Save
Authors:
EIr
Search for other papers by EIr in
Current site
Google Scholar Close
,
Richard H. Rowland
Search for other papers by Richard H. Rowland in
Current site
Google Scholar Close
,
Richard N. Frye
Search for other papers by Richard N. Frye in
Current site
Google Scholar Close
,
C. Edmund Bosworth
Search for other papers by C. Edmund Bosworth in
Current site
Google Scholar Close
,
Bertold Spuler
Search for other papers by Bertold Spuler in
Current site
Google Scholar Close
,
Robert D. McChesney
Search for other papers by Robert D. McChesney in
Current site
Google Scholar Close
,
Yuri Bregel
Search for other papers by Yuri Bregel in
Current site
Google Scholar Close
,
Abbas Amanat
Search for other papers by Abbas Amanat in
Current site
Google Scholar Close
,
Edward Allworth
Search for other papers by Edward Allworth in
Current site
Google Scholar Close
,
Peter B. Golden
Search for other papers by Peter B. Golden in
Current site
Google Scholar Close
,
Ian Matley
Search for other papers by Ian Matley in
Current site
Google Scholar Close
,
Ivan M. Steblin-Kamenskij
Search for other papers by Ivan M. Steblin-Kamenskij in
Current site
Google Scholar Close
,
Gerhard Doerfer
Search for other papers by Gerhard Doerfer in
Current site
Google Scholar Close
,
Keith Hitchins
Search for other papers by Keith Hitchins in
Current site
Google Scholar Close
, and
Walter Feldman
Search for other papers by Walter Feldman in
Current site
Google Scholar Close

(79,951 words)

This series of articles covers Central Asia.

This series of articles covers Central Asia.

A version of this article is available in print

Volume V, Fascicle 2, pp. 159

CENTRAL ASIA

CENTRAL ASIA i. Geographical Survey

The central expanse of the Asian continent, the land mass situated approximately between 55° and 115° E and 25° and 50° N, comprises two geographically distinct areas. The western part includes the Transcaspian plains and the low tablelands between the Aral Sea and the Tien Shan (lit. “heavenly mountains”) range; it is generally equivalent to the territory of western Turkistan (the Turkmen, Uzbek, and Tajik Soviet Socialist Republics and the southern and western portions of the Kazakh Soviet Socialist Republic). The eastern part encompasses the high plateaus and mountainous perimeters of the Tarim basin (approximately equivalent to Eastern Turkistan, now Sinkiang [Xin-jiang] Uighur province of the People’s Republic of China) and Tibet, the area north of the Tien Shan mountains as far as the southern Siberian plains and the Altai mountains (the northern and eastern portions of the Kazakh S.S.R.), and the Gobi desert (comprising parts of the Mongolian People’s Republic and Chinese Inner Mongolia), along with the high mountain ridges thrusting east and south into China and Southeast Asia.

Although the imprecise term Central Asia has been used to designate various regions within this vast area, most Western scholars apply it to western Turkistan, designating the area between 70° and 100° E and 25° and 45° N as Inner or Innermost Asia (Haute Asie, Inner-Asien, Tsentral’naya Aziya; See chinese turkistan).

Land forms. The Eurasian continental plate over-thrusts the Indian subcontinent, and Central Asia thus belongs to an earthquake zone that extends through the Caucasus and Anatolia and across the Mediterranean and the Atlantic to the Caribbean islands; particularly along the ridges of the Alai range, the Hindu Kush, and the mountains bordering Persia earthquakes are common, though not accompanied by volcanic activity. The landscape of western Turkistan is characterized by mountain ranges separated by high plateaus, an extreme continental climate, severe aridity, and resulting scarcity of vegetation and animal life.

The crest lines of the great Central Asia mountain systems generally lie on east-west axes, which seem to radiate from the Pamirs. A cluster of parallel ranges curving to the southwest, then to the west, are generally known as the Hindu Kush, the Paropamisos of the Greek geographers (cf. Pauly-Wissowa, XVIII/4, cols. 1778-79). The southernmost of these ranges bends sharply westward, then south, continuing as the Sulaiman mountains, which separate the Iranian plateau from the lowlands of the Indus (Sindh), though, as there are few elevations higher than 2,000 m, they do not constitute a major geographical barrier. The range extends in various chains south and west through Baluchistan, thus linking the Pamirs with the Zagros system along the southern and southwestern margins of the plateau. The main ranges of the Hindu Kush, however, extend directly southwest, gradually declining in elevation to an average of around 4,000 m; they are linked to the volcanic Transcaucasian plateau by the Bīnālūd, Alborz, and Qarā Dāḡ chains, which form the northern boundary of the Iranian plateau, 1,000 m lower on the average than the Zagros. The highest peaks of the Hindu Kush include Tirich Mir at 7,750 m in Pakistan on the Afghan border and a number exceeding 6,000 m in northeastern Afghanistan; those in the western portion of the range and in the Bīnālūd rarely exceed 3,000 m. In Tajikistan the Alai range extends straight west from the Pamirs, with a fairly even crest line; it is paralleled just to the south by the Trans Alai and still farther south by the Darvaz (Qarā Tegīn) range, which, however, traces an arc similar to that of the Hindu Kush, trailing off to the southwest in several minor subranges. The highest peaks in this region are Pik Kommunizma (Communism, formerly Garmo, peak, 7,495 m) in the Darvaz chain and Pik Lenina (Lenin peak, 7,127 m) in the Trans Alai. In the western Alai, 110 km southeast of Samarkand, Pik Chimtarga rises to 5,490 m, west of which the range trails off in a southwestern branch and a shorter western one.

The steppes on the northern and western flanks of the great mountain ranges extending east from the Caspian Sea, respectively the Kopet-Dag/Bīnālūd mountains, the Alai, the Hindu Kush, and the Tien Shan, are at elevations of 100-500 m. South of these ranges the Iranian plateau, which extends into Afghanistan, lies between 1,000 and 1,500 m. The uplands of the Pamirs are at elevations of 3,000-4,500 m. In all three zones the land generally tilts from south to north. The soil from central northern China to the Atlantic Ocean at every altitude, even beneath the sand of the deserts, is mostly yellow loess and very fertile when there is enough water.

Water. Because of increased precipitation at higher elevations, which may reach a maximum of 1,000 mm a year, the snow cover and particularly mountain glaciers are the main sources of water for Central Asia. In the Pamirs the snow line lies as high as 5,500 m. In the latitudinal ranges it is often 500-700 m lower on the northern slopes than on the sunny southern slopes. The glaciers, most of them valley glaciers, are associated with the highest peaks. In Soviet Central Asia alone 1,700 glaciers total 11,000 km2, five times the surface of the Caucasian glaciers; they also melt at a much more rapid rate than those in the Caucasus or in Switzerland. The largest is the Fedchenko glacier, discovered in the early 20th century near Communism peak north of the Yazgulem pass; it is 77 km long, 2-5 km wide, and 550 m thick, with a volume exceeding 200 km3. From an altitude of 5,330 m it extends down to 2,904 m and is fed by thirty-seven tributary glaciers, some of them more than 10 km long. The Fedchenko glacier melts at a rate of 27 cm a day, 100 m a year; in 1924-35 it lost a total of 282 m. Also in the northwestern Pamirs are the Garmo (29 km long), Fortambek (25 km), and Finsterwalder (16 km) glaciers. In the Alai range the Zeravshan glacier, 25 km long and 200 m thick, is located in the valley between the Turkistan and Zeravshan chains. It melts at a rate of 27 cm a day and is the source of the Zeravshan river. Glacial melting at high altitudes is responsible for the summer flow maximum of Central Asia rivers. A spring maximal flow occurs when these rivers are fed by the melting snow on the lower slopes of the mountains bordering the fertile loessial piedmont plains on which most of the irrigated agriculture of Central Asia is concentrated.

The major rivers in western Central Asia are the Syr Darya (Jaxartes) and Amu Darya (Oxus; see āmū daryā) rivers. The Syr Darya, almost 1,400 km long begins as a small stream in the Farḡāna mountains and is fed by the Naryn, which rises farther east on the southern slopes of the Terskey Ala Tau, south of Lake Issyk Kul, and flows west through the Farḡāna valley. At Leninabad, the western gateway to the Farḡāna valley, it turns sharply northward, bypasses the oasis of Tashkent, and flows northwest, emptying into the northeastern tip of the Aral Sea (in 1987 40 m above sea level). Important towns in the Syr Darya basin include Chimkent and Turkistan in southern Kazakhstan and Kazalinsk in the delta. Farther south the much longer Amu Darya (ca. 2,500 km) drains the Pamirs and the northern Hindu Kush. Its upper course, which defines the border between Afghanistan and Tajikistan, is called Pyandzh as far as the confluence with the Vaḵš west of Kirovabad; the Vaḵš rises in the Alai as the Kyzyl Su (Pers. Sorḵāb, lit. “red water”) and follows the Darvaz range south. Dushanbe, the capital of Tajikistan is situated 1,220 m above sea level on the Kafirnigan river, which joins the Amu Darya west of the confluence with the Vaḵš. From there the Amu Darya flows generally northwest and empties into the Aral Sea from the south; along its lower course, in ancient Choresmia (q.v.), the important towns of Khiva (Ḵīva) and Urgench (Ūrganj) are located. Both the Syr Darya and Amu Darya have fed the irrigation systems of the fertile Central Asian plains since antiquity; furthermore, as they flow through these unforested expanses they lose a considerable volume of water through evaporation. These natural loesses, however, are modest compared to the diversion of nearly all the water of these rivers to irrigated agriculture in the Soviet era. One result of this is that virtually no water from either the Syr Darya or Amu Darya flows into the Aral Sea, which is shrinking at a rapid rate. From 1960 to 1987 it lost two-fifths of its area and two-thirds of its water volume.

Between the two great rivers the Zeravshan (Zarafšān) rises at the western end of the Alai range and flows west, watering the entire cultivated zone between Samarkand and Bukhara, before disappearing in the desert sands. The main population centers in Uzbekistan, aside from Tashkent, lie along its course, including Ziyauddin (Żīāʾ-al-Dīn), Meymana, Karmina, and Bukhara. South of the Amu Darya two former tributaries, the Morḡāb and Harīrūd (Tedzhen) rivers, drain the western Hindu Kush but also dry up in the sands of the Kara Kum (“black sand”) desert in Turkistan. The impressive ruins of ancient Marv are situated not far from where the Morḡāb disappears. Still farther south the Atrak rises between two ranges of the Bīnālūd mountains northwest of the town of Qūčān and flows west across the barren steppes east of the Caspian, forming part of the border between Persia and the Turkman S.S.R.; it empties into Gasan-Kuli bay near Chikishlyar. Stormy Lake Karakul (elev. 3,780 m) in the northeastern Pamirs, southeast of Lenin Peak, is the source for a number of streams that water the high plateaus in that region, a land of rural valley settlements of essentially uniform size.

Climate. An extreme continental climate, with hot summers, cold winters, and sharp differences in temperature between day and night, characterizes the whole of Central Asia. At Kazalinsk near the mouth of the Syr Darya the mean temperatures in July and January are respectively 26° C and - 11.8° C (slightly higher than the - 15.1° C in Arkhangelsk, at 64.5° N); almost directly south, at Turtkul on the lower Amu Darya 28° C and - 5.1° C; and at Termez on the Soviet-Afghan border 31.5° C and 1.7° C. The winter lows preclude growth of most temperate-zone vegetation. Temperatures also rise and fall with extreme rapidity.

Because of the distance from the oceans, there are few clouds and little atmospheric moisture or precipitation, and solar irradiation is correspondingly strong. Much of the land surface is thus arid; the lowlands are preponderantly steppes, interspersed with stretches of sandy desert, particularly the Kara Kum on the Transcaspian plains (Turkmen SSR) and the Kyzyl Kum east of the Amu Darya. Wherever the natural water supply and precipitation are sufficient, the steppes and the fertile plateaus of the lowest zone are covered with grass. South of 45° N, the steppes begin to dry out by early July and have usually remained barren until late October or November, when some rain-bearing cloud formations move in from the Atlantic zone in the northwest. Over the last fifty years, however, there have been noticeable climatic changes, and it is now not unusual for compact cloud formations or overcast skies to appear in midsummer in some areas of western Turkestan as far south as the central course of the Syr Darya and the oasis of Tashkent. This change represents an extension of the influence of the northwestern maritime climate that formerly never penetrated beyond Moscow and central Russia in summer. Only rarely do monsoon clouds diverted from the regular southwest-northeastern path of the South Asian monsoons reach Central Asia. In Soviet Turkmenistan and western Uzbekistan the severe aridity of the Central Asian climate has also been mitigated to some extent by large-scale irrigation and the construction of dams (e.g., the Kara Kum canal). Evaporation from these projects contributes a certain amount of atmospheric moisture and has led to a more humid climate throughout the westernmost portions of Central Asia. Historical climatology shows that current arid conditions result from a recent period of desiccation lasting 1,500-2,000 years, which particularly affected northern Africa and Arabia but also to a lesser degree the Asiatic continent.

EIr

Bibliography

  • B. P. Alisov, Klimat SSSR, Moscow, 1969.
  • V. Chupakin, Fizicheskaya geografia Tyan’ Shanya, Alma-Ata, 1964.
  • Fizikogeograficheskiĭ atlas mira, Moscow, 1964.
  • N. Gvozdetskiĭ, Fizicheskaya geografiya SSSR. Aziatskaya chast’, Moscow, 1970.
  • P. Lydolph, Climates of the Soviet Union, Amsterdam, 1976.
  • P. Micklin, “Desiccation of the Aral Sea: A Water Management Disaster in the USSR,” Science 241/4870, September, 1988, pp. 1170-76.
  • E. Murzaev, ed., Srednyaya Aziya, Moscow, 1968.
  • M. P. Petrov, Pustyni Tsentral’noĭ Azii, 2 vols., Moscow, 1966.
  • A. A. Rafikov, “Ratsional’noe izpol’zovanie i okhrana vodnykh resursov Uzbekistana,” Geografia i prirodnye resursy 4, 1986, pp. 44-47.
  • S. P. Suslov, Fizicheskaya geografia SSSR. Zapadnaya Sibir’, Vostochnaya Sibir’, Dal’niĭ Vostok, Srednyaya Aziya, 2nd ed., Moscow, 1956.

CENTRAL ASIA ii. Demography

The combined population of the Uzbek, Kirgiz, Tajik, and Turkmen Soviet Socialist Republics totals more than 30 million people, one tenth of the population of the Soviet Union. Owing to a relatively high birth rate and a low death rate, it is also the most rapidly growing segment, at an average annual rate of approximately 3 percent a year, comparable to those of the most rapidly growing nations of the underdeveloped world. In general the Central Asian population is characterized by a low level of urbanization, very uneven geographical distribution of the oasis type, a high density among the rural population, and a large Muslim majority, including a number of Iranian-speaking peoples. The chief Russian-language sources on the demography of Central Asia are the Russian and Soviet censuses of 1897, 1926, 1939, 1959, 1970, 1979, and 1989, as well as annual population estimates (Tsentral’nyĭ Statisticheskiĭ Komitet, 1899-1905; Tsentral’noe Statisticheskoe Upravlenie SSSR, 1929, 1962-63, 1972-74, 1984-87; Pravda, 1989; 1939 data were published in the 1959 census and elsewhere; additional 1979 data have also been published in Goskomstat SSSR, 1988). Although there was no official population census before 1897, there have been estimates of the population of Central Asia for this period. For example, Karakhanov (pp. 226-28) provides annual total population estimates for the four republics for 1865-1982, and official estimates were made for political units of Central Asia in 1885 (Tsentral’nyĭ Statisticheskiĭ Komitet, 1887). The republic of Kazakhstan is sometimes included in “Central Asia,” but it will not be considered here, as it is considered today by the Soviets as a separate economic region from that of Central Asia, which consists of the four republics listed above, and has an insignificant Iranian-speaking population (in 1979 there were only 19,293 Tajiks and 17,692 Kurds; no other Iranian-speaking peoples were listed for Kazakhstan in 1979).

General demography of Central Asia. Since the turn of the century the population of Central Asia has increased fivefold, from approximately 6 million to 14 million in 1959 and 33 million in 1989. In recent years the number of births per 1,000 peoples has hovered at around 30-40, while females in the reproductive years have continued to bear an average of four to seven children. Conversely the number of deaths has dropped to fewer than 10 per 1,000. Migration has played only a small role in this rapid growth.

Unlike most other regions of the Soviet Union, the majority (ca. 60 percent) of the population in Central Asia still lives in rural areas. There are some fairly large cities in the region, however, the most important of which is Tashkent, the fourth largest city of the USSR, with more than 2 million people. Settlement is concentrated in river valleys and along canals, especially the Amu Darya, Syr Darya, and Zeravshan (Zarafšān) river valleys and the Kara Kum canal zone. The majority of the population is located in the relatively small area extending eastward from Bukhara through Samarkand to the Zeravshan and Fergana (Farḡāna) valleys and along the upper Syr Darya and, on a branch of the Syr Darya, in Tashkent and its surrounding countryside. Rural population densities in these areas often exceed 200 people per km2, the highest in the USSR. Other notable areas of concentration are the lower Amu Darya valley just south of the Aral sea, in the Kara Kum canal zone between the Amu Darya and the Caspian, and in the Vakhsh valley of the Tadzhik SSR. On the other hand, great stretches of territory are uninhabited or very sparsely populated, especially the Kara Kum desert in the southwest between the Amu Darya and the Caspian Sea and the Kyzyl Kum desert between the Syr Darya and the Amu Darya, and the mountainous zones of the extreme south and east, including the Gorno-Badakhshan area of the Tadzhik SSR (Glavnoe Upravlenie Geodezii i Kartografii, pp. 130-31; Akademiya Nauk Tadzhikskoĭ SSR, pp. 118-19). Approximately 60 percent of the population is in the Uzbek SSR and 10-16 percent in each of the other three republics.

In 1979 81 percent of the population of Central Asia was Muslim; 15 percent, or nearly 4 million, consisted of ethnic Russians, concentrated chiefly in cities (1989 nationality data are not yet available). Of the Muslim nationalities the Uzbeks were clearly the largest; with more than 12 million people in 1979, they had become the third largest national group in the USSR after the Russians and the Ukrainians. There were nearly 3 million Tajiks, and the Kirghiz and Turkmen populations totaled about 2 million each. Other notable Muslim nationalities included Kazakhs (740,000), Tatars (730,000), and Karakalpaks (300,000).

