- 1 L. Ju. Mankovskaja, in: Bukhara, A Museum in the Open, Tashkent 1991, p. 70 (in Uzbek, Russian and (...)
- 2 Narshakhi, The History of Bukhara, trans. R.N. Frye, The Medieval Academy of America, Cambridge, Ma (...)
1“Bukhara is one of the most ancient cities in Central Asia... Bukhara is at least 2,500 years old, just like Samarkand”1. This seems to be the official Uzbek version, based on finds of ancient coins and other remains, but is it plausible? If we examine the text of Narshakhi’s history of the city, however, we find the following: “This place, which today is Bukhara, [formerly] was a swamp; part of it was a bed of reeds and part planted with trees and a meadow. Some places were such that no animal could find footing there, because the snows melted on the mountains of the districts near Samarkand and the water collected... The area which is Bukhara was filled [with mud carried by the river] and the land became level. That river was the great river of Soghd and the filled area became Bukhara”2. We are speaking of the city of Bukhara rather than the large oasis designated by the same name. What should we believe?
- 3 A.A. Askarov (ed.), Gorodishche Paikand, Tashkent 1988, pp. 21-22.
- 4 Narshakhi, op. cit., transl. p. 18, text p. 26.
2Let us begin with geography and irrigation. There is no evidence that the Zarafshan river reached the Oxus in historical times and the existence of swamps and lakes in early times is almost certain. Ptolemy’s Oxian lake could have been situated in the oasis of Bukhara, although more likely it was at the combined delta of the Zarafshan and Kashka rivers. Before canals were created for draining the swamps, one may conjecture that only the high land in the oasis of Bukhara was occupied by settlements. Since the site of the city of Bukhara was not as elevated as the land around Paykand, it appears reasonable that such a site as Paykand, near the Oxus, and with easy access to Samarkand, and in the deltas of both the Zarafshan and Kashka rivers, would attract settlers searching for places to live in the oasis of Bukhara. Furthermore Paykand was situated on a low plateau of circa 100 sq.km.3 Also Narshakhi reports that Paykand was older than Bukhara and every ruler (of the oasis) made Paykand his capital, presumably in early pre-Islamic times4. We should remember that archaeologists believe that the deltas of rivers, such as the Tejen, Morghab and Hilmand rivers, were the places of earliest settlements in Central Asia, and the Zarafshan river delta would most likely conform to this pattern.
3Most people would agree that Dushanbe, in Tajikistan, only became a city in the Soviet period. However, archaeological excavations on a high mound on the banks of the river have revealed the existence of a settlement dating from the early years of our era, if not earlier. Likewise, finds of art objects testify to the existence of ancient craftsmen plying their arts in this area. But can we assert that Dushanbe is over two millennia old? Most likely people lived on or near the site of the city of Bukhara even in prehistoric times, but does it signify that the city is 2600 years in age? Semantics can be many faceted, or even misleading, but let us say that various settlements in the oasis of Bukhara were very early, but a full blown city on the site of the present Bukhara is later than proposed above. The exact date can hardly be ascertained, but the period of the great Kushan empire is a likely time, because earlier the Greco-Bactrians, as far as one can tell, had not established a strong presence in the Zarafshan basin of Central Asia beyond Marakanda, and even there their extended rule is questionable. Whether the discovery of Greco-Bactrian coins in the oasis of Bukhara can be proof of such rule there is also problematic.
- 5 On both sides of the river, but especially on the west side, south of the present Charjuy (Amol), a (...)
4Bactria was called a land of “a thousand cities”, under Greek rule, but north of the Hissar mountain range few traces of Greek settlement have been found. After the Saka expansion of the first century B.C., under the Kushans we find archaeological evidence of settlements along the Oxus river, with an extension into the oasis of Bukhara5. At the time of the Kushans in Bactria, urban life continued to flourish and irrigation was the key to the wealth of the land. The vast Kushan empire of the first two centuries of our era was a rival to the Roman and Chinese counterparts.
