Sec. iv] DARFL OLD AND NEW 21
ladders to be climbed above precipices and its ` bridge of ropes ', agrees closely with what modern accounts we possess of the very difficult tracks leading down the deep rocky defiles, never yet visited by any European, through which the Indus has cut its way from below Tangir. That Fa-hsien's description refers to this, the most direct route connecting Darél with the central portion of Swat and its old capital, marked by the present Manglaor, is placed beyond all doubt by its exact agreement with what Hsüan-tsang tells us of the journey that he took in the reverse direction from the latter place up the Indus to the valley of Ta-li-lo and its great shrine of Maitreya. To this we shall recur presently. Of Fa-hsien's notice of Teo-1i there still remains to be mentioned the interesting statement that, according to the tradition of the local people ` handed down by their fathers from of old ', the propagation of Buddhist doctrine eastwards began from the setting up of that sacred image of Maitreya ` rather more than three hundred years after the Nirvana of Buddha'.
The correct identification of Fa-hsien's T`o-li lit* with Darél was first made by General Cunningham,4 and it soon found complete confirmation when Hsiian-tsang's account of the same territory became accessible.5 ` North-east of Mêng-chieh-li', so the Hsi yii-chi tells us,6 ` over hills and across gulleys ascending the Indus by hazardous paths through gloomy gorges, crossing bridges of ropes or iron chains, across bridges spanning precipices or climbing by means of pegs for steps, a journey of above i,000 ti brings you to the Tali-10 valley, the old seat of the government of Udyana. The district yields much gold and saffron. In the valley is a great monastery by the side of which is a carved wooden image of Tzû-shih Peu-sa (Maitreya Bodhisattva) of a brilliant golden hue and of miraculous powers ; it is above ioo feet high ; it was the work of the Arhat Madhyantika who by his supernatural power thrice bore the artist to the Tushita Heaven to study Maitreya's beautiful characteristics ; the spread of Buddhism eastwards dates from the existence of this image.' ' That Ta-li-lo Iota is as exact a transcription of Darél or an earlier form of the name as Chinese phonetics would permit of, does not require any special demonstration. The close agreement of what both pilgrims tell us of the miraculous image of Maitreya there worshipped leaves no possible room for doubt as to their referring to the same territory, and its location in the present Darél is conclusively proved on topographical grounds by the details that both narratives record as to the bearing, distance, and character of the route connecting it with Udyana or Swat.8
The brief mention which the Tang Annals make of Ta-li-lo as situated to the north-east of Mêng-chieh-li and as forming ` the ancient territory of Udyana ',8a is probably derived from
Hsüan- tsang's account of Ta-li-lo.
Chinese data about Darèl.
4 Cf. J.A.S.B., xvii, Pt. u, p. 19 ; Ladiik, pp. 2, 46 sq. I take these references to publications not at present within my reach from Beal, Si-yu-ki, i. p. 134, note 37.
5 The merit of having recognized the name of Darèl in Hsüan-tsang's Ta-li-lo also belongs to General A. Cunningham ; see his Ancient Geography of India, p. 82.
6 See Watters, Yuan Chwang, i. p. 239. For the location of Mêng-chieh-li, the ancient capital of Swat (restored by Watters into ' Mangkil '), at Manglaor, cf. Colonel Deane's Note on Udyana and Gandhäira, J.R.A.S., 1896, p. 656 ; Serindia, i. p. 13.
7 The text of the Life, which presents substantially the same account, seems to imply that Hsüan-tsang only heard of Ta-li-lo and the road to it ; cf. Watters, Yuan Chwang, i. p. 239. But the detailed and graphic description supports the text of the Memoirs, which apparently implies an actual visit.
8 For references supporting details of the account given of this difficult route by the two Chinese travellers, cf. Serindia, i. pp. 6 (with note 8), 7 sq.
Sung Yün, too, shows a knowledge of this track from Darèl down the Indus to Swat, when he mentions the alternative route leading through Po-lu-le which his party wisely avoided, and describes its formidable difficulties ; cf. Chavannes, Voyage de Song Yun, pp. 28 sq. ; Serindia, i. p. 12.
It is probable that the references in Chinese historical texts of Former and Later Han times to forbidding mountain tracks leading by routes not clearly defined to Chi-pin, where travellers have to pass by ladders, wooden galleries, ropes over frightful precipices, &c., partly at least reflect reminiscences of this Indus river route ; cf. Wylie, J. Anthrop. Inst., x. p. 37 ; Chavannes, T'oung-pao, 1907, p. 217, with note 4.
8a Cf. Chavannes, Turcs occid., pp. 128, 311.