The overwhelming majority of the Muslim populations speaks Turkic languages, especially the Uzbeks, Turkmen, Kirghiz, Kazakhs, Tatars, and Karakalpaks. In 1979 this language group accounted for 86 percent of the Muslim population of the four republics and 69 percent of the total population of Central Asia.

Iranian-speaking populations. Virtually all the remaining Muslim population (11 percent of the total population of Central Asia in 1979) speaks Iranian languages. This group is composed almost entirely of Tajiks (2.9 million, or 98 percent), 77 percent of them living in the Tajik SSR and 21 percent in the Uzbek SSR. Ethnic Tajiks constituted 59 percent of the population of the Tajik SSR, the remainder being mostly Uzbeks (23 percent) and ethnic Russians (10 percent). The Tajiks are among the least urbanized and most rapidly growing nationalities in the USSR. In 1979 only 28 percent lived in urban centers, and the population growth rate, 3.4 percent a year between 1970 and 1979, was the highest among the fifteen nationalities that enjoy SSR status. The native language of 98 percent of them is Tajik, the Central Asian variant of Persian, which includes many archaisms in phonology (e.g., preservation of majhūl vowels) but is strongly influenced, especially in local dialects, by the morphology and syntax (particularly the verbal system) of the surrounding Turkic languages (see xv, below).

It is difficult to trace historical population trends among individual nationalities in the USSR because of changing ethnic classifications and governmental decisions on whether or not to include a particular national population in published census reports (Lewis, Rowland, and Clem, pp. 42-48, 388-92). It is especially difficult to assess long-term population trends among the Tajiks because the nationality was defined differently in the 1897 Russian census and in the Soviet censuses of 1926 and later. One major source of confusion is the language category “Sarts,” which was included among the Turkic languages in the 1897 census. Some sources agree with this classification, but according to others the Sarts, who were regarded as the settled peoples of Central Asia as opposed to the nomadic peoples, spoke either a Turkic language or an Iranian language (Tajik); the name has even been used practically synonymously with Tajik (Akiner, p. 304; Bennigsen and Wimbush, p. 92; Krader, pp. 55-56; Wixman, p. 174). In the 1897 census more than 900,000 Sarts were counted in the Russian empire, and there were also 350,000 Tajiks listed separately. Virtually all of the “Sarts” and the “Tajiks” were in Central Asia. The last figure represents a serious underestimate of the Tajik population at that time, however. Not only may a number of Tajiks have been included in the “Sart” category, but also a large part of what is now the Tajik SSR was then part of the khanate of Bukhara (q.v.), a vassal state that was not officially part of the Russian empire and thus not included in the 1897 census. It has been estimated elsewhere that perhaps 30 percent of the population of Bukhara at that time was Tajik (Becker, p. 7; Lorimer, p. 36). As the total estimated population of Bukhara was 2-2.5 million, the Tajik population must have been 600,000-750,000, approximately twice the official census figure for the Russian contingent. It thus appears that the total Tajik population of Central Asia approached 1 million or perhaps even exceeded it. In 1926, however, it was slightly less than 1 million (978, 680, or 980, 509 if the Yaḡnōbī are included; see below). The figure increased to 1,229,300 in 1939 (Kozlov, p. 285), 1,396,939 in 1959, 2,135,883 in 1970, and 2,897,697 in 1979, when Tajiks were the eleventh largest nationality in the USSR. In that year they accounted for 1.1 percent of the total population of the country, whereas in 1959 they had been only the sixteenth largest nationality, accounting for 0.7 percent of the total population.

In the 1979 census, only three other Iranian-speaking nationalities were recognized in Central Asia: “Persians” (Persy), who have been in Central Asia for centuries, Baluch, who have arrived only since the late 13th/19th century, and Kurds, who reside primarily in Transcaucasus when in the USSR. In 1979 there were 31,313 Persians (Persy) and 18,997 Baluch (Beludzhi). The totals for Persians were not broken down by republic in 1979, but in 1970, when the total of Persians (Irantsy or Persy) in the USSR was 27,501, three-fourths (20,525) lived in the Uzbek SSR and one-fourth in the Turkmen SSR. Virtually all of the Baluch live in the Mary (Marv oasis) area of the Turkmen SSR. Soviet censuses provided figures for these groups in 1926 (43,971 Persians, also called Farsi; 9,188 Iranians, or Irani; and 9,974 Baluch) and 1959 (20,766 Persians or Irantsy and 7,842 Baluch). The listed numbers of Kurds in Central Asia in 1926, 1959, 1970, and 1979 were 2,308, 7,046, 10,907, and 9,544, respectively (the 1959 and 1970 data are for both the Kirgiz and Turkmen SSRs, while the 1979 data are for the Kirgiz SSR only).

Other Iranian-speaking peoples of Central Asia are apparently counted with the Tajiks. Most notable among them are the speakers of East Iranian Pamir languages (Mountain Tajiks; the languages were formerly also called Ḡalča/Ghalchah, see, e.g., Grierson, pp. 3-4) in the Gorno-Badakhshan Autonomous Oblast, eastern Tajik SSR. Eight specific nationalities are recognized there, and the populations of seven of them have been estimated for 1960 (Akiner, pp. 374-79): Šuḡnī (20,000); Rōšānī (7,000-8,000); Wāḵī (6,000-7,000); Bartangī (q.v.; 3,000-4,000); Yazḡulāmī (1,500-2,000); Ḵūfī (1,000-1,500); and Eškāšmī (500). No estimate was provided for the Bajūī. Also now included with the Tajiks are the Yaḡnōbī, who were listed separately in 1926 (population 1,829). In addition, Afghans were listed in the 1926, 1959, and 1970 censuses (populations of 5,348, 1,855, and 4,184, respectively) but not in the 1979 census.

See also AFGHANISTAN iv. ethnography; v. languages.

Richard H. Rowland

Bibliography

  • Akademiya Nauk Tadzhikskoĭ SSR, Atlas Tadzhikskoĭ Sovetskoĭ Sotsialististicheskoĭ Respubliki, Dushanbe, 1968 (detailed population maps of the Tajik Republic).
  • S. Akiner, Islamic Peoples of the Soviet Union, London, 1983.
  • S. Becker, Russia’s Protectorate in Central Asia, Cambridge, Mass., 1968.
  • A. Benningsen and S. E. Wimbush, Muslims of the Soviet Empire, Bloomington, Ind., 1986.
  • A. J. Coale, B. Anderson, and E. Harm, Human Fertility in Russia since the Nineteenth Century, Princeton, 1979.
  • Glavnoe Upravlenie Geodezii i Kartografii pri Sovete Ministrov SSSR, Atlas SSSR, 1986 (maps of population distribution in Central Asia).
  • Goskomstat SSSR, Naselenie SSSR, 1987, Moscow, 1988.
  • G. A. Grierson, Linguistic Survey of India X: Specimens of the Eranian Family, Delhi, etc., 1921; repr. 1968.
  • S. I. Islomov, Demografiya Tadzhikistana, Dushanbe, 1985.
  • M. K. Karakhanov, Nekapitalisticheskiĭ put’ razvitiya i problemy narodonaseleniya, Tashkent, 1983. V. I. Kozlov, National’nosti SSSR, Moscow, 1986.
  • L. Krader, Peoples of Central Asia, Bloomington, Ind., 1963.
  • R. A. Lewis and R. H. Rowland, Population Redistribution in the USSR, New York, 1979.
  • Idem and R. S. Clem, Nationality and Population Change in Russia and the USSR, New York, 1976.
  • F. Lorimer, The Population of the Soviet Union, Geneva, 1946.
  • I. R. Mullyadzhanov, Demograficheskoe razvitie Uzbekskoĭ SSR, Tashkent, 1983.
  • Naselenie Sredneĭ Azii, Moscow, 1985. Pravda, 29 April 1989, p. 2.
  • Razvitie narodonaseleniya i problemy trudovykh resursov respublik Sredneĭ Azii, Tashkent, 1988.
  • Tsentral’noe Statisticheskoe Upravlenie SSSR, Vsesoyuznaya perepis’ naseleniya 1926 goda, Moscow, 1929 (1926 Soviet census).
  • Idem, Itogi vsesoyuznoĭ perepisi naseleniya 1959 goda, Moscow, 1962-63 (1959 Soviet census).
  • Idem, Chislennost’ i sostav naseleniya po dannym vsesoyuznoĭ perepisi naseleniya 1979 goda, Moscow, 1984 (1979 Soviet census).
  • Tsentral’nyĭ Statisticheskiĭ Komitet, Sbornik svedeniĭ po Rossii za 1884-1885 gg., St. Petersburg, 1887.
  • Idem, Pervaya vseobshchaya perepis’ naseleniya Rossiĭskoĭ Imperii, 1897 g., St. Petersburg, 1899-1905 (Russian census of 1897).
  • R. Wixman, The Peoples of the USSR, Armonk, N.Y., 1984.
  • Annual population totals and data on fertility and mortality can be found in the annual statistical series for the USSR, Tsentral’noe Statisticheskoe Upravlenie SSSR, Narodnoe khozyaĭstvo v SSSR (through 1987), as well as those for the individual Central Asian republics. Population data can also occasionally be found in the journal Vestnik statistiki.

CENTRAL ASIA iii. In Pre-Islamic Times

The sources. The main evidence for the history of Central Asia before the coming of Islam comes from archeological excavations, while written sources contain little information. Historical texts include the Old Persian inscriptions by Darius I and Xerxes I (6th-5th cents. b.c.e.), the Chinese dynastic histories (covering events from the 2nd cent. b.c.e. onward; for a summary see Samolin) and accounts of Chinese Buddhist pilgrims (the earliest was Fa-xian/Fa-hsien ca. 400 c.e.; see buddhism i), fragmentary Greek writings of travelers or geographies (such as Ptolemy and Strabo), early Arabic histories and geographies, plus indigenous inscriptions and documents such as the early Sogdian Ancient Letters from the Dunhuang times and the later Sogdian letters from Mt. Mug east of Samarkand. Scattered Sogdian, Bactrian, Choresmian, Parthian, and Middle Persian, as well as Indian inscriptions on wall paintings, ostraca, or on boulders or coin legends give some assistance in reconstructing history (for a Sogdian inscription on a wall painting from Afrāsīāb, old Samarkand, see Frye, 1967, p. 35; for the Parthian inscriptions from Nisa see Dyakonov and Livshits; for Bactrian, Middle Persian, and Prakrit in Kharoṣṭhī script see Staviskiĭ; numerous inscriptions in Sogdian, Bactrian, Middle Persian, Chinese, and Indian Prakrits in both Brahmi and Kharoṣṭhī scripts were found on boulders in the upper Indus valley, Sims-Williams, 1989). Most of the inscriptions contain almost exclusively names of merchants, however, although coin legends in Choresmian and Sogdian also provide names of rulers and towns (Vamberg and Smirnova). The literary texts, that is, the Avesta and the Pahlavi texts, contain only little information. For the Avesta see avestan geography (with further refs.), and for the Pahlavi texts see in particular the geographical chapters of the Bundahišn (chaps. 9 on the mountains, 10 on the seas, 11 on the rivers, 12 on the lakes) and the Pahlavi version of Vidēvdād, chap. 1, with Christensen’s commentary.

The earliest times. From excavations much information has been obtained about the daily life of both rulers and ruled. The discovery of Neanderthal remains in a cave at Teshik-Tash, Uzbekistan, indicates an early human settlement in Central Asia (cf. Okladnikov). Although subsequent finds are few, one may suppose a continuous occupation through the migration of the Indo-Iranian peoples into the area from the northwest at the beginning of the second millennium. Several questions about these migrations are difficult, if at all possible, to answer. First of them is the question about the identity of the aboriginal population in Central Asia at the time of the migrations, and second is the time of the split between Iranians and Indians and the duration of Indian occupation of the area. A generally accepted belief is that the Indians split from the Iranians sometime during the first half of the second millennium b.c.e. and preceded the Iranians, who moved onto the Iranian plateau at the beginning of the first millennium b.c.e. (for references to this theory see Ghirshman). Differing opinions have been expressed by M. Mayrhofer and I. M. Dyakonov. It is uncertain whether ancestors of the Dravidians, such as the Brahui in present Baluchistan, or of the Burusho (see burushaski) in Hunza, inhabited large parts of Central Asia in early times before the expansion of the Indo-Iranians. As we have no records of identifiable pre-Aryan peoples in Central Asia or in Northwest India the suggestion above is nothing more than a plausible guess.

The coming of the Iranian tribes. In any case, neither the aborigines nor the Indians remained in Central Asia; either they were absorbed or were pushed out by the Iranians, who settled in the area by tribes. Archeology is our only source for this pre-literate period of the history of Central Asia. For a survey of early sites in Soviet Central Asia see Kohl, ed., and, for a more comprehensive survey of sites in Central Asia after the early Bronze Age, Koshelenko (1985). Masson, 1966 (1981) contains useful bibliographical surveys of the main archeological sites (pp. 225-30). A. Koshelenko makes the following divisions for this early period: 1. Early Iron Age in the areas of the Marv oasis, northern Parthia, the Sarakhs oasis, northern Bactria, Fergana, the Tashkent oasis, Ustrushana; 2. the Ancient era, in which Choresmia, Fergana, and Sogdiana are added to the Iron Age areas. The Bactrians settle in present northern Afghanistan and Tajikistan, while the Sogdians occupy the Zarafshan river and Fergana valleys.

In Bactria sites along the rivers have provided information about the material culture of the early Iranians who settled there. Along the Surkhan Dar’ya numerous larger and smaller sites have been explored; such as old Termez, Dal’verzin Tepe, Zar Tepe, Aĭrtam (q.v.), and Khalchayan. Numerous sites are also located in the Hissar valley, on the lower Kafirnigan valley, on the right (west) bank of the Vakhsh river, on the lower left bank of the Vakhsh (Lagman, Kafyr Kala, etc.), on the valley between the Tair Su and Kyzyl Su (for references and complete lists of these sites see Koshelenko, pp. 250-72). Among the sites in Sogdiana the following may be mentioned: Kashka Dar’ya with centers Karshi and Shakhrisiyab Kitab (Kalyandar Tepe) and Kurgan Tepe; other sites are located in the vicinity of Samarkand (other than Afrāsīāb) and in the Bukharan oasis (see Koshelenko, pp. 273-92). Numerous sites have been explored in the Fergana valley (Koshelenko, pp. 304-316) and in the Tashkent oasis sites (e.g., Dal’verzin Tepe; Koshelenko, pp. 297-303). The Choresmians settled throughout the region to the south of the Aral Sea around the Oxus River, although they originally may also have roamed south of that region. Among the principal sites of Choresmia are Gyaur Kala, Toprak Kala, Kzyl Kala, and Koĭ-Krylgan Kala. It is noteworthy that the pattern of settlement in Choresmia, unlike in other areas, was the fortified castle or kala. The archeological work in this area was led for many years by S. P. Tolstov and has been described in the reports on the excavations of the Choresmian expeditions, as well as in other publications (e.g., Tolstov), in which a detailed picture of the material culture of that region is drawn. To the north of the oases in the steppes were the nomads called Scythians by the Greeks and Sakas in the Old Persian inscriptions (Herodotus, bk. 1; Pyankov; Kent, Old Persian; see also, e.g., Bongard-Levin and Grantoskij; Litvinskiĭ; Akishev, 1984).

Central Asia in the Avesta. Names of countries in northeastern Iran are listed in several passages in the Avesta. Among the oldest is the one in Mihr yašt (Yt. 10. 14), where “Parutian Iškata, Haraivian Margu, Sogdian Gava, and Choresmia” are mentioned (Gershevitch, 1967, p. 81) as parts of the Aryan lands that MiΘra surveys when he approaches over Mount Harā in front of the rising sun. In the Vidēvdād (Vd. 1.4-7) we are told that among the best places created by Ahura Mazdā were Gava, inhabited by Sogdians, Mouru (Margiana), the strong and truthful, Bāxδī (Bactria), the beautiful with upraised banners, and Nisāya, which lies between Mouru and Baxδī. Not much can be concluded from these brief mentions, however, other than that the Iranians knew and inhabited these areas in pre-Achaemenid times.

Achaemenid times. History really begins in this area with echoes of the conquests in Central Asia by Cyrus founder of the Achaemenid empire. Herodotus (1.205-14) tells us that Cyrus lost his life fighting against the Massagetai, presumably a group of the Sakas, in 530 b.c.e., and the existence of a town in the Fergana valley called Cyropolis or Cyreschata in Arrian (4.3.1) and in Curtius’ (7.6.16) Latin history of Alexander (medieval Kurkath) suggests that the conquests of Cyrus extended at least that far into Central Asia (Benveniste).

In his inscriptions Darius, and after him Xerxes, mentions the following northeastern parts of his empire (the order varies in the inscriptions): ParΘava (Parthia), Zra(n)ka (Drangiana), Haraiva (Herāt), Margu (Marv), Uvārazmiy (Choresmia), Bāxtriš (Bactria), Suguda (Sogdia), and two Saka peoples, the haumavarga and the tigraxauda; on the trilingual golden plates from Persepolis and Hamadān (DPh and DH, Kent, Old Persian, pp. 136-37, 147) he boasts that his empire reaches from the Sakas who are beyond Sogdia to Ethiopia and from Sind to Sardis. In the Bīsotūn inscription (DB 5.20-30, Kent, Old Persian, pp. 133-34) Darius says: “Afterwards with an army I went to the land of the Sakas after the Sakas who wear a pointed hat. These Sakas went from me. When I arrived at the sea, then I crossed beyond it with all my army. Afterwards I defeated the Sakas exceedingly. Another I took captive who was led bound to me and I slew him.” We may presume that the Central Asian provinces of the empire were not uniformly quiet down to the invasion of Alexander the Great, but we have no information about revolts or any events here during the rest of Achaemenid rule, so whether Darius actually fought against the Sakas in Central Asia as well as in south Russia, as suggested by the Bīsotūn inscription, is uncertain. According to the Greek sources (Herodotus, 3.92; Arrian, 3.8.3) Central Asian contingents and officers served in the Achaemenid armies.

