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. ii]   THE TUN-HUANG LIMES SINCE ITS CONSTRUCTION   733

in a temple outside Barkul town, which M. Chavannes has edited and translated, records a great yictory gained in A.D. 137 by a prefect of Tun-huang over the Hsiung-nu king Hu-yen, and claims that this secured order and calm for the frontier territory.1°

It is this very year which has furnished us with the last accurately-dated document from the Tun-huang Limes, T. xv. a. i. 6 (Doc., No. 536). The date A. D. 153 proposed for another record, Doc. No. 68o, depends on an inferential calculation of a calendar, and is in any case not far removed from the latest certain date. "Thus the records recovered from the ruined watch-stations bring us down to the period from which onwards historical notices of the Tun-huang Limes seem to cease in our accessible sources. Whatever the direct cause may have been, it seems safe to assume that the regular guarding of the wall and its towers did not continue beyond the middle of the second century A. D. The progressive disintegration of the empire under the last two Han emperors (A.D. 168-22o), and still more during the Epoch of the Three Kingdoms ' (A.D. 221-77), might furnish an adequate explanation for this. Or else some connexion may be sought with the receding danger from the Hsiung-nu, whose great westward movement was soon about to begin.

But if the line of the Limes itself fell into neglect, it was different with the route to Lou-lan which led along it. The discussion of this route above has shown us that its continued contemporary use is proved by the account of the Wei lio composed between A. D. 239-65.20 We have also seen that, in the light of the'conclusive evidence furnished by the dated documents found at the Lou-lan Site and ranging from A.D. 263 to 330, Lou-lan must be assumed to have retained a small Chinese garrison and a direct line of communication with Tun-huang and China right down to the first third of the fourth century A. D.21 I n one of these documents, the fragment of a letter dated A. D. 312 (Doc. No. 912), a departure from the ` Jade Gate barrier ' is still directly mentioned. But whether this famous station was then still situated at T. )(Tv or had been moved nearer to Tun-huang, it is impossible to say. Fa-hsien on his journey in A.D. 400 to Shan-shan, i.e. the Charkhlik tract, must have passed by the caravan-track leading along the line of the abandoned Limes.22 But though he correctly describes, as we have seen, the character of the desert crossed en route, he makes no mention of the Jade Gate.

I am unable to trace any later reference either to the Tun-huang Limes wall or to the Jade Gate until we come to Hsüan-tsang's •start for the Western regions, in or soon after A. D. 630. I shall have occasion below to discuss fully the interesting account which the great pilgrim's Life furnishes of his secret departure past the Yü-mên barrier.23 Here it will suffice to point out that this account clearly places what was then locally known as the Jade Gate to the north of the old town of Kua-chou and on the left bank of the Su-lo Ho, not far from the present walled town of An-hsi. The five watch-towers pushed far out into the desert north-westwards which also figure in that narrative have to be looked for on the line of the still used road from An-hsi to .1-15.mi.23a But the watch kept there must in all essential points have closely resembled that which centuries earlier was maintained at the outlying guard-posts beyond the ancient Yii-mên.

That nevertheless the old position of the Jade Gate was still remembered in Tang times, at least in learned tradition, is proved by the interesting passage of the Chin Tang shu, already referred to, which correctly places Yii-mên 118 li to the north-west of Shou-ch`ang, the present Nan-hu, and thus exactly at the site of T. xiv.24 That at Tun-huang itself popular notions on the subject were

Last dated record from Tun-huang Limes,

A. D. 137.

Continued use of Loulan route along Limes.

Hsüantsang's Yfl-min located near An-hsi.

Old Yii-min remembered in Tang times.

'9 See Chavannes, Dix Inscriptions, pp. 17 sqq.

=0 See above, pp. 555 sqq,

21 Cf. above, pp. 407 sqq.; Chavannes, Documents, p. iv.

See above, p. 558.

ra Cf. below, chap. xxvi. sec. ii ; Julien, Vie de Hiouen

Thsang, pp. 17 sqq. ; Beal, Life of Hiuen-lsiang, pp. 13 sqq. 23a Cf. below, chap. xxvin. sec. i.

24 See above, p. 621. The passage [Chia Tang shu, chap. 4o, fol. 47 vO] is known to me only from Dr. Giles's paper on the Tun huang 1u, J.R.A.S., 19x4, p. 713.