- 6 Some Arabic texts tell us that Numijkath was the name of a village some four farsakhs from the city (...)
5So we may tentatively assign the beginning of the prominent settlement on the site of the future city of Bukhara to the first century, but it seems that the city did not become the center of the oasis until, at the earliest, the end of the fifth century, or later. But even then rival towns had maintained their own rulers and local prosperity in trade. Probably Bukhara obtained a hegemony over the oasis only shortly before the Arab conquest at the end of the seventh century. From the geographies written in Arabic (Istakhri, Ibn Hauqal, Yaqut, etc.) we find that the name of the village on the site, or near the site, of modern Bukhara, was Numijkath6. The second part of the word, -kath or -kand, means “town” in Sogdian, while the first part may be related to the Sogdian word for “ninth”, or less likely to the word for “law” (from Greek nomos), or it may be a name of the river.
6The site of the city of Bukhara would have been a natural place for settlement, however, since the Zarafshan river divided here into several arms, according to the Arabic geographies. Because of the great fertility of the oasis, reported by the geographers, we may speculate that the whole area was called *pwk’r (Christian Sogdian fwq’r) in Sogdian, meaning something like “excellent, splendid”, which the Arabs made fâkhira with a similar meaning in their language, as they were wont to do, similar to the English. This is my suggestion for the origin of the name Bukhara, rather than the Volksetymologie, deriving the name from Indian Buddhist vihara. This appellation, applied to the district, was then transferred to the principal city, just as the name of the province of Parsa was given to the site of Persepolis.
7The rise of the city of Bukhara to great prominence, in my opinion, dates from the Arab conquests and the coming of Islam to Central Asia. Other than its favorable location and its affluence, as both a rich agricultural and textile area and a trading city, Bukhara became the great focus of Islamic learning in Central Asia. Why did it become such a center rather than Samarkand, or some other city? I believe this can be answered by the circumstances of the Arab occupation of Bukhara, as contrasted with other cities, except Marv which was the first site where the Arabs were settled in the homes of the local people. During almost a century of Umayyad rule, from the 660s to 750 of our era, the whole oasis of Marv was divided by the government, which then settled the Arab tribes in various villages. The Arabs were mainly interested in raiding across the desert to the Oxus river and beyond, using Marv as a base. There was no principal city in the Marv oasis, and Marv had been a military outpost of the Sasanian empire, as it continued to be under Arabs.
8Later than Marv, Arab tribesmen were placed in the homes of the inhabitants of the city of Bukhara rather than scattered throughout the oasis. The people of wealth and influence living there, including those in the city, were merchants as well as landlords, and trade with China and elsewhere was an important source of the wealth of the towns in the oasis. This does not mean that Marv was unimportant in trade relations, but the end of its position as the military and economic outpost of an empire based in Iran in a sense shifted the frontier to the east, and Bukhara became the “dome of Islam” in the east. Furthermore, the political conflicts of the Arab tribes settled in Marv did not help that oasis to retain its once pre-eminent position in the caliphate. Bukhara and Samarkand were no longer on the frontier but in the center of Sughd (Sogdiana), the richest and most populous part of Central Asia even unto this day. So one may conclude that Bukhara under the ‘Abbasids usurped the role held by Marv under the Umayyads.
- 7 On the common trade and textile interests of both Arabs and local people, cf. H. Mason, “The Role o (...)
9To recapitulate, the Arab tribesmen maintained their tribal organization in the Marv oasis, while in Bukhara the Arabs remained in the city. Many of the local nobility moved out of the city into their castles in the oasis of Bukhara, according to Narshakhi. In my opinion one of the factors which elevated the city of Bukhara into a center of Islamic learning was the early acceptance of Islam by its inhabitants, brought about in great degree by the intermarriage of local people with the Muslims, both Arabs and Persians, as well as others7. In any case, after initial resistance conversion proceeded apace, and the way was prepared for the flowering of Bukhara under the Samanids.