It is uncertain how far Iranian tribes, presumably mostly the Sakas, extended into the steppes to the north of the oases of Central Asia, but archeological finds at such sites as Issyk Kurgan near modern Alma Ata (Akishev, 1978) and Pazyryk in Siberia (Rudenko) suggest that Iranians dominated the steppes as well as the settled regions to the south. From Pazyryk comes the oldest relatively well preserved carpet in the world (late 4th-early 3rd centuries b.c.e.) with Achaemenid motifs but probably manufactured locally (see carpets vi). In the Issyk kurgan (burial tomb) was a prince with a tall hat and armor of gold, presumably a Saka prince. The movement of various tribes or peoples on the steppes, however, can only be guessed, as we have no written sources. The time of the movement of the “Tokharians,” an Indo-European people speaking a language of the European, rather than Asian, type (Centum language) into Chinese Turkistan is much disputed; estimates range from the third millennium to the second century b.c.e. If the movement took place in the 8th century b.c.e., as has been suggested, then the “Tokharians” would have moved from west to east before the expansion of the Sakas into the steppes from the south as Herodotus tells us (for historical questions connected with the “Tocharians” see Ivanov and Winter).

With the invasion of Alexander and his campaigns in Central Asia in 329-27 b.c.e. the Alexander historians increase our knowledge of the area; some of the meager information in Ptolemy and Strabo comes from the reports of Alexander’s conquests in this part of the world. The resistance of Central Asian peoples to the conqueror was strong and stubborn, necessitating the establishment of garrisons in the conquered areas after heavy fighting. The center of Alexander’s control, and that of his successors, was Bactria, a rich and strategic region for trade routes from all directions (Tarn, pp. 118-121; Narain, pp. 12-13).

Excavations at the Bactrian site of Aĭ Khanom (Āy Ḵānom) on the Kokcha river in northern Afghanistan, as well as other Bactrian sites on the Oxus, have revealed the purely Greek character of the culture of these sites under the Seleucid successors of Alexander (Francfort, Bernard, et al., eds.; the publications of the Délégation archéologique française en Afghanistan; also Litvinskiy and Pichikan). Inscriptions of Greek poetry, as well as styles of architecture and town planning, were just as characteristic of Central Asia as of mainland Greece (Bernard, 1975, p. 457; idem, 1978, p. 453). Hellenic art influences, such as Corinthian columns and volutes in architecture, had a great influence on Central Asian art (Bernard in Francfort and Bernard, eds., I, 1973, pp. 211-13) just as the beautiful coinage of the Greco-Bactrian rulers influenced all later coinage in Central Asia, both by the use of the Attic weight system and by style of obverses and reverses of the money (Mitchener, I, pp. 10-18). Trade with China probably developed more under the Greco-Bactrians in the 3rd-2nd centuries b.c.e. than earlier, as suggested by the presence of Greek words in several languages of Chinese Turkistan (see china and iran i).

Central Asia and its eastern neighbors. It is probable that the Silk Route to China first came into prominence in the time of Greek rule in Bactria, and the three main routes from China to the west continued to be used. The first was the northern route from the present steppes of Kazakhstan through the Ili valley and Jungaria to Dunhuang and the Gansu corridor into central China. The second went from the Fergana valley to Kashgar and along the northern oasis route through Aqsu, Kucha, Turfan, and Komul (Hami) to Dunhuang. The third went from the Indus valley over the Karakorum range to Kashgar or Yarkand, then through Khotan, Cherchen, and Lop Nor to Dunhuang (Klimkeit, 1988, pp. 68-69; Haussig, pp. 6-7).

In the second half of the 2nd century b.c.e. Central Asia was the scene of nomadic invasions from the north and east (see, e.g., Bivar, 1966, pp. 51-52). The Sakas moved south into the land which bears their name, Sīstān (from Proto-Iranian *Sakastāna), and into India where small Saka (Śaka) kingdoms were created. Probably at this time they also established a kingdom in (see also aśoka iv). The Greco-Bactrian state collapsed under this invasion but other rulers maintained power in the Hindu Kush mountains and in India (cf. Narain, pp. 145-47). After a period of confusion one of the invading peoples, called Yue-zhi (Yueh-chih) in Chinese sources, established a state north of the Oxus River and then in the first century c.e. spread to the south under one of the tribes, which gave its name to the new Kushan empire (Gafurov, ed., esp., I, pp. 182-98, and II, pp. 42-46; Samolin, see index). North of Sogdiana it seems a confederation of settlements and nomadic tribes came into existence called Kang-qu (K’ang-chü) in Chinese sources (Litvinskij, 1972, 1976; Samolin, see index). It is uncertain how much control the Kushans exerted on the settled folk of Central Asia, but because of the scarcity of Kushan coins in Choresmia, as well as the striking of their own coins, we may infer that the Choresmians were independent or in a loose vassal relationship with the Kushans (see Vaĭnberg for a history of the land as well as a catalogue of coins). The Sogdians too probably were in a vassal relationship with the Kushans, who directed most of their efforts of expansion in the Indian subcontinent (Gafurov, ed., II, pp. 9-15).

Sasanian times. In the third century c.e. the Kushan empire seems to have split into two or more parts, and Central Asia was invaded from the Sasanian empire. The data and the extent of Sasanian expansion in the east, however, is uncertain, but probably both Bactrians and Sogdians acknowledged some sort of submission to the western power. The history of Ṭabarī tells us that under the first Sasanian ruler Ardašīr, the Kushans submitted to him (Nöldeke, Geschichte der Perser, pp. 4-6), and in the inscription of Šāpūr on the Kaʿba-ye Zardošt at Naqš-e Rostam it is said that the empire of Iran included the Kushan domains (Kušānšahr) as far as Kāš (Kashgar ?) and Sogdiana (ed. Back, pp. 288-89). In the inscriptions of Šāpūr I and his successors the following northeastern provinces of the Sasanian empire are listed: Marw, Harēw, Abaršahr, Kušān, Kāš, Sugd, Čāč, Xwārazm (inscriptions of Šāpūr on the Kaʿba-ye Zardošt at Naqš-e Rostam and of Narseh I at Paikuli, see, e.g., Maricq, pp. 336-37 [78-79], Humbach and Skjærvø, III/2, pp. 122-24). Later conquests by the Sasanians north of the Oxus are unknown, but trade and cultural influences were strong (for trade relations cf. Pigulevskaya, pp. 1-12). The Choresmians maintained their own dynasty, and relations with the Sasanians are unclear for early periods, but after the fourth century their land was independent, if not in some vassal relationship with the Hephthalites or later the Turks (Vaĭnberg, pp. 89-93; Muninova, pp. 40-50). To the north of the oasis states we have no information, but with the continuing expansion of nomads to the south we may presume the end of any confederation and rather the rule of local dynasts. This is confirmed by coins, the main source of information for the period before the Arab conquests (cf. Zeimal, pp. 233-51).

The end of the fourth century saw a change on the steppes of Central Asia with the expansion from the east of the Altaic-speaking peoples, under the pressure of which the Iranian nomads moved south and west, although it seems some were ruled by Altaic nomads or were absorbed by the newcomers (McGovern, pp. 399-419). Sogdiana and Bactria were invaded by Chionites, probably a nomadic group with Altaic rulers but Iranian common folk. It seems that the name Chionite, which appears in Byzantine sources, is related to the word Hun, and the name of the Huṇas, who invaded India in the middle of the fifth century (Moravscik, s.vv. Chionites, Huns). The Chionites were succeeded by the Hephthalites, again probably a mixed horde (Enoki, 1959, 1969). According to Enoki Chinese accounts of the Hephthalites indicate that they were primarily mountaineers from the lands to the west of the Pamirs, but others argue for their steppe origin (Bivar, 1983, pp. 213-15). In spite of these invasions and foreign rulers the local people of Central Asia developed their trading activities, with the Choresmians establishing extensive relations with the Volga river and eastern European areas, bringing furs, amber, beeswax, and other commodities back, while the Sogdians were the traders of the east, extending their activities and their trading colonies into China and Mongolia (Frye, 1972, pp. 266-68). The trade of the Sogdians was in cloth, precious stones, and spices from India in exchange for silk and various craft objects from China. The Bactrians, of course, were more concerned with trade to the south, but it seems that all of the Central Asians were primarily middlemen in trade in all directions, and the trade was primarily in luxury objects, as the risks of long distance trade in those times required large returns.

Trade, religion, culture. The extensive trading activities of the Central Asians coincided with the expansion of the universalist, missionary religions and the Central Asian were instrumental in spreading those religions. The earliest was Buddhism, which reached Central Asia under the Greco-Bactrians; later Bactria became a center of Buddhism under the Kushans (Klimkeit, 1986, pp. 8-10). Archeology has revealed numerous Buddhist remains in present northern Afghanistan and Tajikistan, and even though we may not have any Buddhist writings in the Bactrian (q.v.) language (the contents of the so-called Hephthalite fragments have not yet been identified for certain, see bactrian, p. 346b bottom), we may presume that they existed but have not survived. One fragment from Turfan in Manichean script but in the Bactrian language contains a Manichean text (Gershevitch, 1984; bactrian). Buddhism continued to flourish in Bactria into the Islamic period and only in the ninth century of our era did Balḵ, the largest city of Bactria, become Muslim in religion (Litvinsky, 1968, p. 121). Buddhism apparently made little progress in Sogdiana and Choresmia, and the greatest success was to the north and east of Central Asia (cf. Rapoport, pp. 119-21). The accounts of Chinese Buddhist travelers to the west such as Xuanzang (Hsüan Tsang) give us welcome details about the presence of Buddhism in various oases of Central Asia.

In Sogdiana a local form of Mazdeism was the dominant religion (cf. Henning), although Manicheism and Nestorian Christianity both had adherents, and we may suppose that it was primarily Sogdians who spread both religions to the east by missionaries in the caravans of merchants (cf. Barthold, 1901, pp. 20-26; Hage, p. 518). In Chinese Turkestan Sogdian became a lingua franca, and even the Turkish states in Mongolia and the Altai mountains used Sogdian as their written language, although later they were to develop their own system of writing of their languages using modified Sogdian script. The Sogdian city states of Bukhara, Samarkand, and Čāč (Tashkent) flourished in the period of Hephthalite domination, and this did not change with the Turkish conquest of the Hephthalites in the 560s of our era (Smirnova, 1970, esp. pp. 122-55). The ease with which the Sogdians replaced Hephthalite overlordship with Turkish reveals the attitude of the merchant society which sought stability so trade could flourish (for the expansion of the Turks see Barthold, 1945; for the Sogdian culture see Azarpay).

Archeological excavations of Sogdian sites have revealed the richness of local culture and far-flung relations of the Sogdian merchants, who probably introduced the system of training of slaves to protect their land and property while merchants were away, which later in the Islamic world came to be known as the Mamluk system of slave soldiers and administrators, also famous in the Ottoman and Mughal empires (cf. Pipes; Crone; also barda and bardadārǰ v).

The Choresmians also followed a local form of Mazdaism, although here, as with the Sogdians, universalist religions were active (Rapoport; Livshits). Archeology has revealed the same kind of culture and civilization as in Sogdiana but not as rich or ostentatious. Also the Choresmians were unified under one dynasty unlike the Sogdian city states (Tolstov and the many publications of the Southern Choresmian expedition). Their language (see choresmian) was originally written in a modified Aramaic alphabet, but later, perhaps in the 10th and 11th centuries, the Choresmians began to write their language in Arabic characters, which apparently was not the case with the Sogdians. Missionaries in Choresmian caravans spread Islam to the Volga region beginning in the 9th century, and the Volga Tatars maintained relations with Choresmia in later times (Togan, pp. 217-18, 253-56). One of the articles of trade with the north was silver, especially in the form of bowls, rhytons, and plates (Frye, 1971, pp. 255-62; idem, 1972, p. 266). The Choresmians played the same role in the west as the Sogdians did in the east in the spread of ideas and culture (Yagodin, pp. 128-68).

Although the Sasanians controlled Bactria and some areas to the north of the Oxus River in the late 3rd and 4th centuries through their Kushano-Sasanian governors, who at times were independent rulers, the nomadic invasions eliminated Sasanian rule in Central Asia, and only raids and temporary incursions were made by the Persians in later times (Carter). The raids by Bahrām IV (r. 388-99) left a mark on Bukhara, where the coinage of the Sasanian ruler was adopted as a prototype for the local coinage of Bukhara (Frye, 1949). Later the Sasanian general Bahrām Čōbīn defeated the Turks and obtained booty from his campaigns in 589 (Kolesnikov, pp. 51-53). Sasanian cultural influences, however, grew such that Central Asia in the period before the Islamic conquest was considered by Arnold to be a provincial Sasanian outpost in cultural matters (Arnold, pp. 10-11). Thanks to archeology this view has changed, and we know that the Sogdian city states and Choresmia were centers of their own cultures (cf. Belenitzki/Belenitskiĭ for Sogdiana, Tolstov for Choresmia).

For the rulers of Bukhara, Samarkand, and Choresmia, as primarily revealed by coins, see individual articles. As the coins of Bukhara, known as “Bukhar Khudat” coins, were copied from those of the Sasanian ruler Bahrām IV, we may suppose that he made conquests in Central Asia that did not survive him. Some of the coins of Samarkand, on the other hand, have square holes in the middle, like Chinese coins of the Tang dynasty (Smirnova, 1981).

The coming of Islam. Marv became an Arab center in the east, just as it had been for the Sasanians, and raids across the Oxus began in 673 and continued almost yearly after that (Gibb). The Muslim conquest of Central Asia began with Qoṭayba b. Moslem, who became Umayyad governor of Khorasan in 705, and he established Arab rule firmly in lands to the north of the Oxus. Several times the local rulers rebelled against Arab hegemony, and from one of them, called Dīvāstīč, lord of Panjikent, we have twenty letters in Sogdian telling, among other matters, of his attempts to organize resistance against the invaders, although to no avail, as he lost his life by Arab orders in 722 (Livshits, 1962, pp. 90-91). It is true that the Sogdian city states turned to their nominal rulers the western Turks for aid, but also to no avail. Neither their successors, the eastern Turks, nor the Turgesh later were able to dislodge the Muslims. Conversions to Islam turned Central Asia into a great Islamic cultural center, and gradually other religions died out. Central Asia and its cultures, however, played an important role in the development of an Islamic civilization including art, architecture, literature, and thought. By the 4th/10th century under the Samanids Central Asia was fully Islamicized (Negmatov, pp. 23-66). The Samanids were the last Iranian dynasty to rule in Central Asia with their center in Bukhara through the tenth century. Afterwards only Turkic dynasties ruled this part of the world.

See also art in iran vi; china and iran i.

