National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Tibetan manuscripts.

Tun-huang under Tibetan dominion.

Return of Tun-huang to Chinese allegiance,

A. D. 850.

8i6   EXPLORATION OF A WALLED-UP HOARD   [Chap. XXII

are indications of such links, as we shall see further on, to be traced also in other. remains of the ` Thousand Buddhas'.

But however this may be, there is definite proof in the abundance of Tibetan texts that the Buddhism of Tun-huang must have been subject to an even more powerful influence from the south during a certain period, and fortunately we can determine this from well-established historical facts. The ` miscellaneous bundles ' had from the first proved to contain hundreds of leaves from Tibetan Pôthis. The packets of leaves were usually mixed up in great confusion ; but the greatly varying sizes, as illustrated by the specimens reproduced in Plates CLXXIV–CLXXIV, would help to restore order afterwards. Apart from these books of loose leaves, provided often with string holes but with no strings to hold them together, and from occasional rolls with Tibetan text written crosswise (see Ch. 06, 07, Plate CLXXIV), I soon ascertained that the solid mass of ` library bundles ' still left in the chapel contained also a considerable proportion of packets with large convolutes of Tibetan sheets usually divided into six columns (Ch. 05, Plate CLXXIV).

Not being a Tibetan scholar, I had no means to make sure whether these convolutes contained different portions from the huge canon of Tibetan Buddhist literature or mainly such endless repetitions of favourite religious texts as Tibetan piety still loves to produce in print—or to use on prayer-wheels. But it was easy to notice that the coarse whitish paper of these sheets differed greatly from that of both rolls and Pôthis. The rolls seemed for the most part to be written on thin grey paper of inferior texture, such as had been used also for Chinese rolls of what soon proved to be a later period, or else on the back of Chinese rolls of that thin but tough yellowish paper of superior make which dated colophons before long taught me to associate with Tang times. That both classes of rolls had been written by Tibetan monks established at Tun-huang was an inference which readily offered itself. The paper of the Pôthis was generally of a stronger make, altogether different in appearance, and recalling the Pothi fragments which I had excavated at the Endere and Miran sites.6 In the case of these Pôthis, import from Tibet naturally suggested itself.

But whatever the original place of production of these Tibetan manuscripts might have been, there could be no doubt about the reason for the great number of them in the walled-up library nor about the chronological indication that it affords. From the lucid analysis of Chinese historical notices which M. Chavannes had prefixed to his interpretation of the previously mentioned inscriptions of Chien-fo-tung,7 it was certain that Tun-huang had experienced prolonged periods of Tibetan predominance from the eighth to the tenth century. About A. D. 759 the territory of Tunhuang had been conquered by the Tibetans, who by A.D. 766 definitely established their power over the whole of Kan-su. The possession of Tun-huang was of special importance for the Tibetans, as it secured the gate for their final conquest of Eastern Turkestan towards the close of the eighth century,8 but the administration of the territory was left in the hands of hereditary local chiefs or governors. It was one of these, Chang I-ch`ao, pre-eminently mentioned in the inscription of A. D. 894, who in A. D. 85o broke with the Tibetan power and made his submission to the Chinese empire.

It is of this important event in the history of Tun-huang that a fortunate chance has preserved for us an authentic and almost contemporary record in the large Chinese inscription which, as already noted, was recovered from the hidden chapel, and of which M. Chavannes' generous help has provided an annotated translation in Appendix A. It has proved to reproduce two imperial edicts of A.D. 851, and their contents have so direct a bearing upon the conditions of Buddhism at Tun-huang during the Tibetan period that this seems the most appropriate place for briefly reviewing

° Cf. above, pp. 462 sq. ; Ancient Khotan, i. pp. 425 sqq.   ° Cf. M. Chavannes' discussion of the Chinese historical

I See Chavannes, Dix inscriptions, pp. 12 sqq.   records, Ancient Kholan, i. pp. 534 sqq.

•