The Tibetan History Reader

When the Tibetan Empire reached deep into Inner Asia at the height of its expansion, what did it encounter? What effects did this encounter have upon the political and cultural history of Asia? Along with the concluding chapter to the seminal work The Tibetan Empire in Central Asia, the following article marks a rare attempt to put Tibetan history into a global perspective. Christopher Beckwith is ideally positioned to undertake this, being well-versed in Tibetan, Chinese, Arabic, and Turkic sources. He is concerned especially with a key geographic location in what is now Inner Mongolia: the southern part of the great bend of the Yellow River known as the Ordos. He argues here that the century-long occupation of this region by Tibetans and their allies, as well as their later enduring presence, formed a key bulwark against the spread of Western religions—Manicheanism, Nestorian Christianity, and Islam—into China. This connection helped ensure that the classical Tibetan language and Tibetan Buddhism came to dominate High Asia, stretching eventually from northern Xinjiang to Inner Mongolia, and from the Himalayas to Lake Baikal.

After the outbreak of the An Lushan Rebellion in Tang China in 755 A.D., the Tibetans took advantage of the resulting Chinese military weakness to recapture a vast stretch of Tibetan territory that had been occupied by the Tang during the preceding two or three decades.1 They did not stop at their old borders, however, but—perhaps seeking revenge—pressed on deeper into China, where they captured the capital, Chang’an, in December of 763.² It is not often appreciated that the eastern border of the Tibetan Empire then stayed fixed for a century at a point only a short distance to the west of the Chinese capital. This was the cause of constant worry to the Tang, since the Tibetans could and often did threaten the capital more or less at will. More important than this, however, is the fact that at the same time Tibet also controlled a vast territory further to the north and northwest of Chang’an, the borderlands between the northern steppes and the traditional Chinese realm south of the Great Wall. This Tibetan military domination of the southern Ordos and neighboring regions of northwest China during the late eighth and early ninth centuries seems to have had a long-lasting effect on the history of East Asia.

The Tibetan capture of Liangzhou in 764,3 and the consequent Tang loss of Hexi dao to Tibet meant that the direct routes from China to Central Asia (and, thus, the Western world) were all in Tibetan hands until nearly the end of the Tang dynasty. Japanese historians have long ago noted the significance of this fact. In an article published in 1956, K. Nagasawa argues that this event was a major turning-point for the history of East-West trade because, he says, the Tibetans held onto Liangzhou long after the rest of their Empire had broken up, and furthermore the Tanguts inherited the same area of control from the Tibetans, and kept it even longer.⁴ Tibetan control of the area meant—according to Nagasawa—that the bulk of Tang China’s silk exports had to go west via the so-called “Uighur route”: from North China via the Ordos or Taiyuan (in Hedong, present-day Shanxi) to Zhong Shouxiang cheng (the “Middle City for Receiving Submission”), which was located just north of the great bend of the Yellow River. From there the route passed northward to the Uighur capital on the Orkhon, and thence westward to the Arab caliphate.⁵ Although Nagasawa’s interpretation is basically correct, the story is somewhat more complicated, and his conclusions should be modified. In addition, while the debilitating effect on Tibet of the protracted warfare with the Uighurs and Chinese has been duly noted by nearly all writers on the subject, the effect of the same warfare on the Uighurs has received little attention. The Tibetan Empire’s movement northward from Hexi into territory once under the influence of (or actually controlled by) the Turks—the area from Hami to the Ordos—on the one hand had a great impact on the fate of the Uighur Empire, and on the other helped lay the foundations for the Tangut Empire. The eventual results of these changes were indeed of fundamental importance for later East Asian history.

The Tibetan expansion into the Ordos region seems originally to have been merely an extension of campaigns into the area about Chang’an and into Gansu, along the Silk Road into Central Asia.⁶ Through constant use of the Yellow River routes, the Tibetans ended up in an excellent position to raid the prefectures along the Great Wall, both north and south of it. They did so regularly: south from 763 on,⁷ and north from 778 onward.⁸ Most of the raids included large contingents of Tanguts, Tuyuhun, and others along with the Tibetans.⁹ Although the Chinese had settled some Tanguts and Tuyuhun in Guannei [inside the passes] and across the Wall in the southern Ordos—in order to keep them away from their former Tibetan overlords¹⁰—the Tibetan army apparently brought new contingents of these peoples with them from northeastern Tibet. After the Chinese refused to honor the Tang’s part of the bilateral agreement of 783–784 concerning payment to Tibet for military assistance against the rebel Zhuci and his Uighur allies (which Tibetan assistance was decisive in the rebels’ defeat),¹¹ the Tibetan army of Zhang Gyeltsen again entered to attack the prefectures to the near northwest of the capital district.¹² When this move was blunted, he turned northward to begin a campaign of conquest in the southern Ordos. On or about December 10, 786, Yanzhou was taken and garrisoned;¹³ by December 26 of the same year, Xiazhou, Linzhou, and Yinzhou were likewise taken.¹⁴ The Tang was duly alarmed; when Zhang Gyeltsen suggested a peace treaty, the Chinese snapped at the chance. By the summer of 787, everything was set for a treaty to be signed at Pingliang. But the Tibetans then turned the tables on the Chinese on July 8, 787, by capturing and carrying off a great number of the Tang generals and other attendants sent to the intended treaty-signing ceremonies.¹⁵ Immediately afterward, his goal largely accomplished, Zhang Gyeltsen ordered the Ordos fortresses destroyed and the garrisons withdrawn.¹⁶ So ended the first period of Tibetan forays into the Ordos proper.

