The Gates of India: Being an Historical Narrative

by Sir Thomas Hungerford Holdich

The valley of Bamian stands at a level of about 8500 feet; the passes out of it northward to Balkh or southward to Kabul rise to 11,000 and 12,000 feet. It is the mystery of its unrecorded history and the local evidences of the departed glory of Buddhism, which render Bamian the most interesting valley in Afghanistan. Massive ruins still look down from the bordering cliffs, and for six or seven miles these cliffs are pierced by an infinity of cave dwellings. Little is left of the ancient city but its acropolis (known as Ghulghula), which crowns an isolated rock in the middle of the valley. Enormous figures (170 and 120 feet high) are carved out of the conglomerate rock on the sides of the Bamian gorge. Once coated with cement, and possibly coloured, or gilt, these images must have appealed strongly to the imagination of the weary pilgrim who prostrated himself at their feet. "Their golden lines sparkle on every side," says Huen Tsang, who saw them in the year A.D. 630, when he counted ten convents and 1000 monks of the "Little Vehicle" in the valley of Bamian.

Twelve hundred and fifty years later the great idols were measured by theodolite and tape, and duly catalogued as curiosities of the world's museum. We know very little of the later history of Bamian. The city was swept off the face of the valley by Chengiz Khan; and Nadir Shah, in later times, left the marks of his artillery on the face of cliffs and images. Moslem destroyers and iconoclasts have worked their wicked will on these ancient monuments, but they witness to the strength and tenacity of a faith that still survives to sway a third of the human race.

Chahilburj and Shahr-i-Babar (31 miles above Chahilburj at the junction of the Sarikoh stream with the Band-i-Amir) with the ruined fortresses of Gawargar and Zohak, wonderful for the multiplicity of its lines of defence, all attest to the former position of Bamian in Afghan history and explain its prominence in mediæval annals. And yet there is not much said about the road thither from Balkh, or onward to the "Indian city" of Kabul.

Idrisi just mentions the road connecting Balkh with Bamian, which he describes as follows: "From Balkh to Meder (a small town in a plain not far from mountains) three days' journey. From Meder to Kah (well-populated town with bazaar and mosque) one day's journey. From Kah to Bamian three days." Bamian he describes as of about the same extent as Balkh, built on the summit of a mountain called Bamian, from which issue several rivers which join the Andarab, possessing a palace, a grand mosque, and a vast "faubourg"; and he enumerates Kabul, Ghazni, and Karwan (which we find elsewhere to be near Charikar) amongst others as dependencies of Bamian.

It is not easy to identify Meder and Kah. The total distance from Balkh to Bamian is at least 200 miles by the most direct route via the Darra Yusuf. Forty miles a day through such a country must be regarded as a fine performance, even for Arab travellers who would think little of 50 or 60 miles over the flats of Turkistan. However, we must take the record as we find it, and assume that the camels of those days (for the Arabs never rode horses on their journeys) were better adapted for work in the hills than they are at present.

The inference, however, is strong that not very much was really known about this mountain region south of the Balkh plain. To the pilgrim it offered no terrors; but to the merchant, with his heavily laden caravan, it is difficult to conceive that 800 or 900 years ago it could have been much easier to negotiate than it is to the Bokhara merchants of to-day, who take a much longer route between the Oxus and Kabul than that which carries them past Bamian.

The province of Badakshan to the east (the ancient Baktria) is still but indifferently explored. It is true that certain native explorers of the Indian Survey have made tracks through the country, passing from the Pamir region to the Oxus plains; but no English traveller has recently done more than touch the fringe of that section of the Hindu Kush system which includes Kafiristan and its extension northwards, encircled by the great bend of the Oxus River. Kafiristan has ever been an unexplored region—a mountain wilderness into which no call of Buddhism ever lured the pilgrim, no Moslem conqueror (excepting perhaps Timur) ever set his foot, until the late Amir Abdurrahmon essayed to reduce that region and make it part of civilized Afghanistan. Even he was content to leave it alone after a year or two of vain hammering at its southern gates. Kafiristan formed part of the mediæval province, or kingdom, of Bolor; but it is always written of as the home of an uncouth and savage race of people, with whom it was difficult to establish intercourse. Kafiristan is, however, in these modern days very much curtailed as the home of the Kafir. Undoubtedly many of the border tribes fringing the country (Dehgans, Nimchas, etc.), who are now to be numbered amongst the most fanatical of Moslem clans, are comparatively new recruits to the faith, and therefore handle the new broom with traditional ardour; but they were not so long ago members of the great mixed community of Kafirs who, driven from many directions into the most inaccessible fastnesses of the hills by the advance of stronger races north and south, have occupied remote valleys, preserving their own dialects, mixing up in strange confusion Brahman, Zoroastian, and Buddhist tenets with classical mythology, each valley with apparently a law and a language of its own, until it is impossible to unravel the threads of their complicated relationship. Here we should expect to find (and we do find) the last relics of the Greek occupation of Baktria, and here are certainly remnants of a yet more ancient Persian stock, with all the flotsam and jetsam of High Asia intermingled. They are, from the point of view of the Kabul Court, all lumped together as Kafirs under two denominations, Siahposh and Lalposh; and not till scientific investigation, such as has not yet reached Afghanistan, can touch them shall we know more than we do now. No commercial road ever ran through the heart of Kafiristan, but there were two routes touching its eastern and western limits, viz. that on the east passing by Jirm, and that on the west by Anjuman, both joining the Kokcha River, which are vaguely referred to by our Arab authorities. That by Jirm is certainly impracticable for any but travellers on foot.

Badakshan (i.e. the province) was apparently full of well-populated and flourishing towns 1000 years ago. The names of many of them are given by Idrisi, but it is not possible to identify more than a few. The ancient Khulm (50 miles east of Balkh) was included in Badakshan. In Idrisi's day it was a place "of which the productions and resources were very abundant: there is running water, cultivated fields, and all sorts of vegetable productions." From thence to Semenjan "a pretty town, in every way comparable to Khulm, commercial, populated, and encircled with mud walls," two days' journey. Then we have "from Balkh to Warwalin" (a town agreeable and commercial with others dependent on it), two days. From Warwalin to Talekan, two days. Talekan is described as only one-fourth the size of Balkh, on the banks of a big river in a plain where there are vineyards. And then, strangely enough, we find "from Balkh to Khulm west of Warwalin is a two-days' journey. From Semenjan to Talekan, two days."

This is a puzzle which requires some adjustment. From Balkh to Khulm is about 50 miles and may well pass as two days' journey. But from Balkh to Warwalin is also said to be a two-days' journey, and from Warwalin to Talekan two days, whilst Khulm is two days west of Warwalin. The difficulty lies in the fact that all these places must be on a line running almost due east from Balkh. It was and is the great high-road of Badakshan in the Oxus plains. Moreover, Talekan has been fixed by native surveyors at a point about 150 miles east of Balkh which fully corresponds in its physical features to the description given of that place above. If, however, we assume 150 miles to represent six days' journey instead of four, the difficulty vanishes. We then have Balkh to Khulm, two days; Khulm to Warwalin, two days; and Warwalin to Talekan, two days. This would place Warwalin somewhere about Kunduz, which is, indeed, a very probable position for it.

Semenjan is important. Two days from Talekan; two days from Khulm; five days from Andarab.