yün (YJCS 26. 1939. 129-142). The 319 rhyme groups, divided by Luh in the same paper, were, as he noticed it, only “the starting point of a study” and not “the result of a study.” The author of this paper classifies the rhymes of Cb'ieb Yün, with reference to the Luh's 51 initial groups and not to the Sung rhyme tables, as follows:
I Group chiefly uses the Division II characters.
II Group chiefly uses the Division II characters and Group B of the Division I characters.
III Group chiefly uses the Divsion I characters.
As to the so-called “fan-ch'ieh doublets” that appear in the C and D initial Groups of the III rhyme Group, some scholars assume the difference in medials (-i-, -ï-) and someone in the principal vowels. Mostly, the lower ch'ieh characters represent the distribution of the class, and A initial Group belonged undoubtedly to the palatal (or closer vowel) class, and B1 Group (series) to the neutral (or opener vowel) class. But the position of the B2 Group (series) within III Group, being not clearly pointed out by fan-ch'ieh, can not be decided. The examination of fanch'ieh show the following pattern:_??_The author, then, interprets the doublets as the difference in the initial consonant phonemes: ki-, kji-; pi-, pji-(instead of ki-, ki-; pi), tji-, tci-, tsi-, li-, etc.
As the former interpretations, this one also does not fit the distribution of the Sung rhyme tables. The Yün Ching arranges as follows:_??_(N. B. A1 includes, sub-groups and A2 includes sub-groups. In the above table, however, sub-group belongs to A2).
The main problem is the position of A1 in III Division (except. sub-group), and according to the author's view, if such arrangement of the Yün Cbing is acknowledged, the sound shift, from the dorsal consonant (in the Cb'ieb Yün) to apical, must be assumed to have occured between the time of the Cb'ieb Yün and that of compilation of the Yün Cbing: tG->t∫-, tG->t∫'-, dz>d3-, G->∫-, z->3-. az->3. This phenomena will explain easily why the “30 (or 36) initial alphabets” compound (tG-) series with (ts-) series and not with (ts-) series.
The Yün Cbing is, therefore, not a mere indication of the Cb'ieb Yün system, but in the interpretation of the latter system it reflects the phonemic system of the time of its compilation.