Lay Students and the Making of Written Vernacular Narrative: An Inventory of Tun-Huang Manuscripts

LAY STUDENTS AND THE KAKDlG OF WRITTEN VERRACOLAR NARRATIVE: AN INVENTORY OF TUN-HUANG MAHUSCRIPTS Victor Mair Problem University of Pennsylvania For more than fifty years, the almost universally accepted explanation of the raison d'~tre of Tun-huang pien-wen ~,t 'J:! ~..t:' ~ has been that they were the promptbooks of monks who used them for evangelistic purposes. This view was first given broad currency by Cheng Chen-to ~r 1~ ~l in his Ilustrated History of Chinese Literature 1~ I~ J} f l~ J.. , '/... (Peiping, 1932) and History of Chinese Popular Literature l' @ 1iJ- ~ ~ f.. (Changsha, 1938). It has been freshly reiterated by Richard E. Strassbelg in an article entitled "Buddhist Storytelling Texts from Tun-huang," Chinoperl Papers, 8 (1978), 39-99, and in the most recent study of tEe subj!3ctmfrom China by Chang Hsi-hou ,f,"% if. called Tun-huang Literature R 'K~ :z. \~ (Shanghai, 1980). In two forthcoming works, Tun-huang Popular Narratives and Tun-huang Transformations, I re-examine the evidence for this and a number of other assertions about pien-wen, such as that pien means "popularization" and that the genre died out dramatically in the early Sung period because it was prohibited by Chen-tsung ~ :1: (r. 997-1022). I have been unable to substantiate any of these claims. To avoid repetition, I shall address myself in this paper to a single much-discussed and much-vexed issue, namely, who wrote down the Tun-huang popular narratives and why? The data which I have assembled in the following inventory point squarely to lay students1 who were enrolled in monastery schools at Tun-huang. So as not to prejudice the issue, I shall avoid, for the moment, further elaboration on this point. The mas included in the inventory fall into several categories. The first is pien-wen f~~ -st ("transformation texts"). ~ All texts that have been identified by responsible scholars as pien-wen have been entered in the inventory. I consider irresponsible any estimate of the number of extant pien-wen mss that exceeds two hundred (published guesses have run higher than 8,000!). My own studies will show that there are no more than thirty verifiable pien-wen mss in the world.3 One of the reasons why the number I give is so low is that I make a sharp distinction between pien-wen and chiang-ching-wen i~ ~t -J:.. ("stitra lecture texts"). Both are prosimetric (an English word [see OED, s.v.], resur~~cted PI the czeCh. school of Sinology, which means the same as chantefable and ,&f\J / ~~ o~ ~ ~ ' i •e. a literary form that al terna tes between spoken or intoned prose s,ections and sung verse sections), both are part of the medieval popular Buddhist tradition, and both are written in a language that is more or less vernacular. But there the resemblance ceases. Pien-wen are the written descend~nts of oral narrative performed by popular entertainers; they may deal with secular or religious subjects. Chian -chin -wen are records of or notes for religious lectures for laymen (su-chiang lJti r7 ) • They are basically exegetical texts closely tied to a single sutra but may contain a certain narrative component. Pien-wen may normally be identified by the pre-verse formula "please look at t~e place where X X hapt:?ens; how does it go?" .ii ~ X X ,t ~ ~ rt. 1ft or some variation thereof. 4 This formula is directly derived from the use of pictures as visual aids in the type of storytelling known as chuan-pien fj, ~li.~ ("turning transformation [scrolls]"). The characteristic quotative formu\a for chiang-ching-wen is "please sing" C7~ ~-! * (addressed to the cantor) at the end of verse sections and preceding the citation of the sTItra which is being explicated.5 In using the word "formula" here, I make no claims about the nature of the orality of these texts. The Parry-Lord theory of oral composition has yet to be properly and rigorously applied to the Tun-huang popular narratives. Due to the nature of the transcription process by which they were recorded and copied, the inadequate size of the corpus, and other technical reasons, I am skeptical...