国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

Sec. viii]   CHINESE DOCUMENTS FROM THE LOU-LAN SITE   407

which led from Tun-huang, on the extreme west of Kan-su, to the main line of oases north of the Tarim River. From passages of the Chinese Annals which we have already had occasion to discuss,' it was certain in a general way that the first expansion of Chinese influence into Central Asia had proceeded by a route which was opened about 110 A.C. through the desert west of Tun-huang to Lou-lan or the Lop region, and that this route had remained in use through the whole of Han times. But as renewed reference to those passages will presently show us, the indications available in the texts would not by themselves suffice to determine the exact direction of the route. Strong as the archaeological evidence was which pointed to the ` Lou-lan Site ' as the western terminus of the desert route, confirmation by documentary evidence was particularly needed to meet the serious doubts which the absolutely barren nature of the desert further east was bound to raise about the correctness of such a location of the route.

From all that the Chinese Annals have to tell us, and from broad geographical facts which remain unchanged to the present day, it is clear that, throughout the successive periods of China's control of the Tarim Basin, it was always the great route leading along the southern foot of the Tien-shan and through the string of big oases from Korla westwards to Kashgar which claimed most importance for Chinese traders, administrators, and soldiers. It was by this route, the Pei-lu or ` Northern Road ' of the Annals, that the bulk of the silk trade, for the sake of which the first advance of Chinese political and military power into Central Asia was made, moved to Farghana or Ta yuan and into ancient Sogdiana and Bactria.3 The protection of this great trade route against the inroads of the Huns and their nomadic successors north of the Tien-shan was the main purpose for which the Tarim Basin was held, and it always remained the chief aim of the Chinese administration set up in the ` Western Countries'.

A reference to the map suffices to show that the shortest way to reach that line of oases from Tun-huang, the westernmost cultivated area on the Kan-su marches of China, lay along the foot of the Kuruk-tagh and through the Lop desert, past the ` Lou-lan Site ', to where the Tarim bends south-eastwards. But nowadays a marching distance of over two hundred and forty miles of wholly waterless desert, even on the most direct line, intervenes between the last-named point and the nearest well on the Tun-huang-Charkhlik caravan track, and this would render the use of the ancient route wholly impracticable for caravan traffic at the present time.

In Han times and in the century immediately succeeding, it is true, the existence of the

Kuruk-darya delta, just as it accounts for the occupation of the ` Lou-lan Site ', also removed all difficulties about water and grazing for the western half of that distance ; for there the ancient route undoubtedly led along the bed of the ` Dry River', then still carrying water, to Yingp`an, where in 1915 I traced ruins of the same early period, within easy reach of the present Konche-darya. But for the eastern half of the old route, the hundred and twenty odd miles separating the ruins of the ` Lou-lan Site' from the Kum-kuduk wells on the Tun-huang caravan track, the total absence of water must have been as serious an obstacle in ancient times as it now is. It was only by my explorations of 1914 that definite archaeological proof was obtained for the ancient Chinese route having actually crossed this most formidable of deserts, a wholly lifeless waste of salt and clay, and meanwhile we should have found it hard to believe in the possibility of such a route having ever been followed by those early pioneers of Chinese trade and influence westwards, had the fact not been so clearly proved by the documents recovered from the site.

The physical change which has come over this portion of ancient Lou-lan by the drying-up of the delta to which it once owed water and life is so striking and of such wide geographical interest

s See above, pp. 336 sqq. '   Kingsmill, Intercourse of China, jR.A.S., 1882, pp. 4 sqq.

3 Cf. the extracts from Ssû-ma Ch`ien's history in

Importance of ancient silk trade route.

Shortest line passes Loulan Site.

Difficulties of ancient desert

route.

Documents proving occupation, e. n. 263- 70.