Richard N. Frye

Bibliography

  • A. K. Akishev, Kurgan Issyk, Moscow, 1978.
  • Idem, Iskusstvo i mifologiya Sakov, Alma-Ata, 1984.
  • T. W. Arnold, Survivals of Sasanian and Manichaean Art in Persian Painting, Oxford, 1924.
  • G. Azarpay, Sogdian Painting, Berkeley, 1981.
  • M. Back, Die sassanidischen Staatsinschriften, Acta Iranica 18, Tehran and Liège, 1978.
  • W. Barthold, Zur Geschichte des Christentums in Mittel-Asien, Tübingen, 1901.
  • Idem, Histoire des Turcs d’Asie Centrale, Paris, 1945.
  • A. M. Belenitskiĭ (Belenitzki), Zhivopis’ drevnego Pendzhikenta, Moscow, 1954.
  • Idem, Mittelasien. Kunst der Sogden, Leipzig, 1980.
  • E. Benveniste, “La ville de Cyreschata,” JA, 1947, pp. 163-66.
  • P. Bernard in Comptes rendus de l’Académie des Inscriptions et Belles-lettres, April-June, 1975, pp. 167-97 and April-June 1978, pp. 421-63.
  • A. D. H. Bivar, “The History of Eastern Iran,” in Camb. Hist. Iran III, 1983, pp. 181-231.
  • Idem, Zentralasien, in G. Hambly, ed., Fischer Weltgeschichte, vol. 16, Frankfurt am Main, 1966.
  • G. M. Bongard-Levin and E. A. Grantoskij, De la Scythie à l’Inde, Paris, 1981.
  • M. L. Carter, “A Numismatic Reconstruction of Kushano-Sasanian History,” American Numismatic Society, Museum Notes 30, 1985, pp. 215-81.
  • A. Christensen, Le premier chapitre du Vendidad et l’histoire primitive des tribus iraniennes, Det Kgl. Danske Videnskabernes Selskab, hist.-filol. medd. 29/4, Copenhagen, 1943.
  • P. Crone, Slaves on Horseback, Cambridge, 1981.
  • I. M. Dyakonov, “Die Arier im Vorderen Orient: Ende eines Mythos,” Orientalia 41, 1972, pp. 91-120.
  • I. M. Dyakonov and V. A. Livshits, Parthian Economic Documents from Nisa, Corpus Inscr. Iran. III/2, Moscow, 1976.
  • K. Enoki, “On the date of the Kidarites,” Memoirs of the Research Department of the Toyo Bunko 27, 1969, pp. 1-26.
  • Idem, “On the Nationality of the Ephthalites,” ibid., 18, 1959, pp. 1-58.
  • H. P. Francfort, P. Bernard, et al., eds., Fouilles d’Aï Khanoum, vols. 1-7, Paris, 1973-87.
  • R. N. Frye, Notes on the Early Coinage of Transoxiana, American Numismatic Society, New York, 1949.
  • Idem, “The Significance of Greek and Kushan Archaeology in the History of Central Asia,” Journal of Asian History 1, 1967, p. 35.
  • Idem, “Sasanian Silver and History,” in E. Bosworth, ed., Iran and Islam in Memoriam V. Minorsky, Edinburgh, 1971, pp. 255-62.
  • Idem, “Byzantine and Sasanian Trade Relations with Northeastern Russia,” Dumbarton Oaks Papers 26, 1972, pp. 265-69.
  • Idem, The History of Ancient Iran, chaps. 3, 7, 9 and 12. A. von Gabain, “Irano-Turkish Relations in the Late Sasanian Period,” in Camb. Hist. Iran III, 1983, pp. 613-24.
  • B. G. Gafurov, ed., Tsentral’naya Aziya v kushanskuyu èpokhu, 2 vols., Moscow, 1974-75.
  • I. Gershevitch, The Avestan Hymn to Mithra, Cambridge, 1967.
  • Idem, “The Bactrian Fragment in Manichean Script,” in Harmatta, ed., 1984, pp. 273-80.
  • R. Ghirshman, L’Iran et la migration des Indo-Aryens et des Iraniens, Leiden, 1977.
  • b H. A. R. Gibb, The Arab Conquests in Central Asia, London, 1923.
  • W. Hage, “Das Nebeneinander christlicher Konfessionen im mittelalterlichen Zentralasien,” in ZDMG, Supp. 2, Wiesbaden, 1969, pp. 517-25.
  • J. Harmatta, ed., Prolegomena to the Sources on the History of Pre-Islamic Central Asia, Budapest, 1979.
  • Idem, ed., Studies in the Sources on the History of Pre-Islamic Central Asia, Budapest, 1979.
  • Idem, ed., From Hecataeus to al-Huwarizmi, Budapest, 1984.
  • H. W. Haussig, Die Geschichte Zentralasiens and der Seidenstrasse in vorislamischer Zeit, Grundzüge 49, Darmstadt, 1983.
  • W. B. Hennig, “A Sogdian God,” BSOAS 28, 1965, pp. 242-54.
  • H. Humbach and P. O. Skjærvø, The Sassanian Inscription of Paikuli, 3 vols. in 4, Wiesbaden, 1978-83.
  • V. V. Ivanov, “Yazykovye dannye o proiskhozhdenii kushanskoĭ dinastii i tokharskaya problema,” Narody Azii i Afriki 3, 1967, pp. 106-19.
  • H.-J. Klimkeit, Die Begegnung von Christentum, Gnosis and Buddhismus an der Seidenstrasse, Opladen, 1986. Idem, Die Seidenstrasse, Cologne, 1988.
  • P. L. Kohl, ed., The Bronze Age Civilization of Central Asia, Armonk, N.Y., 1981.
  • A. I. Kolesnikov, “Iran v nachale VII veka,” Palestinskiĭ sbornik 22, 1970, pp. 1-110.
  • A. Koshelenko, ed., Drevneĭshie gosudarstva Kavkaza i Sredneĭ Azii, Moscow, 1985.
  • B. A. Litvinsky, Outline History of Buddhism in Central Asia, Moscow, 1968.
  • B. A. Litvinskiĭ, Drevnie kochevniki “Kryshi mira,” Moscow, 1972.
  • B. A. Litvinskij, “Das K’ang-chu-Sarmatische Farn,” Central Asian Journal 16, 1972, pp. 242-89, 20, 1976, pp. 47-74.
  • B. A. Litvinskiy and I. R. Pichikan, “Monuments of Art from the Sanctuary of Oxus,” in Harmatta, ed., 1984, pp. 25-84.
  • V. A. Livshits, Sogdiĭskie dokumenty s gory Mug, Moscow, 1962.
  • Idem, “The Khwarezmian Calendar and the Eras of Ancient Chorasmia,” Acta Antiqua Hungaricae 16, 1968, pp. 433-46.
  • A. Maricq, “Res Gestae Divi Saporis,” Syria 35, 1958, pp. 295-360; repr. in Classica et Orientalia, Paris, 1965, pp. 37-101.
  • V. M. Masson, Das Land der tausend Städte. Baktrien, Choresmien, Margiane, Parthien, Sogdien. Ausgrabungen in der südlichen Sowjetunion, Wiesbaden and Berlin, 1987 (tr. by M. T. E. Seitz of Strana tysyachi gorodov, Moscow, 1966, repr. 1981).
  • M. Mayrhofer, Die Arier im Vorderen Orient—Ein Mythos?, Vienna, 1974.
  • W. M. McGovern, The Early Empires of Central Asia, Chapel Hill, North Carolina, 1939.
  • M. Mitchener, Indo-Greek and Indo-Scythian Coinage, 9 vols., London, 1975-76.
  • G. Moravscik, Byzantino-Turcica, Berlin, 1958.
  • I. M. Muninova, Istoriya Khoresma, Tashkent, 1976.
  • A. K. Narain, The Indo-Greeks, Oxford, 1957.
  • N. N. Negmatov, Gosudarstvo Samanidov, Dushanbe, 1977.
  • A. P. Okladnikov, “Neandertal’skiĭ chelovek i sledy ego kul’tury v Sredneĭ Azii,” Sovetskaya arkheologiya 6, 1940, pp. 5-19.
  • D. Pipes, Slave Soldiers and Islam, New Haven, 1981.
  • N. V. Pigulevskaya, Vizantiya i Iran na rubezhe VI-VII vekov, Moscow, 1946.
  • B. B. Petrovskiĭ and G. M. Bongard-Levin, eds., Tsentral’naya Aziya. Novye pamyatniki pis’mennosti i iskusstva, Moscow, 1987 (numerous contributions on art, archeology, and texts; note E. V. Zeĭmal’, “K periodizatsii drevneĭ istorii Sredneĭ Azii”).
  • I. V. Pyankov, Srednyaya Aziya v izvestiyakh antichnogo istorika Ktesiya, Dushanbe, 1975.
  • Yu. A. Rapoport, Iz istorii religii drevnego Khorezma, Moscow, 1971.
  • I. Ronca, Ptolemaios. Geographie 6, 9-21. Ostiran and Zentralasien, Rome, 1971.
  • S. I. Rudenko, Kul’tura naseleniya tsentral’nogo Altaya v skifskoe vremya, Moscow, 1960.
  • W. Samolin, East Turkistan to the Twelfth Century. A Brief Political Survey, Central Asiatic Studies 9, London, etc., 1964.
  • N. Sims-Williams, Inscriptions in Sogdian and Other Iranian Languages from the Upper Indus, Corpus Inscr. Iran. III/2, London, 1989.
  • O. I. Smirnova, Ocherki iz istorii Sogda, Moscow, 1970. Idem, Katalog monet s gorodishcha Pendzhikent, Moscow, 1973.
  • Idem, Svodnyĭ katalog sogdiĭskikh monet. Bronza, Moscow, 1981.
  • V. Y. Staviskiĭ, Kara Tepe, 4 vols. 1-4, Moscow, 1964-75.
  • W. W. Tarn, The Greeks in Bactria and India, Cambridge, 1951.
  • A. Z. V. Togan, Ibn Fadlan’s Reisebericht, Leipzig, 1939.
  • S. P. Tolstov, Drevniĭ Khorezm, Moscow, 1948.
  • Idem, Po sledam drevne-khorezmiĭskoĭ tsivilizatsii, Moscow and Leningrad, 1948; Germ. tr. O. Mehlitz, Auf den Spuren der altchoresmischen Kultur, Berlin, 1953.
  • B. I. Vaĭnberg, Monety drevnego Khorezma, Moscow, 1977.
  • W. Winter, “Tocharians and Turks,” in D. Sinor, ed., Aspects of Altaic Civilization, Indiana University Publications, Ural and Altaic Series 23, pp. 239-51.
  • V. N. Yagodin, Nekropol’ drevnego Mizdakhkana, Tashkent, 1970.
  • E. V. Zeimal, “The Political History of Transoxiana,” in Camb. Hist. Iran III, 1983, pp. 232-62.

CENTRAL ASIA iv. In the Islamic Period up to the Mongols

In early Islamic times Persians tended to identify all the lands to the northeast of Khorasan and lying beyond the Oxus with the region of Turan, which in the Šāh-nāma of Ferdowsī is regarded as the land allotted to Ferēdūn’s son Tūr. The denizens of Tūrān were held to include the Turks, in the first four centuries of Islam essentially those nomadizing beyond the Jaxartes, and behind them the Chinese (see Kowalski; Minorsky, “Tūrān”). Tūrān thus became both an ethnic and a geographical term, but always containing ambiguities and contradictions, arising from the fact that all through Islamic times the lands immediately beyond the Oxus and along its lower reaches were the homes not of Turks but of Iranian peoples, such as the Sogdians and Khwarezmians. Equally imprecise was the Arabic designation Mā Warāʾ al-Nahr “the land beyond the river” (i.e., Amu Darya, the Oxus), which passed also into Persian literary usage and was used until post-Mongol times, e.g., by Ḥāfeẓ-e Abrū and by Bābor. At the outset, however, those nearby parts of Central Asia with which the Arabs were familiar were often subsumed into the vast and ill-defined province of Khorasan, embracing all lands to the east of Ray, Jebāl, and Fārs.

On the eve of the first Arab incursions across the Oxus in the second half of the 1st/7th century, the ethnically, linguistically, and culturally Iranian lands of Ḵᵛārazm and Transoxania were still thriving entities linked with the Eurasian steppes, which ran from eastern Europe to the borders of China, by a nexus of commercial routes, benefiting from the religious and intellectual stimuli of both the Iranian and Indian worlds (see also buddhism; choresmia; sogdia). Arab raiders penetrated north of the Oxus in the caliphate of ʿOṯmān and in the governorship over Khorasan of ʿAbd-Allāh b. ʿĀmer b. Korayz (q.v.), but Yazīd b. Moʿāwīa’s governor Salm b. Zīād (61-64/681-83) was the first Arab commander actually to winter across the Oxus. Disturbances in the heartland of the caliphate meant that it was not till the time of Qotayba b. Moslem Bāhelī (86-96/705-15; q.v.) that a firm hold was secured over Transoxania and the upper Oxus provinces, together with the first Arab attack on Ḵᵛārazm in 93/712 (Gibb, pp. 42-43; also choresmia). But only after 133/751 was Arab rule in Transoxania finally free from challenge. In that year, at the battle of Talas (Ṭarāz), the ʿAbbasid forces under Zīād b. Ṣāleḥ defeated the Chinese general Kao-hsien-chih, for the Ṭʿang emperors claimed suzerainty over Central Asia and had responded to appeals made to Peking by the threatened Sogdian princes (Ebn al-Aṯīr, IV, p. 449; Barthold, Turkestan, pp. 195-96). The Chinese threat was thus averted, as had been, shortly before this, the threat of the Western Turks of Türgeš, who as a steppe confederation had been liable to press on the frontiers of Transoxania at times of political weakness and instability there and who had been called in, like the Chinese, by local rulers (Barthold, Turkestan, pp. 186-87; Gibb, pp. 59-87; Grousset, pp. 165-72). Yet whereas the Chinese retreated permanently back behind the Ṭʿien-Shan Mountains, Turkish pressure was only momentarily dispelled and was later to be exerted by the individual tribes who succeeded to the heritage of the Western Turkish empire in the steppes; certain of these tribes such as the Qarluq (Ḵarloq), the probable progenitors of the Islamic Qarakhanids and the Oghuz, from whom the Saljuqs sprang, were later to have decisive and lasting effects on the historical and demographic evolution of Transoxania.

The ʿAbbasids integrated Transoxania in their empire as a province, at first appointing over it a series of ephemeral governors, who had to cope with several movements of social and religious protest, some Islamic in nature, such as those of the , others distinctly heterodox, such as the uprising of Moqannaʿ and his “Wearers of white,” the Mobayyeża or Sapīdjāmagān (q.v.). The integration persisted, helped by the gradual rallying of the landowning or dehqān classes in the eastern Iranian lands to Islam and the Islamic ruling order: The members of the abnāʾ al-dawla, supporters of the ʿAbbasid revolution, came from Transoxania, and in the 3rd/9th century Transoxanian Iranian elements in the caliphal armies were perpetuated through local princes like the Afšīn Ḵayḏār of Osrūšana and contingents from specific areas like the “men of Farḡāna” and the men of Osrūšana” (Faraḡena and Osrūšanīya in the sources); these were undoubtedly free Iranians rather than Turkish slave guards (ḡelmān, mamālīk; see Ayalon, pp. 29-32).

The Taherid governors of Khorasan in the 3rd/9th century deputed their authority over Transoxania to an Iranian dehqān family from Ṭoḵārestān in the upper Oxus valley, the Samanids, who after the capture of the Taherid capital Nīšāpūr by the Saffarids in 261/875, were recognized by the ʿAbbasids as their official representatives in Transoxania and Khorasan (R. N. Frye, in Camb. Hist. Iran IV, pp. 137-38). The increasing enfeeblement of the caliphs in Sāmarrā and Baghdad meant that the Samanids, while always proclaiming their allegiance to the caliphate and to Sunni Islam, could behave as virtually independent rulers. The Arab historians and geographers praise the beneficence of Samanid rule during the later 3rd/9th and 4th/10th centuries: low taxation, helped by the fact that Samanid amirs could raise money from import and transit dues levied on traffic in the products of inner Asia, above all, on Turkish military slaves (see barda and bardadārī v. military slavery); cheap and plentiful provisions from the rich irrigated lands and oases of the Oxus, Zarafšān, and Jaxartes (Syr Darya) valleys; and the generally enlightened rule of the Samanids themselves, which involved respect for scholars and litterateurs and, at least in the years until the decay of the amirate, regularly paid salaries for officials and the troops (see Barthold, Turkestan, pp. 234-40). Socio-religious protest, while not disappearing completely (Ismaʿili propagandists dispatched by the Fatimids seem to have secured a foothold in Transoxania during the reign of Naṣr b. Aḥmad, 301-31/914-43; see ibid., pp. 242-44), was much diminished. The frontiers of Transoxania and Ḵᵛārazm were maintained by strong defense lines of rebāṭs or frontier fortresses, from which punitive raids could be launched into the steppes when the nomads proved recalcitrant; alliances were made with some Turkish tribes on the frontiers whereby they received subsidies in exchange for acting as frontier guards; and, although little is known about this, a certain amount of evangelism in the pagan steppes may have been undertaken by individual Sufi shaikhs and similar enthusiasts for the faith, such as the Shaikh from Nīšāpūr Abu’l-Ḥasan Moḥammad Kalemātī (d. some time before 350/961), who worked among the Qarluq and may have played a part in the conversion of the Qarakhanid prince Satuq Boḡra Khan (see ibid., pp. 254-56; and Grenard). Thus the northeastern bastions of Islamic faith and civilization, maintenance of which had always been a prime concern of governors and rulers in the East, held firm during these early centuries, and from the second half of the 4th/10th century onward Islamic religion began to influence, even if only superficially, the animistic and shamanistic beliefs of Turkish and other peoples of the steppes like the Oghuz and Qarluq and, further west, the Khazars and Bulghars. Although the complete Islamization of Central Asia was hardly achieved even by the Timurid period, Central Asian Islam began to evolve its own special nature and emphases, seen, e.g., in the Sufi order founded in Transoxania in the later 6th/12th century by the followers of the Turkish holy man Shaikh Aḥmad Yasavī (d. at Yasī, the later town in Turkestan, in the middle Jaxartes valley in 562/1166; see F. İz, “Aḥmad Yasawī,” in EI 2, pp. 298-99).

However, the collapse of the Samanid amirate as a result of internal tensions and financial crisis at the end of the 4th/10th century meant a distinct weakening of the defenses against pressure from the outer steppes. The Qarakhanids or Ilek-khans, only recent converts to Islam, appeared in the Jaxartes valley, temporarily occupying the Samanid capital Bukhara as early as 382/992, and in the early decades of the next century took over Transoxania. The Samanid lands south of the Oxus fell to the Ghaznavids, a Turkish dynasty of military slave origin. Iranian rule in Transoxania came to an end with the fall of the Samanids and in the neighboring kingdom of Ḵᵛārazm in 408/1017, when the Ghaznavids destroyed the Iranian line of

Maʾmunid Ḵᵛārazmšāhs (see āl-e maʾmūn). In the middle decades of the 5th/11th century Turkish power in these regions was strengthened through the establishment of the Great Saljuq empire in Iran and the central Arab lands from Iraq to Syria (see Bosworth, in Camb. Hist. Iran V, chap. 1). For varying periods, under such rulers as Alp Arslān, Malekšāh, and Sanjar, the Great Saljuqs exercised suzerainty over the Qarakhanids in Transoxania, and in Ḵᵛārazm a line of Turkish Ḵᵛārazmšāhs came to power under Qoṭb-al-Dīn Moḥammad, son of Anūštigīn, a slave of the Saljuq Malek Shah.

These political events had profound consequences for Transoxania and Ḵᵛārazm. The incoming of steppe nomads with their herds, first of Turkish tribesmen and then, in the 7th/13th century, of the Mongols, was bound to have long-term economic and demographic effects. A certain degree of pastoralization may have begun under the Qarakhanids, as there is mention of the setting-up of royal hunting grounds (ḡūroqs) by Šams-al-Molk Naṣr b. Ebrāhīm Ṭamḡāč (Ṭamḡāj) Khan (460-72/1068-80; see Naršaḵī, p. 35, tr. Frye, p. 29, cf. p. 125). Since the Qarakhanids were a tribal confederation and never formed a centralized state they had several centers of power, from Khotan to Samarqand, but these were only semipermanent (see Pritsak, pp. 23, 37). A continuator of Naršaḵī (p. 39, tr. p. 33) states that taxes were everywhere lightened when the Qarakhanids replaced the Samanids, and it is possible that the indigenous Iranian landed classes, the dehqāns, enjoyed a temporary resurgence of power. Nevertheless, the long-term trends of the 5th/11th and 6th/12th centuries militated against the preservation of the Iranian character of Transoxania and Ḵᵛārazm. Turkish elements continued to be attracted into these lands from the steppes, with the ultimate effect of the disappearance of the Sogdian and Choresmian languages and the confining of Iranian speech to the mountainous refuge-areas of the upper Oxus, what is now the Tajikistan SSR and the Pamir region (cf. xiii, below).

On the other hand, the strength of Islamic culture and religion exerted a pull in the reverse direction. Once converted to Islam, dynasties like the Qarakhanids and Saljuqs came to share fully in the Islamic heritage, which had always been strong in Khorasan and Transoxania. Persian poets flourished at the courts of the Ilek-khans, and Neẓāmī ʿArūżī cites thirteen poets who glorified the Āl-e Ḵāqān, as he calls it, among whom ʿAmʿaq of Bukhara was the eulogist of Šams-al-Molk Naṣr and his successor Ḵeżr Khan b. Ebrāhīm (472-73/1080-81; Čahār maqāla (ed. Qazvīnī, text, pp. 44-45; cf. Browne, Lit. Hist. of Persia II, pp. 335-36); but it was also among the Qarakhanids that the first Islamic Turkish imaginative literature appears, with Yūsof Ḵāṣṣ Ḥājeb’s Qutadḡu bilig, completed at the local court of Kashghar in 462/1069-70 (see Bombaci, pp. 83-96). In the sphere of toponymy, increased Turcisization in Central Asia is reflected in the appearance—at a point which cannot be precisely documented—of the term Turkestan for the Turkish lands of Central Asia comprising the former Transoxania and Ḵᵛārazm, while after the Mongol invasions that of Moḡolestān appears, more specifically for the steppes to the north of the Oxus-Jaxartes basins and Turkestan proper (Mīrzā Ḥaydar Doḡlāt, introd. pp. 51ff., tr. pp. 36-37; and Bosworth, “Mogholistān,” in EI 2).

