Historical data in Chinese documents.
370 THE ANCIENT SITE BEYOND THE NIYA RIVER [Chap. XI
But with all these indications to guide me, there was reason to feel particularly gratified when the approximate dating thus arrived at received confirmation and precision by incontrovertible chronological evidence. It came, as I had always hoped, from the Chinese records on wood found in our ancient rubbish-heap. Their first cursory examination at Kdshgar, effected with Mr. Macartney's help through Sun Ssû-yeh, showed that their contents were mainly of an official nature, and that their characters bore distinct resemblance to the type of writing which Chinese palaeographers associate with the Han epoch. Their subsequent scrutiny in the British Museum by Sir R. Douglas enabled me to form a more exact idea of the nature of the brief police orders, requisitions, &c., recorded on some of these slips of wood. But it was only when in December, 1902, Dr. S. W. Bushell had occasion to examine the documents that one of them, N. xv. 326 (see Plate CXII) was discovered to be fully and precisely dated. The initial characters, as since verified by M. Chavannes, who subsequently was kind enough to undertake the systematic decipherment and translation of all Chinese records brought to light by me, unmistakably indicate the fifth year of the T`ai-shin period of the Emperor Wu ti, the founder of the Western Chin dynasty, corresponding to 269 A. D. The accuracy of this reading is established against any possible doubt, as M. Chavannes has proved, by the specification of a chronological detail in the rest of the dating : ` the twentieth day Ling--ch`ou of the tenth month, of which the first day is the day wu-wu30.' Thus a precise date has been found which fixes the period when this remarkable collection of documents accumulated. N. xv. 326 itself was discovered in a layer of refuse about one foot above the original floor.
The highly instructive notes with which M. Chavannes has accompanied his full transcripts and translations of the Chinese documents on wood, reproduced in Appendix A, make it possible for me to attempt here a brief delineation of the general character of these records, as well as of the main historical data they furnish 31. Among the latter, to which we may well turn first, the precise date of N. xv. 326 is undoubtedly of exceptional value, in connexion with a number of antiquarian questions. But wider historical importance may be claimed for what the whole collection proves as to the re-assertion of Chinese authority in Eastern Turkestdn under the Emperor Wu ti (265-289 A. D.). The Chin Annals, as noted by Chavannes in his remarks on N. xv. 93, record of Wu ti that he caused his supremacy to be recognized in the ` Western Countries' ; they also record embassies with tribute in 2 70 A. D., from Farghâna and Kara-shahr,
Discovery of precise date, 269 A. D., in Chinese document.
site, has since been identified by Dr. Bushell as bearing the legend CHienyuan ch`ung pao `Copper money of the Chien-yiian period' (758-763 A. n.). This coin would have to be accepted as a proof that the deserted site was visited during the Tang period, if its evidence could quite be relied upon. My notes show that this particular coin was brought to me at the site by Turdi Khwaja, when he joined me from Khotan on February 7, 1901 (see Ruins of Khotan, pp. 406 sq.). He said he had picked it up on his march to my camp, then near N. xv., and as I understood him, since he had left Imam Ja`far Sadiq. Whether this was really the case, and at what particular part of the area he found it, I could not ascertain at the time, and the question must thus remain open. My treasure-seeking guide, always on the lookout for antiques, was honest enough on most matters, but not exactly burdened with an ` archaeological conscience'.
30 See App. A, part ii, note on N. xv. 326.
71 I regret that in the Plates CXII—CXIV, which reproduce the whole of the Chinese records on wood, they are
shown in an order different from that in which Prof. Chavannes has arranged his translations in App. A. While the latter arrangement is based on a systematic classification of the slips according to their contents, the Plates show them as grouped according to the various types of handwriting recognized by Mr. Rionin Kohitsu, a Japanese expert in Chinese palaeography, who early in 1903 kindly offered his help for the reproduction of the documents. I regret that I could not find time previous to my departure from England to arrange for the preparation of fresh negatives when M. Chavannes' translations had become available, and was subsequently obliged by practical considerations to have the collotype plates printed from the available negatives. As the context does not suffice to establish connexion between any two slips, the criterion of handwriting may possibly yet prove of some use. On the other hand, the reproduction of the first grouping has resulted in N. xv. 328 and 75, which are the broken halves of one slip and exactly fit each other, being shown erroneously in different plates.