62 FROM THE OXUS TO KHOTAN [Chap. III
spread everywhere. The land produces beans and corn ; it favours trees and fruits ; it produces excellent horses. The people's eyes show a greenish iris. . . . This territory is on the route which leads from the `Four Garrisons ' (or Chinese Turkestan) to the Tu-huo-lo (or Tokharistan). At one time it was dependent on the Tibetans.'
It requires no detailed demonstration to recognize how closely the geographical data here furnished agree with Wakhan. The mention made of the great route leading through it and the reference to the Tibetan influence at one time exercised possess distinct historical interest. The capital Sai-chia-shen undoubtedly corresponds to the present Ishkashim, a large group of villages on the western extremity of Wakhan.8 The historical data which the Tang Annals' notice furnishes, and of which a brief summary will suffice here, help to bring out on the one hand the hold exercised by the Chinese administration over Wakhan during the seventh and eighth centuries, and on the other the strong Turkish influence prevailing in the ruling family, probably through the close connexion with Badakhshan. When the territory in A. D. 656-6o was turned into a Chinese administrative district under the designation of Niao-fei with the king as prefect, his name is given with the Turkish title Czieh-li fa. A string of Turkish names and titles is borne also by the king who is mentioned in A.D. 72o as receiving his brevet of investiture from the Emperor. Offerings of homage are recorded in the years A.D. 728 and 729, and in 741 the king Hu-chên-ean came in person to the Imperial Court.
For the year A. D. 742 the encyclopaedia Ts`ê fu yuan kuei has preserved the text of a brevet issued by the Imperial chancellery to an envoy from Hu-mi or Wakhan, who had been sent by the son of the ruling chief to express his desire of breaking with the Tibetans.9 From this it is clear that Tibetan aggression must have made itself felt on the uppermost Oxus years before Kao Hsien-chih started on his memorable expedition of 747 to close the Tibetan line of advance across the Darkôt and Barôghil Passes. Probably in consequence of this great success Hu-chên-t`an presented himself again at court in A. D. 749 and obtained the honour of a command in the Imperial guards. Even as late as 758 the visit of a Wakhan `king' to the Imperial capital is recorded. That during this whole period Wakhan was directly dependent on Tokharistan, just as in modern times it always shared the political fortunes of Badakhshan, is made evident by a petition which the brother of the Jabgu of Tokharistan in A. D. 718 addressed to the throne and of which the text is preserved in the Ts`ê fie yuan kuei." In this Hu-mi is distinctly claimed as one of the chiefships which for generations past have acknowledged the suzerainty of Tokharistan.
The same close connexion with Tokharistan is reflected in the detailed account which Hsüan-tsang has left concerning Wakhan." The identity of Wakhan with the territory of
Ta-mo-hsi-t`ie-ti )►, through which the pilgrim passed on his way from Badakhshan
to the Pamirs and Sarikol about A. D. 642, was recognized from the first by General Cunningham and accepted by all those who, like V. de Saint-Martin and Yule, followed him in the elucidation of this part of Hsüan-tsang's itinerary. Though a satisfactory explanation of the name Ta-moshih-t`ie-ti still remains to be sought,72 its application to Wakhan is established beyond doubt by
The identity of the names was first recognized by Marquart, Erân.fahr, p. 224. Most editions of the Tang shu
give the erroneous form Han-chia-shen pp. ; see
Chavannes, Turcs occid., p. 165.
9 Cf. Chavannes, Turcs occid., p. 212.
10 See ibid., p. 200.
" Cf. Julien, Mémoires, i. pp. 201 sqq. ; Watters, Yuan Chwang, ii. pp. 279 sqq.
12 For attempts to connect this name with Mastüj, see V. de Saint-Martin in Julien, Mémoires, ii. p. 425; Yule in J.R.A.S., N. S. vi. p. i i 2 ; Marquart, Erân,fahr, p. 225. But there is no evidence that Wakhan or the uppermost part of it ever bore the name of *Darah-i-Mastiij or *Dar-i- Mastij, as has been assumed. The use of this term would be particularly strange in the case of Hsüan-tsang, who did not even visit that part of the valley from which the route