10Thus, the end of the Islamic caliphal frontier in Marv after the advent of the ‘Abbasids, together with the prosperity and tranquility, gave the “land across the river”, as the Arabs called Transoxiana, a possibility of cultural flowering, which is what happened. The pre-Islamic traditions of religious tolerance, and the prominence of various cultures in the area made it a breeding ground for intellectual growth among the people.
11By the end of the third century of Islam the number of Islamic scholars with the nisba “Bukhâri”, “Marvazi”, “Balkhi”, or another east Iranian or Central Asian city, was legion, whereas we hear of few “Shirâzi”, “Esfahâni”, or other west Iranians. I have suggested that an important reason for this was the continued organization of Sasanian state Zoroastrianism in villages as well as in some cities in the west, while in the east, where Christianity, Buddhism, Manichaæism and other faiths had flourished before Islam, the latter religion flourished and spread in a more tolerant milieu. In the west the Muslim communities were distinct from others which retreated into ghettoes, while in the east such clear separations were not as evident. This was the situation with the advent of the Samanids in the ninth century.
- 8 The lack of the sources on the Samanids is frustrating, for there are many questions one would like (...)
12The patronage of scholars writing in Arabic, and of poets and literati composing in Persian at the court of the Samanids in Bukhara is well known. The nature of the loose Samanid state, continuing traditions of local principalities of Central Asia with many minor courts, also provided such patronage, and that too has long been recognized. So a principal contribution of the Samanids to Islamic history was the fusing of Arab Islam with the cultures of Iran and Central Asia to produce an ecumenical Islamic civilization. One might object that the ‘Abbasids started such a movement, but under the Samanids Persian was accepted as an Islamic tongue, and the previous formula that to become a Muslim meant to become an Arab was expanded. Not only was Persian written in the Arabic alphabet, but also local languages such as Khorezmi and Mazandarani, but everything was swept aside by the simple lingua franca of Persian. There is confusion in the sources about the use of Persian as a written language at the court of the Samanids in Bukhara. Presumably Arabic remained the official written tongue throughout the tenth century, but Persian probably was an official spoken language. Yet surely some literati wrote their compositions in Persian, using the Arabic alphabet. That, however, is another story8.
1 L. Ju. Mankovskaja, in: Bukhara, A Museum in the Open, Tashkent 1991, p. 70 (in Uzbek, Russian and English).
2 Narshakhi, The History of Bukhara, trans. R.N. Frye, The Medieval Academy of America, Cambridge, Mass., 1954, p. 6; and text in Târikh-e Bokhârâ, ed. Modarres Reżâvî, Tehran 1351 Sh./1973, pp. 7-8.
3 A.A. Askarov (ed.), Gorodishche Paikand, Tashkent 1988, pp. 21-22.
4 Narshakhi, op. cit., transl. p. 18, text p. 26.
5 On both sides of the river, but especially on the west side, south of the present Charjuy (Amol), a number of sites with Kushan remains have been surveyed, and some excavated.
6 Some Arabic texts tell us that Numijkath was the name of a village some four farsakhs from the city of Bukhara.
7 On the common trade and textile interests of both Arabs and local people, cf. H. Mason, “The Role of the Azdite Muhallibid Family in Marw’s Anti-Umayyad Power Struggle”, Arabica 14 (1967), especially p. 204.
8 The lack of the sources on the Samanids is frustrating, for there are many questions one would like to have answered, such as the reasons for the silver shortage at the end of the tenth century, a precursor of the madrasa incipients guilds in Bukhara, and others.Haut de page
Pour citer cet article
Richard N. Frye, « Early Bukhara », Cahiers d’Asie centrale, 5/6 | 1998, 13-18.
Richard N. Frye, « Early Bukhara », Cahiers d’Asie centrale [En ligne], 5/6 | 1998, mis en ligne le 01 octobre 2010, consulté le 26 mars 2017. URL : http://asiecentrale.revues.org/527Haut de page
© Tous droits réservésHaut de page