The Mongol invasions of Transoxania were not a cataclysm, in that this appearance of non-Turkish, non-Islamic peoples from remote Inner Asia in Transoxania had been prefigured by the arrival there, some eighty years before Čengīz Khan’s time, of the Qara Khitay (Ḵeṭāy), probably also of Mongol but conceivably of Tungusic stock. In 536/1141 the Qarakhanid Maḥmūd Khan b. Arslan of Samarqand and his suzerain, the Saljuq Sultan Sanjar, were defeated by these incomers at one of the great battles of Central Asia, that of the Qaṭvān Steppe in Ošrūsana to the south of the middle Jaxartes (Rāvandī, pp. 171ff.; Ebn al-Aṯīr, XI, pp. 81-86; Barthold, Turkestan, pp. 326-27). (The news of this event filtered through dimly to the Christian West and gave an impetus to the legend of Prester John, the powerful anti-Islamic monarch who supposedly ruled in Inner Asia). Since the Qara Khitay stemmed from the people of northern China called in Chinese annals the Liao (Wittvogel and Fêng), they were partly sinicized, and the decentralized rule which they established in the eastern parts of Transoxania during the later decades of the 6th/12th century had a distinct Chinese imprint (e.g. in regard to the copper coinage of the Qara Khitay Gür Khans; ibid., pp. 661-62, 664, 672-73), bringing yet another element into what was becoming the ethnic, religious and cultural melting pot in Central Asia.

For specific details of the course of events of Central Asian history, insofar as it impinges on Iran and Iranian culture, see also individual place names and dynasties; cf. ARAB ii; BUKHARA.

C. Edmund Bosworth

Bibliography

  • D. Ayalon, The Military Reforms of Caliph al-Muʿtaṣim. Their Background and Consequences, commun. to the International Congress of Orientalists, New Delhi, 1964; offset Jerusalem, 1963.
  • A. Bombaci, Storia della letteratura turca dall’antico impero di Mongolia all’odierna Turchia, Milan, n.d. [1956].
  • Barthold, Turkestan 3.
  • Idem, Zwölf Vorlesungen über die Geschichte der Turken Mittelasiens, Berlin, 1935; French tr. Histoire des Turcs d’Asie Centrale, Paris, 1945.
  • Idem, Four Studies on the History of Central Asia, Leiden, 1962.
  • C. E. Bosworth, “Mā warāʾ al-nahr,” in EI 2. Camb. Hist. Iran IV, chaps, 1-5; V, chap. 1. Mīrzā Ḥaydar Doḡlāt, Tārīḵ-e rašīdī, tr. N. Elias and E. D. Ross, A History of the Moghuls of Central Asia, London, 1895.
  • H. A. R. Gibb, The Arab Conquests in Central Asia, London, 1923.
  • F. Grenard, “La légende de Satok Boghra Khân et l’histoire,” JA, sér. 8, no. 15, 1900, pp. 5-79.
  • R. Grousset, L’empire des steppes, 4th ed., Paris, 1952.
  • G. Hambly, ed., Zentralasien, Fischer Weltgeschichte, Frankfurt, 1966, Eng. tr. Central Asia, London, 1969, chap. 4.
  • T. Kowalski, “Les Turcs dans le Šāh-nāme,” Rocznik orientalistyczny 15, 1939-49, pp. 84-99.
  • Le Strange, Lands, pp. 433-89. Markwart, Ērānšahr, pp. 47-93, 199-304.
  • V. Minorsky, “Tūrān,” in EI 1. O. Pritsak, “Die Karachaniden,” Der Islam 31, 1953-54, pp. 17-68.
  • Moḥammad b. ʿAlī Rāvandī, Rāḥat al-ṣodūr wa āyat al-sorūr, ed. M. Eqbāl (Iqbal), London, 1921.
  • F. H. Skrine and E. D. Ross, The Heart of Asia. A History of Russian Turkestan and the Central Asian Khanates from the Earliest Times, London, 1899, pp. 34-148.
  • K. A. Wittvogel and Fêng Chia-Shêng, History of Chinese Society. Liao (907-1125), Transactions of the American Philosophical Society, N.S. 36, Philadelphia, 1946.

CENTRAL ASIA v. In the Mongol and Timurid Periods

At the death of Čengīz (Chinggis) Khan in 624/1227 the territory he had conquered was divided between his sons. To Čaḡatai (d. 642/1244-45) was allotted the region of Transoxania, from Bukhara and Samarkand in the west north to the Chu river and Lake Balkhash and as far east as the land of the Yellow Uighurs (approximately equivalent to present-day Sinkiang/Xinjiang), and Čaḡatai’s successors ruled in the eastern part of Central Asia almost without interruption until 1089/1678 (see chaghatayid dynasty).

Period of the Great Khans. In the middle of the 7th/13th century Čaḡatai’s territorial holdings (ulus) included the area around Lake Issyk-Kul (Īseq-Kūl) and the region of Yetisu (“seven rivers,” approximately coterminous with the modern Alma-Ata district in southeastern Kazakhstan) and, to the east and northeast of Lake Balkhash, adjoined the land belonging to his brother Ögedei (Ögödei; Ūkadāy, Ūkatāy), who was Great Khan from 626/1229 to 639/1241. Ögedei’s ulus included the Tarbagatay mountains and extended north to the valley of the Kara Irtysh river and the Altai mountains. The two domains were not sharply delineated; they were economically interdependent and inhabited by Turks who spoke the same language. The brothers’ residences were located in close proximity: in winter on the lower Ili south of Lake Balkhash, in summer farther upriver near Kuldja (I-ning) or east of the lake in Īmīl (Emel; Chuguchak, T’a-ch’eng/Pin-yin: Ta-cheng) on the upper course of the river of the same name. Each brother received allowances from the tax revenues and had at his disposal an army of 4,000 men, and both permitted various lesser dynasties to rule as vassals; the Il-khanids (654-736/1256-1336) did the same in their own territory.

Čaḡatai was recognized throughout the Mongol empire as the guardian of the old customs and the code of law (yasa, yāsā), formulated by Čengīz Khan, and gained a reputation for active hostility toward Islam. The governor appointed by Ögödei over the urban settlements in Transoxania and the Turfan oasis, the Ḵᵛārazmian Maḥmūd Yalavač (Yalavāj, d. 652/1254 in China), thus became the defender of his fellow Muslims against the hostility of the nomadic Mongols; he was supported by the Uighur Čenqei (Jīnqāy; executed in 649/1251). Maḥmūd’s son Masʿūd took over this role from about 637/1240 until his death in 688/1289, when he was succeeded by three of his sons. During this period nothing was done to prevent Muslims, Nestorian Christians, and Buddhists from practicing their religions, as the Mongols, who were mostly nomadic shamanists, must have found it advantageous to watch over their economic interests. Gradually the region recovered from the particularly severe devastation that it had suffered during the Mongol conquests in 615-18/1219-21, as well as during subsequent internecine conflicts, yet it never regained its former position as a center of Islamic learning and economic prosperity. It was linked to the great khan’s newly established capital at Karakorum (Qarāqūrom), near the headwaters of the Orhon river, by means of the Mongol postal system (yam/yām).

Most of the princes of the branches of Čaḡatai and Ögedei did not take part in the election in 649/1251 of the great khan Möngke (Mūngkā; Mengü/Mangū; 649-57/1251-59), a grandson of Čengīz Khan by his youngest son, Tolui (Tūlī), and suffered death or exile for their opposition. Much of their territory was confiscated and passed into the hands of Möngke or of his supporters from the branch of Čengīz’s eldest son, Joči (Jūjī), who reigned over the so-called Golden Horde in western Asia. The accession to the great khanate in 658/1260 of Möngke’s brother Qubilai (Qūbīlāy; d. 693/1294), who ruled from Ta-to (Daidu) near modern Peking/Beijing, unleashed prolonged disputes, first between Qubilai and his brother Arïḡ Böke (Arīq Būkā) and later between Qubilai and his successor Temür (Teymūr), on one hand, and Ögedei’s grandson Qaidu (Qāydū; d. 702/1303), on the other.

The Chaghatayids entered into an alliance with Qaidu, who was recognized as the rightful great khan in Central Asia in or soon after 667/1269, and took part in his campaigns against Qubilai and his clients in the east. These included the Uighur rulers, who were forced to abandon their territories of Beshbalyk (Bīšbālīḡ; north of the Tien Shan mountains near present-day Urumchi), Turfan, and Kucha (Kūčā, Kūjā), which had passed into the hands of the Chaghatayid khan Du’a (Dūʾā) by about 689/1290. Čaḡatai’s descendants also engaged in lengthy conflict along the northern border of their territory with the White Horde, the eastern branch of the Golden Horde, ruled by the line of Joči’s son Orda (Ūrda). In conjunction with Qaidu they crossed the Oxus and established a foothold in northern Afghanistan, where they fought against the Il-khans and from where they further launched repeated incursions into northern India. Following Qaidu’s death in 702/1303 the descendants of Čaḡatai and Tolui reached an agreement, which was then briefly extended to become the Pax Mongolica announced to western European monarchs by the II-khan in 704/1305 (Mostaert and Cleaves, cited in Camb. Hist. Iran V, p. 399).

In comparison with Persia and the Near East, the Chaghatayid khanate remained quite backward, both commercially and agriculturally. Sunnite Islam gradually spread into this territory from the west, progressing as far as the Tarim basin. Christianity declined, Buddhism retreated toward the east, although at the same time it was enjoying some success in Persia under the Il-khans (see buddhism ii. in islamic times).

The advance of Islam in the Chaghatayid khanate. The struggle between the lines of Qubilai and Ögedei came to an end with the defeat of Qaidu’s son Čapar (Čāpār) in 708/1309; the ulus of Ögedei then passed to the descendants of Čaḡatai. At that time the Mongol empire in Central Asia consisted of two distinct regions, which continued to evolve along different lines. In the west, in the original ulus of Čaḡatai, Sunnite Islam gradually prevailed. In 726/1326 the Chaghatayid khan ʿAlāʾ-al-Dīn Tarmaširin (Tarmaširīn) converted, though he was overthrown only eight years later (cf. Haidar) by opponents in the east. The Mongols’ acceptance of Islam helped to further their integration with the Muslim Turkish peoples, who were predominant among the population. Doubtless the remnants of the Iranian languages in the area disappeared at about the same time. Under the influence of Islam the rulers were encouraged to give greater consideration to the interests of the cities; for much of this period their capital was in Transoxania, at Naḵšab (Qaršī/Karshi, southeast of Bukhara), where they created a new administrative organization, based on small individual units, and minted a new type of coinage. Nevertheless, there was continuing opposition between settled inhabitants and nomads (which persisted into the 14th/20th century), and religious leaders, officials, and the army also sought to further their own interests. At the same time the power of certain Turkish clans, the Barlās, Arlat, and Süldüs (Sūldūs), increased substantially, and they were even able to extend their influence over large portions of Afghanistan, to the detriment of the Kartid rulers there (see āl-e kart).

In the eastern part of the Chaghatayid khanate, around Lake Issyk-Kul, in the former ulus of Ögedei, Islam met with bitter opposition and remained a minority religion in the early decades of the 8th/14th century. The struggle to establish it continued for some time longer and led to conflict and considerable destruction in the Chu and Talas (Ṭarāz) valleys, which became visibly depopulated. Nevertheless, in 739-40/1339 the increasing power of the Muslim faith led to the dissolution of the Roman Catholic missionary center at Almalyk (Almalïḡ, Almālīḡ). The Nestorian Christians, who had been represented in Central Asia for hundreds of years, died out completely in the 8th/14th century. Buddhism, which had also played an important role in the region, declined as well. The yasa was gradually replaced in importance by the Šarīʿa (Islamic law). In 747-48/1347 a (so-called) prince gained recognition as khan and converted to Islam, along with a large portion of the population of this region. A powerful eastern monarchy now confronted the ruling clans in the west. The contrast was so marked that the eastern area was given the name Moḡolestān, which remained in use for some time. There, too, the spread of Islam was followed by a Turkicization of the general population, so that the Turkish-language area continued to extend its eastern limits. The Turfan oasis was converted to Islam, and even in western China Muslim groups appeared (e.g., the Dungans), which still survive today. Only the definitive conversion of the Mongols in Mongolia to Buddhism toward the end of the 10th/16th century brought the eastward expansion of Islam to a halt.

Tīmūr and his successors. In 1360 the new khan of Moḡolestān, the Muslim Tuḡluq Temür (Tūḡlūq Tīmūr; 760-71/1359-70), succeeded in taking Transoxania, thus reuniting the previously divided Chaghatayid khanate under his rule, though the structural differences between the two regions persisted for some time. He made his son Elyās Ḵᵛāja governor in Transoxania and appointed Tīmūr, a young amir of the Barlās tribe, as the young man’s aide, without suspecting that this action marked the beginning of a new era in the region.

Tīmūr (b. 736/1336 near Kaš, now Shakhrisabz/Šahr-e Sabz south of Samarkand) extended his power as far as Čāč (modern Tashkent) and Balḵ, at first in alliance with Amīr Ḥosayn, one of the powerful Turkish princes, but he allowed the latter to be assassinated at Balḵ in 771/1370. Between 773/1372 and 790/1388 Tīmūr then conquered Ḵᵛārazm, which was divided into two realms at that time. He was thus in control of all Transoxania; throughout his life a large proportion of his troops came from Chaghatayid territory and from the Barlās tribe. Although Tīmūr reinstated the yasa, the influence of Islam nonetheless continued to increase, and he himself was a lifelong adherent of the Sunni branch. Tīmūr appointed to the nominal position of Chaghatayid khan princes of the line of Ögedei, whereas he himself bore only the title beg, or amīr, and after 790/1388 solṭān.

From this territorial base the great conqueror extended his power far to the west in the early 9th/15th century, up to the borders of Egypt and into Asia Minor and eastern Europe. On the other hand, despite all his efforts and repeated advances into the Tarim basin (especially in 801-2/1399-1400) and as far as Lake Issyk-Kul, he was unable to conquer Moḡolestān (cf. Manz). The political and to some extent the cultural and structural differences between eastern and western Central Asia thus persisted. In the west Tīmūr crowned his empire with a splendid capital, the reconstructed city of Samarkand, which continued to play an important role in history until quite recent times.

Tīmūr’s successors, who were, unlike him, essentially peace-loving, devoted themselves to the support of culture, the arts, and religion and to the preservation of his territorial legacy. His fourth son, Šāhroḵ (807-50/1407-47), succeeded him as ruler of Transoxania, though he lived in Herat, and earned a great reputation as a friend of scholars and poets and as a patron of architecture. He installed his son Oloḡ (Ulūg) Beg as governor at Samarkand, where, following in the footsteps of his grandfather, he enlarged his palace and took steps to prevent the deterioration of many of his ancestor’s monuments. His personal interest was astronomy, to which he made significant contributions (cf. Barthold, 1935). Like his father, Oloḡ Beg was entirely integrated into Persian Islamic cultural circles, and during his reign Persian predominated as the language of high culture, a status that it retained in the region of Samarkand until the Russian revolution of 1917. Many works of poetry, history, and other learned subjects were composed there in Persian (as later in the empire of the Great Mughals in India). By contrast, Persian was disappearing in Anatolia at the same period, increasingly supplanted by Ottoman Turkish.

Despite occasional forays (particularly an expedition to Lake Issyk-Kul in 828/1425), the Timurids generally tried to effect a reconciliation with Moḡolestān and to strengthen trade relations with it, as well as with China. In 822-25/1419-22 Šāhroḵ dispatched an embassy to the capital of the newly installed Ming dynasty (1368-1644) to dispel fears of an imminent Mongol attack and to prevent the rulers of Moḡolestān from enlisting Chinese assistance against the Timurids. Šāhroḵ experienced considerable difficulties with the Sufis in Transoxania, who had succeeded in gaining considerable economic influence, particularly the order of the Naqšbandīya and their spiritual leader Ḵᵛāja Aḥrār. This situation changed radically after his death in 850/1447 and the murder of his son Oloḡ Beg in 853/1449. After 855/1451 a great-grandson of Tīmūr, Abū Saʿīd, ruled in Samarkand, though only with the help of the Uzbeks. This Turkish tribe had settled in the winter of 808/1405-6 on the northern bank of the Syr Darya (Jaxartes) and had conquered Ḵᵛārazm in 834/1430-31; by about 859/1445 it had taken the whole northern bank of the Syr Darya, freed itself from the domination of the White Horde (see above) in the north, and ravaged parts of Transoxania. It was impossible for Abū Saʿīd to punish these incursions; in fact, the Uzbeks had obtained a substantial voice in his affairs. Under the leadership of Ḵᵛāja Aḥrār the influence of the Sufi orders, which had been strong enemies of the khan’s predecessors, increased considerably, and tolerance in religious matters came to an end (cf. Gross; Chekhovich; Paul).

In contrast to western Turkestan, Moḡolestān continued to maintain itself as an independent power in the region of the Ili and its tributaries, Lake Issyk-Kul, the Tarim basin, and the Turfan oasis as far west as the Ala Tau mountains and the upper Yenisei (Naryn) rivers. An internal reorganization of the country occurred in connection with the establishment of Islam and the Turkish language under the khan Esen Buqa (Īsen Būqā) II (833-67/1429-62),though these changes led to a long struggle with the “pagan” Oïrats (Ūyrāts, Qalmaqs) on the Ili and, after 855-57/1451-53, to incursions into Transoxania, which caused serious devastation there. Beginning in 860-61/1456-57, the khan had the support of the Uzbek ruler of Ḵᵛārazm, Abu’l-Ḵayr Khan, who eventually, in 873/1468, fell in battle against the rebels, or Kazakhs, as they have been known since that time. In the meantime Abū Saʿīd had managed to establish Esen Buqa’s brother Yūnos as a counterclaimant to the throne of Moḡolestān. Between Yūnos and the Oïrats, who had advanced into the region of the Amu Darya (Oxus), Esen Buqa’s power was increasingly curtailed. The Turkish amirs and clans joined forces with Yūnos, who, after his brother’s death and ten years of warfare, became ruler of Moḡolestān in 876-77/1472. During these battles the power of the tribes in Moḡolestān had increased significantly; the khan was forced to allow the Doḡlāt clan to form a kind of vassal state in the southwestern Tarim basin, which was, however, weakened by internal conflict.

