Prosody or pharyngealization in old Chinese? The origin of the distinction between Type A and Type B syllables.

• Author: Pulleyblank, Edwin G.
• Position: Response to Jerry Norman, Journal of the American Oriental Society, vol. 114, p. 397, 1994

It is a pleasure to welcome Jerry Norman(1) to the small group of scholars who are now prepared to reject the ubiquitous yod of Karlgren's reconstruction of Old Chinese, which was taken for granted by Li Fang-kuei and is still adhered to by William Baxter and others. As Norman fully recognizes, I have been advocating this for a long time. He correctly states that my first idea was that long vowels in Old Chinese had diphthongized in Middle Chinese: *a:n[right arrow] *ien, etc. What he says about my more recent views about Type A and B syllables (corresponding to those with unyodized and yodized initials in Karlgren's system), which he claims that I have never "substantiated by either arguments from linguistic typology or by the marshalling of material evidence," is, however, quite misleading. On this question he cites only the 1973 paper in which I first adopted this terminology and made the suggestion that the contrast between the two types originated as a prosodic distinction in Old Chinese. What he neglects to mention is that even at that time I had already also rejected Karlgren's yod for the reconstruction of either Early Middle Chinese (EMC) of the Qieyun or late Middle Chinese (LMC) of the rhyme tables, with abundant arguments and much marshalling of material evidence (1970, 1970-71, 1984). This conclusion, which he ignores, undermines his basic assumption, following Karlgren, that palatalization was the defining characteristic of Division III in the rhyme tables and of the corresponding rhymes in the Qieyun.

The crux of the matter is the so-called chongniu, cases where, as he puts it, there are "double entries" in some Qieyun rhymes for some labial and guttural initials divided between Grade III and Grade IV in the rhyme tables. As I have shown in another paper (1995), Karlgren's assumption about palatization as the characteristic of Grade III was something he inherited from his predecessors, Kuhnert (1890) and Schaank (1897-1902). He combined it with his own discovery that the fanqie spellers in the Qieyun for (mainly) velar initials fall into two sets corresponding (roughly) to Grades I, II and IV, versus Grade III. It became an idee fixe that he stubbornly adhered to in spite of the plain evidence of the chongniu that obviously contradicts it, since it shows that the supposedly palatalized initial spellers occur in both Grade III and Grade IV. Karlgren did his best to minimize the importance of the chongniu distinction...