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. i]   OLD REMAINS IN WAKHAN   865

or Maulai sect of Islam, which is widely spread in the Hindukush valleys from Hunza to Chitral and worships its spiritual head in H.H. the Aga Khan of Bombay. Hence frequent intercourse is maintained across the snowy range to the south, notwithstanding the close watch which was said to be maintained by Afghan posts below all the more accessible passes. Thus I found it easy, while halted for one night at Zang, to arrange for the transmission of a mail to India via Chitral. My letters were carried across the difficult Wüst pass to Shuyist by two hardy Wakhis who had swum the Oxus on skins at night-time to escape Afghan attention, and reached Peshawar safely within little more than a week.

My march of September 2nd down the valley illustrated throughout that frequent interchange of shingle-covered fans, rocky promontories, and stretches of riverine sands with fields and meadow land to which Hsüan-tsang's description of Wakhan refers.8 Four miles below Zang we passed in full view of Kala-i-Panja (Fig. 398), the seat of the Mirs of Wakhan in modern times and now occupied by the chief Afghan post in the valley. Counting only some 15 households and situated by the side of a huge fan of detritus, this ` capital ' of Afghan Wakhân looked a very desolate place. But here, as farther down opposite Ishmarg (Fig. J97) and at other points of the valley, grand vistas opened to the south. Towering above narrow side valleys, and seemingly quite near, magnificent ice-clad peaks of the Hindukush main range could be seen, looking just like peaks of jade or silver, as Sung-yün and his fellow pilgrim Hui-shêng describe them, on their passage down the ` kingdom of Po-ho ' or Wakhân, A. D. 519.9 After passing verdant patches of cultivation at the hamlets of Shergin, Daresh, and Nichgar, separated by stretches of rocky or sandy waste, we reached a pleasant camp at Warang (Fig. 399), about 9,700 feet above sea-level.

From there I visited on the same day a reported ` Kafir ' fort situated high up on a spur to the north-west; which overlooks the canon-like debouchure of the Warang stream. The fort, also called Zangibar, was found to consist of a roughly built line of wall defending the narrow southern end of a rocky ridge on those sides on which unscalable cliffs falling off towards the foot of the Warang gorge do not afford protection (see the sketch-plan, P1. 46). Within the enclosed area, which measures about 108 yards with a width of only 20 yards, crumbling walls, built, like the enclosure of unhewn stone, without any plaster, divide small quarters. The remains of a much-decayed square tower and a narrow oblong platform with several small mounds on the top were also observed within the enclosure. Some hundred yards to the north, a modern looking tower, known as ` Tiip-khana', was said to have been used down to recent times to watch a track leading across the spur against raiders from the Shughnan side.

On September 3rd I visited a series of small cave-dwellings carved into the conglomerate rock face which overlooks the debouchure of the Warang stream about three-quarters of a mile west of the village. They extend in irregular groups for about half a mile, most of them situated on levels but little above the top of the talus slope below more or less vertical cliffs. The approach to them lies either over ` Rafaks ' or little galleries built with sticks, &c., now almost all broken, or through narrow passages cut in the rock and connecting them. All these caves are rough excavations, and none that I was able to examine exceeded 15 feet or so in depth or width. Many have partially fallen in, owing to erosion proceeding along the deep ravines which cut up the face of the cliffs. That these caves were occupied down to quite recent times was evident both from local

March to Warang.

`Kafir' fort of Zangibar

Cave-dwellings near Warang.

sqq.), and Sarikol. As graziers Wakhis are to be found in places along the higher slopes of the K`un-lun as far east as the Sanju pass.

This infiltration of an Iranian element to the south of the Hindukush and into the south of the Tarim basin has a distinct

linguistic and historical interest as regards an earlier period also. But the mere mention of it must suffice here.

8 Cf. Serindia, i. p. 63.

9 Cf. Chavannes, Voyage de Song Yun, p. 23 ; Marquart, Erdn-.lahr, p. 224.