The death of Yūnos in 891/1486 or 892/1487 in Tashkent, where he had resided during his last years, was followed by a civil war between his two sons, who were also forced to fend off attacks from the Timurids and the Chinese in turn. They also fought an indecisive war with the Doḡlāt clan (until about 904/1499). During these conflicts political order in Moḡolestān deteriorated, and the situation in Transoxania became increasingly unstable as well. The fading of Central Asia from the main arena of world history was at hand.

Western Central Asia still enjoyed a period of flourishing cultural life under the Timurids, despite many external difficulties. But after Abū Saʿīd was killed fighting the Qara Qoyunlū in 873/1469, his two sons engaged in fratricidal conflict; after frequent clashes with Yūnos as well, they were conquered by the Uzbeks, under the leadership of Moḥammad Šaybānī (Šïbanī) Khan, at the beginning of the 10th/16th century. He took Transoxania in 906/1500 and western Moḡolestān in 914/1508. Only the area east of the Ili and south of the Tien Shan mountains remained in the possession of the house of Čaḡatai.

During this period of struggle the last important Timurid ruler, Ḥosayn Bāyqarā (875-912/1470-1506), controlled large portions of eastern Persia and Central Asia from his capital at Herat. He encouraged the development of Persian literature and literary talent in every way possible; among the outstanding literary figures who benefited from his patronage were the poet ʿAbd-al-Raḥmān Jāmī (817-98/1414-92) and the historian Mīrḵᵛānd (836-903/1433-98). At the same time Sultan Ḥosayn also allowed his famous vizier, the noted poet ʿAlī-Šīr Navāʾī, to further the cause of his mother tongue, the Turkish spoken by the Chaghatay people (See chaghatay language and literature) and to champion its importance as a language of high culture. In fact, Navāʾī’s own works and the memoirs of the first Mughal emperor, Bābor (932-37/1526-30), guaranteed the establishment of this branch of Turkish, now known as Chaghatay, as a literary language (cf. Barthold, 1938; ʿAlī-Šīr Navāʾī; Bertel’s). This development was certainly related, at least in part, to the fact that in the early 10th/16th century Persia was converted by the Safavid dynasty to the Shiʿite branch of Islamic teaching, whereas Central Asia remained strictly Sunnite. Chaghatay became to some extent the language of this religious community, and Persian literary works from the Safavid realm had an aura of heresy. The influence of Persian was thus substantially undermined in Transoxania, though inhabitants of the region continued to look to the Persian of the earlier period, culminating in the poetry of Jāmī, and to accept it as the standard against the later “innovations.”

The end of the khanate. After vain attempts by Bābor, a grandson of the Timurid Abū Saʿīd who later founded the Mughal dynasty in India, to seize power in Transoxania in 917-18/1510-12, the Shaibanid (Shibanid) princes were in undisputed control of both this province and Ḵᵛārazm. ʿAbd-Allāh Khan b. Eskandar (991-1006/1583-98) in particular proved a very energetic ruler, first as regent for his father, then as his successor; it was he who took Balḵ, Samarkand, and the Farḡāna valley. After his death in 1006/1598, however, the power of the dynasty in Transoxania swiftly collapsed, and the territory was divided into several smaller individual states.

In the 10th/16th century only Moḡolestān remained in the hands of Čaḡatai’s descendents: The Tarim basin was allotted to Saʿīd Khan in 908-9/1503, when the patrimony was divided; the Yetisu, the Yulduz, and the Turfan oasis, to which Komul (modern Hami) was annexed in 919/1513, were ruled by his brother Manṣūr. As the two brothers worked together in harmony, the country enjoyed several decades of peace. During this period the culture of Transoxania, strongly influenced by that of Persia, spread through the region, whereas Chinese influence was imperceptible.

Saʿīd embarked on an invasion of Ladakh; his successor (ʿAbd-al-)Rašīd had to deal with a branch of the Doḡlāt clan in Kashmir and lost the Ili valley to the Kazakhs, retaining control only of Kāšḡar. He and his descendants gradually lost ground to clans claiming descent from the Prophet (sayyeds) and the first four caliphs (ḵᵛājas), the latter divided into two groups, the Aqtaḡlïk (White Mountain) ḵᵛājas and the Qarataḡlïk (Black Mountain) ḵᵛājas. With various members of the Chaghatayid house as figureheads, these groups were able to take control of different regions of the country, creating a number of small city-states, especially in the Tarim basin. In general the Aqtaḡlïk were allied with the Kazakhs and the Qarataḡlïk with the (Qara-)Kirghiz tribe. The history of the country during the several decades dominated by conflicts between these groups is obscure. Only in the 11th/17th century is it reported that the Aqtaḡlïk called the Dzungars to their aid and forced the last Chaghatayid khan, Esmāʿīl, to relinquish his throne, in 1089/1678. The head of the Aqtaḡlïk then declared himself khan, initiating the “holy state” of the ḵᵛājas (cf. Hartmann; Schwarz; McChesney).

The line of Čengīz Khan and his son Čaḡatai thus came to an inglorious end in eastern Central Asia. During the centuries of its rule, sometimes only nominal, the whole of Central Asia was converted to Sunnite Islam, which made possible its cultural development and the unification of its population within the framework of the Chaghatay language.

Bertold Spuler

Bibliography

  • ʿAlī-Šīr Navāʾī, Izvestiya Nauk SSSR, Otdelenie literatury i yazyka 6/6, 1947.
  • V. V. Barthold, Istoriya Turkestana, Tashkent, 1923.
  • Idem, Istoriya kul’turnoĭ zhizni Turkestana, Leningrad, 1927.
  • Idem, Ulug Beg and seine Zeit, tr. W. Hinz, Leipzig, 1935a.
  • Idem, Zwölf Vorlesungen über die Geschichte der Turken Mittelasiens, tr. T. Menzel, Berlin, 1935b.
  • Idem, Herāt unter Ḥusain Baiqara, dem Timuriden, tr. W. Hinz, Leipzig, 1938.
  • Idem, Four Studies in the History of Central Asia, ed. V. Minorsky, Leiden, 1956-62.
  • Idem, Turkestan 4, London, 1977, chap. 5. E. E. Bertel’s, Navoi, Moscow and Leningrad, 1948.
  • B. Brentjes, Mittelasien. Eine Kulturgeschichte der Völker zwischen Kaspischem Meer and Tien-Schan, Vienna, 1977.
  • M. Bussagli, Cultura e civiltà dell’Asia Centrale, Turin, 1970.
  • O. D. Chekhovich, ed., Samarkandskie dokumenty XV-XVI vv. o vladeniyakh Khodzhi Akhrara v Sredneĭ Azii Afganistane (Documents from Samarkand of the 15th and 16th centuries on the possessions of Ḵᵛāja Aḥrār in Central Asia and Afghanistan), Pamyatniki pis’mennosti Vostoka 31, Moscow, 1970.
  • B. A. Gafurov, ed., Istoriya i kul’tura narodov Sredneĭ Azii, Moscow, 1976.
  • J. Gross, Khodja Ahrar. A Study of the Perceptions of Religious Power and Prestige in the Late Timurid Period [Central Asia], Ph.D. diss., New York University, 1982.
  • Grosser historischer Weltatlas II: Mittelalter, ed. J. Engel, Munich, 1970; III: Neuzeit, Munich, 1957.
  • M. Haidar, “The Mongol Traditions and Their Survival in Central Asia, XIVth/XVth Centuries,” Central Asiatic Journal 28/1-2, 1984, pp. 57-79.
  • M. Hartmann, Ein Heiligenstaat im Islam. Das Ende der Čaghataiden and die Herrschaft der Choğas in Kašgarien, Der islamische Orient 6-10, Berlin, 1905.
  • K. Herrmann, Historical and Commercial Atlas of China, Cambridge, Mass., 1935.
  • P. P. Ivanov, Ocherki po istorii Sredneĭ Azii (XVI-seredina XIX v.), Moscow, 1958.
  • L. Kwanten, Imperial Nomads. A History of Central Asia 500-1500, Leicester, 1979.
  • R. D. McChesney, “The Amirs of Muslim Central Asia in the XVIIth Century,” Central Asiatic Journal 26, 1983, pp. 33-70.
  • F. Machatschek, Landeskunde von Russisch-Turkestan, Stuttgart, 1921.
  • B. F. Manz, “The Ulus Chaghatay before and after Temür’s Rise to Power. The Transformation from Tribal Confederation to Army of Conquest,” Central Asiatic Journal 27/1-2, 1983, pp. 79-100.
  • A. Mostaert and F. W. Cleaves, Les lettres de 1289 et 1305 des ilkhans Aṛγun et Öljeitü, Cambridge, Mass., 1962.
  • J. Paul, Die politische and soziale Bedeutung der Naqšbandiyya in Zentralasien im 15. Jh., Ph.D. diss., University of Hamburg, 1989.
  • P. Pelliot, Haute Asie, Paris, 1931. A. Schultz, Die natürlichen Landschaften von Russisch-Zentralasien, Hamburg, 1920.
  • G. Schwarz, “The Khwājas of Eastern Turkestan,” Central Asiatic Journal 20/4, 1976, pp. 266-96.
  • D. Sinor, Inner Asia. A Syllabus. History—Civilization—Languages, Bloomington, Ind., 1969.
  • B. Spuler, “Geschichte Mittelasiens seit dem Auftreten der Turken,” in HO 1/5: Geschichte Mittelasiens, pp. 123-310; tr. by F. R. C. Bagley in The Muslim World. A Historical Survey, ed. Bagley, III: The Last Great Muslim Empires, Leiden, 1969, pp. 219-59; IV/1: Modern Times, Leiden, 1981, chap. IV.
  • Westermanns Atlas zur Weltgeschichte, Brunswick, 1956.

CENTRAL ASIA vi. In the 16th-18th Centuries

I. The appanage state system. An overview (Table 28a; Table 28b).

Table 28 (part 1). The appanage "khans."Table 28 (part 1). The appanage "khans."View full image in a new tab

Table 28 (part 2). The appanage "khans."Table 28 (part 2). The appanage "khans."View full image in a new tab

With the beginning of the 10th/16th century a significant political transformation occurred in Central Asia. After governing for more than a century the Timurids were expelled from Transoxania, Balḵ, and Khorasan and founded the Mughal state in northwest India. The Chaghatay khans, the legitimizers of Timurid rule in Central Asia, were dislodged from eastern Transoxania and established a new state centered in Kashgar and Yarkand. Khorasan, which had been oriented for over a century toward Central Asia, was annexed to the new Safavid/Qezelbāš state. In the 10th/16th and 11th/17th centuries Central Asia, including Transoxania, Greater Balḵ, and Ḵᵛārazm, witnessed a neo-Chingizid (Jochid) political revival, spearheaded by the ʿArabshahid/Shibanid (Shaibanid) lineage in Ḵᵛārazm (see ʿARABŠĀHĪ) and the Abulkhairid/Shibanid and Toqay-Timurid lines in Transoxania and Greater Balḵ. Though marked by the restoration of Chingizid khanates, by the competition between the khanates and the Safavid/Qezelbāš state for control of Khorasan, and by rivalry between the khanates and the Mughal state in India for influence in Badaḵšān, in the main political life was shaped by the neo-Chingizid appanage system of state and its internal dynamic.

Sovereignty and succession. In the neo-Chingizid state installed by Moḥammad Šïbānī (Šaybānī) Khan at the beginning of the 16th century, sovereignty was corporate, embodied in the ruling or royal clan and shared among its eligible members. Its focus was the khanate (ḵānīyat), an ancient institution, whose archetypal representative was the mythologized figure of Čengīz Khan. Ideally, the khan (or khaqan) was a first among equals who presided over the assemblies (qoreltāy, kangāš) of the royal clan members and their supporters, at which matters of mutual interest were settled. The precepts of neo-Chingizid sovereignty were contained in an unwritten body of ordinances referred to in the sources by such terms as the yāsā(q) and yūsūn, tūra, and āʾīn-e čengīzī or “Chingizid constitution,” which political behavior was measured by and conformed to.

In a system in which sovereignty was corporate, membership in the royal clan was of obvious importance. In the 16th and 17th centuries, the first requirement was lineage through agnates to Jočī, the eldest son of Čengīz Khan. At the beginning of the 16th century circumstances further limited eligibility to the agnates of Šïbān, son of Jočī. The Transoxanian Jochids of the 16th century and the Khwarezmian Jochids of the 16th and 17th centuries legitimized their khanates on the basis of Shibanid descent. The Jochids of 17th-century Transoxania traced their legitimacy to another son of Jočī, Toqāy Tīmūr. The process of formalizing eligibility was complex and fluid. Factors such as the resources controlled by the clan, the outcome of contests for those resources, and the size of the royal cohort determined eligibility.

One of the principal institutions prescribed by the Chingizid constitution and adhered to throughout the 16th and 17th centuries was succession by seniority, which meant that succession could not be predicted. This made it difficult for power to accrue to the khanate and raised the level of conflict among the eligibles. The institution of heir apparency (qaʿalḵān, qaʿalḡa, qonalḡah) evolved in response to this problem. On two of the three occasions in the 16th century when an heir apparent was designated, the heir apparent predeceased the khan (during the reign of Kūčkonjī Khan). On the third occasion (under ʿAbd-Allāh Khan b. Eskandar Khan) the attempt was to circumvent, not facilitate succession by seniority. This indeterminacy of the seniority principle combined with corporate sovereignty and clan eligibility gave the system its special characteristics.

Membership in the ruling/royal clan and succession to the khanate came to be characterized as much by contest between competing “cousin clans” within the royal clan as by strict adherence to the seniority rule. Although the theoretical group of eligibles might be large, the political prestige arising from the outcome of contests and the control of economic and military resources foreordained a much smaller cohort—what M. Dickson (1960) has labeled the “neoeponymous clan,” that is, the periodic rise of new royal clans out of the old.

The appanages. The neo-Chingizids adopted an appanage system of state compatible with their notions of corporate sovereignty and succession, and the territories of the region subordinate to the royal clan were distributed among its male members. The four major appanages of the Abulkhairid/Shibanids and the Toqay-Timurids were the regions of Bukhara, Samarqand, Tashkent, and Balḵ. Lesser appanages included the Mīānkalāt between Samarqand and Bukhara, with the towns of Nūr, Karmīna, Dīzaq (Jīzaq), and Kūfīn; Sogd, with its center Āfarīnkent; Andejān, Aḵsīkat, Khojand, and Šāhroḵīya (formerly Fanākat) in the Farḡāna valley; Orā Tīpa; Ḥeṣār-e Šādmān (near present-day Dushanbe) and the Vaḵš river valley; Šahr-e Sabz (Kaš); and Badaḵšān. These tended to fall under the influence, if not jurisdiction, of the major appanages. Thus the Farḡāna region was contested between the appanage holders of Tashkent and Samarqand for much of the 16th century while Šahr-e Sabz was alternately subordinate to Bukhara and Samarqand.

Within the Arabshahid/Shibanid state of Ḵᵛārazm the major appanages were cities and their surrounding oases. Abu’l-Ḡāzī (pp. 213, 222, 243 [text]; Dickson, 1958, appendix I, p. xi) divides the two major appanage regions into the “mountainside” (oases along the northern flanks of the Kopet-Dag) and the “riverside” (Amu Darya delta towns). The two major “riverside” appanages were Ūrganj and Ḵīva (Ḵīvaq), lesser appanages were Wazīr and Kāt. The appanages were further divided into subappanages, whose distribution was supervised by the head of the branch of the royal clan assigned to the appanage, and, as time passed, the heads of appanages began to be styled “khan,” although it was always clear that there was a distinction between an appanage khan and the khan of the entire Jochid state (who headed his own family’s appanage as well).

The system was fundamentally decentralized. The appanage holder generally had the power to appoint, as well as to assess and collect revenues and to muster the military forces of the appanage. Throughout the period component parts of a combined campaign army are often named after the appanages they come from: laškar-e Balḵ, sepāh-e Tāškent, and so on. The reigning khan’s power of jurisdiction, as well as his ability to initiate and direct military campaigns, derived in large part from his persuasive powers and individual prestige, however, not from any prescriptive right. Nonetheless, a powerful reigning khan could exert direct control over the fiscal and military affairs of the individual appanages through the power to conduct military and fiscal audits, which in turn meant some inherent subordination of the appanages to central control. Control of the ataliqate was another indicator of central control. Every appanage holder had an atālīq/atalïq, an amir, a non-Chingizid, who supervised the administration and the military. During the reigns of powerful khans the atālīqs were appointed by and reportedto the khan. An indicator of appanage autonomy, therefore, is the source of the atālīq’s appointment. When the reigning khan exercised the power of appointment, it shows a greater degree of central authority. This standard also applied at the appanage level when the holder appointed atālīqs to the subappanages. But when the appanage or subappanage holder chose his own atālīq greater local autonomy may be assumed.

Appanage holders mutually consolidated and expanded their power through internal and external alliances, direct contests, and conquests of other appanages. The sources give the impression of autonomous appanages, self-reliant in military and diplomatic affairs, and yet bound into a loose confederation through adherence to the Chingizid constitution and acceptance of the legitimacy of a particular royal clan and its right to the khanate. The appanage system was neither static nor rigid. Although the boundaries of the individual territories retained a certain consistency over the two centuries in which the system held sway and devotion to the Chingizid way remained constant, the actual form in which the appanage state appeared at any given moment was a product of the political circumstances of the time.

Amirs, sultans, khans, ʿālems, and shaikhs. The appanage framework contained two formal, prescribed classes of participants, the Uzbek amirs and the Chingizid khans and sultans, and one group of informal participants: the ʿālems and shaikhs. Probably the most influential class were the amirs, leaders or highly placed members of Mongol and Turkic tribal groupings (ayl, ūlūs, ūymāq, tūmān, qošūn, ṭāʾefa, qabīla, qawm, etc.) with no claim to the Chingizid khanate. Though many of them were descendants of Čengīz’s ascendants or descendants of cognates, this conveyed no legitimacy within the political context of 16th- and 17th-century Central Asia. The tribal attributives of amirs frequently given in the sources: Kerāyt, Mangḡīt, Mīng, Ūyḡūr, Qalmāq, Qerḡez, Dūrmān, Qonḡrāt, Merkīt, Qaṭaḡan, Jalāyer, Qārlūq, Ūḡlān, Qūšjī, Bahrīn, Būyrāk, Qānqlī, Ālčīn, and others, reveal them as the Uzbeks par excellence, or more precisely the “Mongols and Uzbeks” as distinct from the Chingizids. But the sources often mention amirs of non-Uzbek (i.e., Tajik) lineage, as well, including religious figures and members of the main brotherhood organizations, such as Sayyed Atāʾī, Pārsāʾī, ʿAzīzān, Naqšbandī, Jūybārī, Dahpīdī, Ṣāleḥī, and so on, and others with no discernible intellectual orientation but who held either a military or bureaucratic office.

By and large the term amir seems to have been generally limited to those who performed some military function, but the line between military and civil administration is not at all sharp. On one hand we encounter men like Qol Bābā Kūkaltāš, probably the greatest military genius of the 16th century, who was also prominent as a bureaucrat at the khanate level, conducting fiscal audits (taḥqīqāt) in his capacity as mošarref-e dīvān, and held the post of ṣadr-e ḵānī, which supervised the judiciary. On the other hand, men like Ḥasan Ḵᵛāja Naqšbandī, a Tajik and the naqīb of Bukhara in the second half of the 16th century, played a preeminently military role. The differences between familiar concepts like amir and aʿyān or Turk and Tajik become less clear in individual cases. One might say that the distinguishing aspects of amirhood were military rank or participation in military campaigns and administrative rank other than clerical.

There is little evidence that members of the Turko-Mongol tribal groups acted in concert in a way comparable to the contemporary Qezelbāš ūymāqs. On numerous occasions amirs are assigned to military duties “with their ūlūsāt” or “at the head of their qošūn,” but whether military units such as the čohragān or īčakīān were organized along tribal lines is difficult to say. It is clear, however, that within a large tribal grouping like the Naymān, Dūrmān, or Qūšjī, for example, there was little if any solidarity vis-à-vis other tribal organizations or the royal clan. At the height of the civil war between the Janibegid, Soyunjokid, and Kuchkonjid subclans of the Abulkhairid/Shibanid lineage, one finds Naymān amirs fighting alongside each of the contenders. In addition, if we look at the main amirid supporters of any one of the contenders, the Soyunjokid, Bābā b. Nowrūz-Aḥmad, ʿAbd-Allāh b. Eskandar, the Janibegid, or Javānmard-ʿAlī, the Kuchkonjid, no one Uzbek group predominates. Amirid loyalties appear to have been to individuals and families, and sons of amirs often worked for the same subclan as their fathers. The commitment of the amirs as a group to the legitimacy of the Jochids tended to mute expressions of tribal solidarity. This is not to say that tribal solidarity did not exist. The initial campaign of ʿAbd-Allāh b. Eskandar against his Janibegid cousin at Balḵ was precipitated by complaints from Naymān amirs there that they were being collectively persecuted by the Balḵ appanage holder. At the beginning of the 17th century, the Toqay-Timurid leader Bāqī-Moḥammad b. Dīn-Moḥammad established a foothold at Bukhara because of the disaffection of the Dūrmān amirs there from the Janibegid appanage holder. But in both cases, the people involved were few, and neither the Naymān nor Dūrmān as a whole played any role. More of a sense of tribal solidarity is displayed much later, in the late 17th and early 18th century, with the emergence of the Mangḡīt as the preeminent tribal group of Bukhara and the Qaṭagān in Balḵ.

The second formal class of participants in the appanage system is that of the Chingizid eligibles, the khans and sultans. In Shibanid and Toqay-Timurid usage the terms have a very precise meaning. In theory there is but one khan or khaqan, the head of the khanate. As the appanage system evolved, the title khan also came to be applied to the head of a major appanage, especially if that individual was also the senior member of one of the subdivisions of the royal clan. Khan was also a nontechnical term used to indicate Chingizids of exceptional military achievement who were, as junior members of the royal clan, the puissant khan as distinct from the regnant khan. Contemporary sources use phrases like “apparent khan” and “real khan” (ḵān-e ṣūrī, ḵān-e maʿnawī) and “greater” and “lesser” khan (ḵān-e kalān/aʿẓam and ḵān-e ḵord/aṣḡar). There are thus three contexts within which the term is used: 1. to designate the nominal, titular head of state, 2. to designate the head of an appanage, and 3. to single out the most vigorous and military active Chingizid figure, often himself the head of an appanage.

The title sultan was appended to the name of every member of the royal clan as well as other descendants of Čengīz Khan, (e.g., Maḥmūd-Solṭān, Dīn-Moḥammad-Solṭān) and only those members. Thus it was not limited to members of the royal clan but was used by all Chingizids. Within the Abulkhairid/Shibanid state there were at least two prominent groups of sultans who did not belong to the Abulkhairid clan. These were the so-called “Ḥeṣārī sultans,” more properly the Bakhtiyarid/Shibanids and the Toqay-Timurids. As Chingizids their right to the title sultan is always acknowledged, even in the Abulkhairid sources.

There was a third class of participants in the political process, who might be called “the learned,” those whose charisma and influence arose from their identification with sacred lore, whether scriptural or spiritual. These included the ʿālems, or scholars of the disciplines derived from Muslim scripture, logic, rhetoric, exegesis, jurisprudence, and grammar, as well as the sheikhs of the mystic brotherhoods, the leaders of the Kobrawīya and Naqšbandīya suborders in Central Asia, the ḵᵛājagān, members of those orders, and darvīšān, individuals of recognized spiritual merit who avoided denominational affiliation (eremites, qalandars, and the like). In general, the most influential figures on the 16th-century political scene, men like Sheikh ʿAbd-al-Walī Pārsā (Ḵᵛāja Jān-Ḵᵛāja) in Balḵ, Ḵᵛāja Saʿd Jūybārī (Ḵᵛāja Kalān-Ḵᵛāja) in Bukhara, and Ḵᵛāja Hāšem Aḥrār in Samarqand were centrists, with credentials covering a broad part of the spectrum.

The learned class had no formal role in the Chingizid scheme, but the function it served of intercession and mediation was indispensable to the operation of the appanage system. It mediated both between individuals in the other two groups and between the disenfranchised (all those who did not belong to either of these three categories) and the ruling caste. In addition, most nonmilitary offices (šayḵ al-Eslām, qāżī, moftī, raʾīs, modarres, wazīr, dīvān) and the military office of naqīb were filled by members of this class.

II. The Abulkhairid khanate (Table 26).

Table 26. The Abulkhairid/Shibanid DynastyTable 26. The Abulkhairid/Shibanid DynastyView full image in a new tab

The empire of Šïbānī (Šībānī) Khan, r. 907-16/1501-10. In the late 9th/15th century only Tashkent and Moḡūlestān (eastern Turkestan) were subject to the revived Chaghatay khanate. The rest of Central Asia was still under the Timurids: ʿOmar Sheikh Mīrzā (d. 899/1494) governed the Farḡāna valley from Andejān; Sultan Aḥmad Mīrzā (d. 899/1494) governed the rest of Transoxania from Samarqand; and Sultan Ḥosayn Mīrzā ruled the lands south of the Oxus, Balḵ, and eastern Khorasan. The death of ʿOmar Sheikh Mīrzā set off a bitter universal struggle for control of Transoxania and Khorasan. Although Mīrzā Bāysonḡor (d. 905/1499), the son of Sultan Maḥmūd Mīrzā (d. 900/1495), and after him ʿOmar Sheikh Mīrzā’s son, Ẓahīr-al-Dīn Bābor (d. 937/1530), would press Timurid claims to the region for almost two decades, neither was an effective strategist and battlefield tactician.

Moḥammad Šībānī (aka Šāhī Beg, Šaybāq, Šaybak, and Šāhbaḵt), grandson of Abu’l-Ḵayr, had been a mercenary horseman in the service of a Timurid prince before becoming governor of Tashkent on behalf of the Chaghatay/Chingizid khan Sultan Maḥmūd at the beginning of 899/1494, and in 907/1501 he took control of Samarqand, at first on behalf of Sultan Maḥmūd. He launched a series of campaigns that brought him control over the Farḡāna region, Turkestan, and Tashkent (907-8/1502), the Amu Darya delta (Ūrganj and Kiva; 910-11/1504-5), Balḵ (911/1505), and Herat (913/1507), and by mid-913/end of 1507, he held the largest area of any political figure since the time of Šāhroḵ Mīrzā (d. 850/1447).

Moḥammad Šībānī’s achievements were based on military genius and personal charisma, and the loyalties generated by his successes were to his person rather than to his clan. When he was killed at Marv in Ramażān 916/December 1510, the “empire” he had created fell into the hands of competing powers, such as the Safavid shah Esmāʿīl I (eastern Khorasan and Herat) and the Timurid Ẓahīr-al-Dīn Bābor partly backed by Esmāʿīl (Balḵ, Bukhara, Samarqand). In eastern Transoxania amirs loyal to the Chaghatay khans ousted the Abulkhairids from the Farḡāna valley and Tashkent. In Ḵᵛārazm a rival Shibanid clan, the Arabshahid, seized control of the lower Amu Darya region. By the end of 917/1511, the Abulkhairids had lost every major city taken by Moḥammad Šībānī.

The corporate khanate of the Abulkhairids, 918-59/1511-50. In late winter 917/1512, encouraged by popular dissatisfaction with the rulers, Moḥammad Šībānī’s uncle Soyūnjok b. Abi’l-Ḵayr marched against Tashkent and his nephew ʿObayd-Allāh against Bukhara, defeating Bābor at Kūl Malek in spring 918/1512. Bābor fled south to Ḥeṣār-e Šādmān, where his Chaghatay allies also turned on him. In the following winter a large Qezelbāš army was routed by the Abulkhairid forces at Ḡejdovān, just north of Bukhara.

After Šībānī Khan’s death the leaders of the Soyunjokid, Kuchkonjid, Shahbudaqid, and Janibegid clans, as well as some of the non-Abulkhairid clans (e.g., the Bakhtiyarid), resorted to a system incorporating consultation and consensus between the major clans and their amirid backers, an acceptance of the idea of appanage autonomy, and a weak form of overall leadership based on seniority within the Abulkhairid house. The apparent inability of the clans to operate under a strong central leadership and their political success as individual appanages set the pattern for the politics of the 16th century. This pattern was formalized in two qoreltāys of the clan leaders, one probably held in the spring of 916/1511 after Šībānī’s death, in which the senior Abulkhairid, Kūčkonjī Moḥammad (or Kūčūm), was elected khan.

The second qoreltāy was held after the reestablishment of Abulkhairid control over Tashkent, Samarqand, and Bukhara and fixed the shape of the political order for the next two hundred years. It confronted two main issues: succession to the Chingizid khanate and the distribution of territory. According to Ḥāfeẓ Tanīš (Šaraf-nāma, 1983, fol. 33b/p. 87), the election of the khan was based on established principle. “According to ancient [Chingizid] law, the sultans consulted together on the issue of the khanate. Since Kūčkonjī-Solṭān was the eldest they gave him the title khan. The (title) qaʿalḡah which means “heir apparent” they conferred on Soyūnjok-Solṭān (Šaraf-nāma, 1983, fol. 33b). In the distribution of territory, Bukhara went to the Shahbudaqid clan led by ʿObayd-Allāh; Mīānkal and Soḡd-e Samarqand were given to the line of Ḵᵛāja Moḥammad headed by Jānī Beg. Tashkent was allotted to Soyūnjok, and Samarqand was awarded jointly to Kūčkonjī and Moḥammad-Tīmūr, the son of Moḥammad Šībānī. This last arrangement reflected the ancient status of Samarqand as “capital” and therefore the appropriate seat for a head of state, as well as the fact that Samarqand had been Moḥammad Šībān’s center and thus properly belonged to his lineage.

Certain regions consequently became associated with particular Abulkhairid subclans, and the principle of seniority succession was accorded historical roots. In the 930s-40s/1520s-30s the four main Abulkhairid clans consolidated and expanded their appanage holdings. The Janibegids, allotted the least valuable appanage, improved their fortunes considerably when one of the clan members, Kīstan-Qarā-Solṭān, succeeded in wresting control of Balḵ from the Timurids in 932-33/1526. In Samarqand, where appanage control was at first jointly shared by Kuchkonjids and Shahbudaqids, the death of Moḥammad-Tīmūr in 920/1514 left Kuchkonjids in complete control. Kūčkonjī and his successors directed their expansionist ambitions to the south in the direction of Badaḵšān and to the Farḡāna valley, where their interests conflicted with the Soyunjokids in Tashkent. In Tashkent the Soyunjokids, under the leadership of Soyūnjok and then his son, Nowrūz-Aḥmad, first fended off the Chaghatay khans in the east, later they joined forces with them against the “Qazaq” Shibanid khans of the Talas-Chu basin. In Bukhara the Shahbudaqid clan placed its expansionist hopes in Khorasan.

ʿObayd-Allāh’s succession on the basis of seniority as regnant khan in 940/1533, after nearly a decade as puissant khan, signaled a temporary end of the struggle for Khorasan; by the end of his reign the Abulkhairid khanate had solidified into a quadripartite state centered on Balḵ, Bukhara, Samarqand, and Tashkent with the capital moving to the appanage center of each succeeding khan. The khanate maintained its integrity, with succession by seniority within the royal clan, but the regnant khanate was now formal, titular, and for the most part nominal, and real power rested in the hands of the heads of the appanages, now microcosms of the khanate. Nominal leadership was conferred on the senior Janibegid, Soyunjokid, Kuchkonjid, or Shahbudaqid; energetic younger sultans solicited amirid support and exercised authority from their subappanage centers.

After ʿObayd-Allāh two sons of Kūčkonjī succeeded to the nominal ḵānīyat, ʿAbd-Allāh and ʿAbd-al-Laṭīf. Nowrūz-Aḥmad, appanage khan at Tashkent in 930-31/1524, became puissant khan of Transoxania. Few of Nowrūz-Aḥmad’s early efforts are known, but sometime between 959 and 963 (1552-56), he joined ʿAbd-al-Rašīd Khan, Chaghatay khan of Kashgar, in a major successful campaign against the “Qazaq Shibanids.” Of the other appanage khans and sultans during the period 946-57/1540-50, Pīr-Moḥammad, the Janibegid appanage khan at Balḵ, was drawn into the internecine Timurid struggle over Badaḵšān, when, in early 955/1549, Homāyūn, the Mughal leader, led a force from Kabul through Aybak and Ḵolm toward Balḵ, which was routed by a combined Shahbudaqid/Janibegid army.

By 957/1550 the individual appanages in general enjoyed internal autonomy and set their own foreign policies. Succession to the nominal khanate was based on seniority within the royal, Abulkhairid/Shibanid, clan. The appanage structure appeared stable at this point, capable of defending its component parts against external enemies, and able to meet the expectations of its amirid and sultanic elements. After 957/1550 this all changed.

Most of what we know of the appanages in the 940s-50s/1530s-40s pertains to the development of the cities. ʿObayd-Allāh’s son ʿAbd-al-ʿAzīz rebuilt the city walls of Bukhara, developed the shrine of Bahāʾ-al-Dīn Naqšband with a ḥaẓīra and ḵānaqāh (both of which still stand) and built a congregational mosque inside the Outer City, a madrasa (951/1544), and several mosques; the madrasa of Mīr ʿArab was completed in 942/1535. In Balḵ, in the 940s/1530s-40s, Kīstan-Qarā completed work on a congregational mosque, whose foundations were laid by Sultan Ḥosayn Mīrzā Bāyqarā (d. 911/1506), and rebuilt the city’s fortifications.

Civil War in Transoxania and Balḵ, 957-90/1550-82.

1. Elimination of the Shahbudaqids, 957-64/1550-56. In the first four decades of clan rule (918-57/1512-50) four major appanages had formed, loosely federated under the moral authority of the regnant khan but by and large autonomous in internal and international affairs: the Soyunjokids in Tashkent, Turkestan, and the Farḡāna valley; the Kuchkonjids at Samarqand, Šahr-e Sabz, and as far west as Qaršī/Nasaf; the Janibegids represented in 957/1550 by two main subclans, the Eskandarid at Mīānkāl and Sogd-e Samarqand in Transoxania and the Pirmohammadid at Balḵ; and the Shahbudaqids at Bukhara, with two subclans, the descendants of Moḥammad Šībānī and those of his brother Maḥmūd and ʿObayd-Allāh b. Maḥmūd. Of the four clans, the Shahbudaqid was the weakest. Allied with the Kuchkonjids was a fifth Shibanid, but non-Abulkhairid, clan, the “Ḥeṣārī sultans,” at Ḥeṣār.

Succession had been fairly uneventful. Since both great and appanage khans were barely primi inter pares, little seems to have been at stake in the way of resource redistribution at the time of succession. But in 957/1550 the situation changed with the death of ʿAbd-al-ʿAzīz. The senior surviving Mahmudid was his brother Moḥammad-Raḥīm, himself a political nonentity, but important as the stalking horse for his son Borhān’s political ambitions. In 957/1550 leadership of the Bukharan appanage passed to Moḥammad-Yār, grandson of Moḥammad Šībānī and the head of the other Shahbudaqid subclan. The adjustments made would probably not have had any wider significance had not another appanage involved itself in the succession and ignited a civil war involving all four appanages.

War began when Pīr-Moḥammad, the senior Janibegid and the appanage holder of Balḵ, attempted to seize control of Bukhara. It is the first of many recorded instances of one Abulkhairid trying to annex the territory of another. Responding to Moḥammad-Yār’s plea for help, Nowrūz-Aḥmad and ʿAbd-al-Laṭīf led a joint army against Mīānkāl and expelled all the Janibegid sultans. Pīr-Moḥammad returned Bukhara to Moḥammad-Yār and withdrew to Balḵ, where he now had to find appanage room for his brother, Eskandar, and his brother’s sons and amirs. But Moḥammad-Yār’s status was diminished by this episode, and Borhān, the most powerful of the Bukharan sultans, forced him to accept his father as co-ruler.

By the end of 958/1551, the balance of the previous forty years had been destroyed: the Eskandarid/Janibegid appanage had been conquered and divided between the Soyunjokids and Kuchkonjids, and the Shahbudaqids were at war with themselves. Over the next six years, the Janibegids at Balḵ, in particular the displaced Eskandarids, tried to exploit the weaknesses of the Shahbudaqids as well as any opportunities offered them by their Soyunjokid and Kuchkonjid cousins to recoup the position they had lost, and in Rajab 964/May 1557 the subclan led by Jānī Beg’s grandson, ʿAbd-Allāh b. Eskandar, finally established itself at Bukhara. The Shahbudaqid clan was ousted and vanished from the political scene.

2. Elimination of the Soyunjokid and Kuchkonjid clans, 964-90/1557-82. For the next quarter century, appanage politics were factional with the fragmentation of politics in every appanage. Bukhara was soon polarized between the two most powerful Janibegids, ʿAbd-Allāh and Yār-Moḥammad’s son Ḵosrow. At Balḵ Pīr-Moḥammad and his son Dīn-Moḥammad opposed both ʿAbd-Allāh and Ḵosrow. In Samarqand the brothers Javānmard-ʿAlī and Solṭān-Saʿīd and later the sons of Javānmard-ʿAlī emerged as competitors. In Tashkent, Nowrūz-Aḥmad’s sons, Darvīš and Bābā, similarly became the focus of rival factions. Sometimes (as in the case of Samarqand) an internal faction would bring in its allies from other appanages. After the death of ʿAbd-al-Laṭīf, two of his sons and three of his nephews sought appanage leadership. Although the senior member was Solṭān-Saʿīd, Javānmard-ʿAlī soon challenged him. Solṭān-Saʿīd won the backing of Nowrūz-Aḥmad and later even secured, briefly, the backing of the Chaghatay khan at Kashgar, ʿAbd-al-Rašīd. Javānmard-ʿAlī found his allies among both the Janibegids and his Kuchkonjid cousins, but after Solṭān-Saʿīd had been eliminated both cousins and sons challenged Javānmard-ʿAlī. The same pattern of events was repeated all over Transoxania and Balḵ. As soon as one faction was ousted and its territories and amirs redistributed, new factions appeared to challenge the new order. Clan solidarity both between and within the appanages broke down as generational change and the increase in the number of sultans in each appanage put intolerable pressure on available resources.

This continual warfare resulted in the creation of two appanage superpowers: One was the Eskandarid branch of the Janibegids, whose main figures after the death of Ḵosrow were Eskandar’s son ʿAbd-Allāh and his brothers, ʿAbd-al-Qoddūs (Dūstom-Solṭān) and ʿEbād-Allāh, and a cousin, Ūzbak b. Rostam. Throughout a period noted for the lack of cooperation between clansmen these four men maintained an extraordinary degree of unity. The other appanage power was the Soyunjokid family at Tashkent led by Bābā, son of Nowrūz-Aḥmad.

During the latter part of Nowrūz-Aḥmad’s life, a pattern was established in which inter-Kuchkonjid factionalism became the arena for contests between Soyunjokids and Eskandarids. After Nowrūz-Aḥmad’s death the pattern persisted but increasingly seems to have been polarized around the personalities of Bābā and ʿAbd-Allāh. The Kuchkonjid pretenders were eliminated through their own disunity and reliance on Eskandarid or Soyunjokid support. The case of Javānmard-ʿAlī (khan at Samarqand 980-84/1572-76) is illustrative: In 961/1554 he assisted Bābā in a battle against the Janibegids for control of Nasaf. Two years later, while appanaged at Aḵsīkat and Andejān in the Farḡāna valley, he backed his cousin Gadāy b. ʿAbd-al-Laṭīf against his brother, Sultan Saʿīd, who had the support of Bābā, for the khanate of Samarqand, thereby alienating the powerful Soyunjokid leader. He then tried to form ties with the Janibegid ʿAbd-Allāh, sending him troops during the siege of Termeḏ in the spring of 979/1572. Soon after, he and his son Abu’l-Ḵayr were with ʿAbd-Allāh in the army center at the battle of Kūk Gombad against Bābā (Ṣafar 980/July 1572). At the end of that year he brought troops to aid ʿAbd-Allāh against his first cousin Dīn-Moḥammad at Balḵ. But towards the end of 985/early 1578, as he faced pressure from his own son, he again renewed his alliance with Bābā. This alliance and Abu’l-Ḵayr’s request for aid from the Eskandarids led to a month-long siege of Samarqand (Ṣafar 986/April-May 1578), at the end of which Javānmard-ʿAlī surrendered, was imprisoned, and finally put to death at ʿAbd-Allāh’s orders.

The fall of Samarqand and its annexation to the Eskandarid appanage, now including Balḵ and Šahr-e Sabz as well as Bukhara, isolated the Soyunjokids and their Kuchkonjid allies. For the next four years the Janibegids mounted campaigns against Tashkent, at first in defense of the claims of Darvīš b. Nowrūz-Aḥmad to control of the khanate against his brother Bābā, and later in order to annex it to the Janibegid lands of Bukhara, Balḵ, and Samarqand. As Soyunjokid power waned in Tashkent, the “Qazaq” Shibanids, especially Ḥaqq-Naẓar Khan and his son and grandson, Šīḡāy and Tawakkol (Tevkel) reemerged as a potent new force in eastern Transoxania.

By 990/1582 the Eskandarid clan under the military leadership of ʿAbd-Allāh and the nominal khanate of his father Eskandar had consolidated the appanages of Tashkent, Samarqand, Bukhara, and Balḵ into one unified state. But adherence to the appanage principles remained as strong as ever.

The new age of empire, 992-1006/1584-98. Badaḵšān and Khorasan. Badaḵšān the nearest outpost of Mughal influence, had long been viewed by Mughal policy-makers as the staging ground for recapturing the ancestral lands of Balḵ and Transoxania. But it also served equally well, as Bābor, Homāyūn, and Akbar were to discover, as a hotbed for ambitious Timurids to nurture their hopes for Kabul and northwest India. From the Transoxanian perspective, Badaḵšān was an attractive object because of its proximity to Balḵ and to the southern reaches of the Samarqand appanage. In 956/1549 Homāyūn led an army against Balḵ but was defeated in no small part because of the refusal of his half brother, Kāmrān, who resided in Badaḵšān, to send him reinforcements. Later that same year there was a struggle between Kāmrān on one side and his half brother Hendal and two other Timurids, Solaymān Mīrzā (or Solaymān Shah), whom Bābor had first sent to Badaḵšān in 935/1528-29, and his son Ebrāhīm. In 967/1560, apparently trying to capitalize on the inter-appanage problems of the Janibegids, Solaymān marched against Balḵ but was defeated by Pīr-Moḥammad and ʿAbd-Allāh at Čašma-ye Jarzovān (Garzovān), where his son Ebrāhīm Mīrzā was killed.

From then on Badaḵšān’s politics and the policies of Solaymān Shah were generally directed toward Kabul. But in 982/1574, when ʿAbd-Allāh conquered Ḥeṣār-e Šādmān and installed an Eskandarid as appanage holder, Solaymān intervened on the side of the Ḥeṣārī sultans. But his candidate lost, and in a struggle between two parties at his own capital, Qalʿa-ye Ẓafar, he was ousted, and his seven-year-old grandson, Šāhroḵ Mīrzā, was installed in his place. In 986-87/1577-78, returning from a penitential ḥajj by way of Qazvīn, Solaymān obtained a commitment of military support from Shah Esmāʿīl II, but a few months later the shah was murdered. Nevertheless, with the backing of the Mughal sovereign, Jalāl-al-Dīn Akbar, Solaymān regained control of part of Badaḵšān. But his rivalry with the supporters of his grandson continued, which encouraged ʿAbd-Allāh to intervene. The campaign opened at Bukhara on 6 Moḥarram 992/19 January 1584 and concluded ten months later. In the course of the campaign, Qondūz, Ṭālaqān, Kāhmard, Ḡūrī, and Kūlāb were all wrested from Timurid control. Solaymān and Šāhroḵ attempted to restore their rule in Badaḵšān towards the end of Rabīʿ II 994/April 1586 but were dispersed by the army of ʿAbd-Allāh’s son, ʿAbd-al-Moʾmen, the appanage holder of Balḵ.

The expansion into Khorasan involved a greater commitment of resources and produced greater rewards. In the struggle to control the Safavid monarch, ʿAlīqoli Khan Šāmlū, the Qezelbāš governor of Herat, after losing the battle for control of the prince, ʿAbbās b. Moḥammad Ḵodābanda, who was soon to be shah, attempted to form an alliance with ʿAbd-Allāh at Bukhara in 994/1586. But faced with the reality of a Shibanid/Uzbek army come to occupy Herat at his request, ʿAlīqoli Khan in early Rajab 995/June 1587 locked the gates of the city. Nine months later, on 1 Rabīʿ II 996/29 February 1588 Herat was stormed and the garrison killed. From there the Shibanid/Uzbek armies extended their control over all the major cities of Khorasan, Qohestān, and Sīstān. Mašhad fell in the middle of Ḏu’l-ḥejja 997/November 1589, then Nīšāpūr, Sabzavār, and Esfarāʾen. Janibegid allies exploited Qezelbāš factionalism in Qohestān to extend Bukharan control there as well. The maleks of Sīstān acknowledged Transoxanian hegemony, while the Safavids of Qandahār also reached an accommodation with the new lords of Khorasan. From 996/1588 until the Safavid/Qezelbāš reconquest of 1006-7/1598, Transoxania had jurisdiction over most of northeastern and eastern Persia.

The Khorasan period had a pronounced influence on the development of appanage politics. In the siege and conquest of Herat and at Mašhad, Nīšāpūr, Sabzavār, and Esfarāʾen ʿAbd-al-Moʾmen played the leading role. In fact after the fall of Herat, over which ʿAbd-al-Moʾmen himself presided, ʿAbd-Allāh, though regnant khan, played no further part in the conquest of Khorasan but appears to have been satisfied with conferring Herat on his most trusted amir, Qol Bābā Kūkaltāš, turning Qohestān over to the Toqay-Timurid sultans, and then giving his son a free hand in the rest of Khorasan. In the confrontations between Safavid/Qezelbāš and Shibanid/Uzbek forces during the decade ʿAbd-al-Moʾmen represented the Shibanid house. From Iranian sources and letters exchanged by Shah ʿAbbās and ʿAbd-al-Moʾmen it is clear that ʿAbd-al-Moʾmen felt he was not properly rewarded for his services in Khorasan and was especially unhappy at not having been granted Herat, which was adjacent to his own appanage. But at the end of Ṣafar 999/late 1590 ʿAbd-Allāh convened a qoreltāy in Bukhara and named ʿAbd-al-Moʾmen heir apparent.

Another consequence of the Khorasan period was the rise of the Toqay-Timurid house to military prominence. The leading figures in the family (see below and Table 27) were Dīn-Moḥammad and his brothers Bāqī-Moḥammad and Walī-Moḥammad. ʿAbd-Allāh had given the governorship of the strategic city of Qondūz to Dīn-Moḥammad after its capture from the Timurids in Ṣafar 992/February 1584. Later Baḡlān was added to his territory. For his participation in the conquest of Herat in 996/1588, Dīn-Moḥammad was awarded Bāḵarz and Ḵargerd, and during the next decade, with the help of his brothers, he brought all of Qohestān under his control as far west as Ṭabas(-e Gīlagī?) on the road to Yazd. From there he conducted razzias that took him at times into Fārs. In 1003/1595 ʿAbd-Allāh gave Sīstān to Dīn-Moḥammad. By the time of the Safavid/Qezelbāš reconquest of Khorasan, the Toqay-Timurid line with Dīn-Moḥammad as its most powerful sultan had established itself as a credible representative of Chingizid constitutional politics.

Table 27. Toqay-Timurids mentioned in the articleTable 27. Toqay-Timurids mentioned in the articleView full image in a new tab

Collapse of the Abulkhairid/Shibanid khanate, 1006-7/1598-99. ʿAbd-Allāh Khan died in Rajab 1006/February 1598. For the Janibegids the politics of succession were theoretically mitigated by ʿAbd-al-Moʾmen’s appointment as heir apparent. But his activities after his succession destroyed any political advantage he might have had. He immediately began to liquidate his uncles and cousins and execute a number of his father’s amirs, among whom was Qol Bābā Kūkaltāš, but was himself assassinated after about six months by a group of amirs near Żāmen, between Orā Tīpa and Samarqand.

III. The Toqay-Timurid Khanate (Table 27).

The Interregnum, summer 1006-7/1598-spring 1007/1599. In Tashkent, a revived Qazaq Shibanid entity led by Tawakkol Khan b. Šīḡāy Khan had emerged as a major challenger to Janibegid legitimacy as a result of ʿAbd-Allāh’s death and ʿAbd-al-Moʾmen’s purges. Shortly after ʿAbd-al-Moʾmen had put the Janibegid appanage holder at Tashkent, Ūzbak b. Hazāra, to death, Tawakkol Khan occupied the city. On ʿAbd-al-Moʾmen’s death there was no prominent Janibegid sultan with the military and administrative credentials to capture wide amirid support and to combat the immediate Qazaq threat. In Samarqand and Balḵ amirid factions sought a legitimate candidate to sponsor. The Balḵ faction proclaimed the khanate of ʿAbd-al-Amīn, reportedly a son of ʿEbād-Allāh, ʿAbd-Allāh’s brother and until Ramażān 994/October 1586 the appanage holder of Samarqand. At Samarqand, where concern was greatest about the Qazaq approach, a Kanīkas amir and loyal servant of the Janibegids since at least 957/1550, ʿAbd-al-Wāseʿ Bī (Biy), first secured control of Samarqand and then sent a delegation of amirs to Bukhara, where they proclaimed the khanate of Pīr-Moḥammad b. Solaymān b. Jānī Beg. By the end of 1006/mid-1598 there were in effect three khanates in Transoxania and Balḵ, the Qazaq Shibanid khanate of Tawakkol b. Šīḡāy (with growing support from the Uzbek amirs) at Tashkent and the Janibegid khanates of ʿAbd-al-Amīn at Balḵ and Pīr-Moḥammad at Bukhara.

In Khorasan and Sīstān, sometime between February and August 1598, the Toqay-Timurids led by Dīn-Moḥammad b. Jānī-Moḥammad proclaimed their khanate in the name of the senior member, the grandfather of Dīn-Moḥammad, Yār-Moḥammad (who declined the honor) and then in the name of the next senior member, Jānī-Moḥammad. Dīn-Moḥammad moved on Herat at about the same time an army launched by Shah ʿAbbās at the news of ʿAbd-Allāh Khan’s death was approaching the city from the northwest. The Toqay-Timurid leader and his brothers reportedly debated whether to return to the “patrimonial lands” of Balḵ and Transoxania and establish the khanate there or stay and defend Herat. The decision was to stay, and in the ensuing fighting Dīn-Moḥammad was killed and Khorasan reoccupied by the Safavid/Qezelbāš. The surviving Toqay-Timurids and their amirid backers retreated to Transoxania. There, under the leadership of Bāqī-Moḥammad the family offered its services to the Janibegid khan of Bukhara in his fight with the Qazaqs, who in the preceding months had forced Samarqand’s surrender, taken two of the major fortress towns of Mīānkāl, Dabūsīya, and Kūfīn, and laid siege to Bukhara. But now some of those who had joined Tawakkol b. Šīḡāy abandoned him and rejoined the amirs supporting Pīr-Moḥammad Khan, and unable to take Bukhara he withdrew from Samarqand by the fall of 1007/1598. In return for his help Pīr-Moḥammad Khan awarded Samarqand to Bāqī-Moḥammad, who began to rebuild Toqay-Timurid fortunes. In late winter 1007/1598-99, the khan dismissed one of the leading Dūrmān amirs, Moḥammad Bāqī Bī, from his post as dīvānbegī and refused to bow to the appeals of Moḥammad-Bāqī Bī’s fellow Dūrmān amirs to reinstate him. These then invited Bāqī-Moḥammad to oust the khan. The Toqay-Timurid leader brought an army to Bukhara and at the battle of Bāḡ-e Šamāl in the spring of 1007/1599 he defeated the Bukharan force and killed Pīr-Moḥammad, ending a century of Shibanid sovereignty.

The Unified Khanate, 1007-20/1599-1612. The Toqay-Timurid clan immediately met in qoreltāy, elected Jānī-Moḥammad, the senior member after Yār-Moḥammad (who refused the khanate for a second time), regnant khan. Jānī-Moḥammad presided at the first Toqay-Timurid appanage distribution, at which Bāqī-Moḥammad as puissant khan received Bukhara. The new khan sat at Samarqand, the historic taḵt-gāh. The rest of the appanages were distributed in the spring of 1007/1599: Walī-Moḥammad, the next most powerful figure after Bāqī-Moḥammad received Sāḡarj; Pīr-Moḥammad, of unknown lineage, Orā Tīpa; Šahr-e Sabz (Kaš) went to ʿAbbās-Solṭān, and Ḵozār, a dependency of Bukhara, was given to Raḥmānqolī-Solṭān. Tashkent, Balḵ, Ḥeṣār-e Šādman, the Farḡāna region, and Badaḵšān were still outside Toqay-Timurid jurisdiction.

By and large the Uzbek amirs approved the new khanate as a worthy upholder of the Chingizid political constitution. At Balḵ the amirs installed ʿAbd-al-Amīn (see above), but there was no general amirid consensus on his khanate, and within a short period dissident amirs assassinated him and sometime early in mid-1007/1599 installed Moḥammad-Ebrāhīm, a man alternately styled a Janibegid (grandnephew of Jānī Beg) or a Toqay-Timurid (grandson of Yār-Moḥammad), who had been captured by the Qezelbāš in the 1006-7/1598 campaign and gone with Shah ʿAbbās to Isfahan. The shah apparently saw in Moḥammad-Ebrāhīm an instrument for his Khorasan policies in the wake of the recapture of Herat. There is little sign that the Balḵ amirs who conspired in support of Ebrāhīm, were in any way pro-Safavid. Instead the indications are that he was seen as far more capable than ʿAbd-al-Amīn. On Ebrāhīm’s