The next stage followed almost without a pause. During the Tibetan-Uighur war over Tang-held Beshbalïq (Beiting, near present-day Urumchi) the Uighurs attacked and defeated the Tibetans at Lingzhou (present-day Ningxia).17 After the Uighurs captured (not, as often stated, “recaptured”) Beshbalïq from Tibet in 792, they pressed the Tibetans southwards, capturing from them Qocho (Gaocheng or Xizhou, in the Turfan Depression) in the same year.¹⁸ The Tibetans, who apparently still held Hami (Yizhou), counterattacked—recapturing Liangzhou and eventually pushing the Uighurs back to Qocho in the West. Tibet also sent armies again into the Ordos region. Although in 808 the Uighurs were able to take Liangzhou and possibly hold it for a short time,¹⁹ the Tibetans responded in the following year by sending 50,000 cavalry to attack an Uighur embassy on its return home from China, somewhere beyond Piti Spring (located north of Xi Shouxiang cheng, the “Western City for Receiving Submission”).²⁰ Tibetan pressure on the Uighurs’ most critical lifeline was such that in 816 a Tibetan raid is said to have reached a point only two days’ journey from the Uighur capital, Ordubalïq (now known as Karabalgasun).²¹ Tibet kept up the pressure, to the point where the Uighurs felt the need to boast to the Chinese in 821 that not only would they be able to protect the Taihe Princess on her way to the Uighur capital in Mongolia, but they would even send out 10,000 cavalry via Beshbalïq and 10,000 via Anxi, to push the Tibetans back.²² In fact, however, the Uighurs sent at most 3,000 men to the Chinese border near Fengzhou, and there is no record of any actual Uighur move against the Tibetans at this time. Moreover, it was undoubtedly only due to the conclusion of the Sino-Tibetan peace treaty of 821–822, and the Sino-Uighur and Tibetan-Uighur accords of the same year, that the Tibetans did not continue their raids in the Ordos region.²³ It is clear from these events that Tibetan influence then extended across the southern Ordos and the neighboring area south of the Great Wall, southwestward throughout the whole of Gansu and Hexi, and westward as far as Hami and Qocho. With the conclusion of the new treaty, the Tibetan military presence was theoretically restricted to the parts of that territory which were under actual Tibetan administration in 821. Finally, between 849 and 863 most of the Tibetan-ruled areas outside present-day ethnic Tibet were lost.²⁴

The most important immediate effect of the nearly century-long Tibetan presence in the Ordos and North China was the movement of a great number of people from areas further west or south into the borderlands of the northern steppe. Some of these people had fled from Tibetan control and had been moved by the Tang in order to keep them away from Tibet. Such were most of the Tuyuhun and Tanguts, moved in the seventh century,²⁵ and the Shatuo Turks, moved in the early ninth century.²⁶ The Shatuo and Tuyuhun were soon moved across the Yellow River to the east into what is now Shanxi and Hebei,²⁷ where they were eventually to become power brokers in Chinese dynastic politics, to help set geopolitical patterns followed by a long succession of northern Chinese dynasties. The Tanguts, however, unlike the Tuyuhun and (ultimately) the Shatuo, seem normally to have retained a close working relationship with the Tibetan Empire during its period of domination over them, and many Tanguts, remnants of the Tibetan armies, were apparently more recent arrivals from their homeland in the northeastern marches of Tibet. Together with some Tibetans who stayed behind, they continued to be active as rebels, bandits, or raiders of one kind or another, long after the conclusion of the international peace of 821–822 and the cessation of imperial Tibetan military activities in the area. The Tanguts’ power grew proportionately as that of the Tang declined, so that by the end of the latter dynasty they were for all practical purposes independent. The Tangut Empire established in the eleventh century—with a Chinese-style dynasty later known as Xixia—was territorially more or less a reincarnation of the former Tibetan zone of influence there. The Tangut Empire lasted until its conquest by the Mongols in the early thirteenth century.

The presence of the Tibetan Empire (and later the Tangut Empire) in the lands bordering northwestern China had several far-reaching consequences. One of these was the redirection of international trade: caravans to or from China were forced either to go through Tibetan territory or to go around by a very circuitous route through Uighur Mongolia, and so westward.28 (Seaborne commerce was not affected at all, except to be encouraged.) Another, perhaps more dramatic, effect of Tibet’s movement northward was its impact on the Uighur Empire. Tibet’s military presence in Hexi and the neighboring regions helped to separate Mongolia from Zungharia; thus, with simultaneous Tibetan pressure on both the southern Mongolian steppe and the more fertile and prosperous western part of their empire, the Uighurs ended up divided, apparently keeping the bulk of their forces in the West. The Tibetan raids into Mongolia were thus designed to divide, or at any rate had the effect of dividing, the Uighur Empire into two separate halves. The eastern part, with the capital, Ordubalïq, was eventually weakened to the extent that it was easily crushed by the Kirghiz in 840.²⁹ The destruction of Uighur rule in Mongolia meant the nearly total elimination of Western religious pressure, in the form of Manicheism, on Eastern Eurasia. The remaining followers of Manicheanism, and the numerically weak Nestorian adherents, ultimately disappeared in the face of another world religion, South Asian in origin, namely Tibetan Buddhism.

Of all the effects of the Tibetan expansion, the most long-lasting was not primarily political, but rather cultural in nature. The late Tibetan Empire, its successor states in Liangzhou and Siling (Xining), and finally the Tangut Empire, were all strongly Buddhist states, powerful enough to resist the encroachment of Islam. They allowed Buddhist clerics to pass through and cross-fertilize, for example between Tibet and the Buddhist centers on Wutai shan.30 Tibetan Buddhist activity in the colonial areas just outside Amdo continued uninterrupted after the breakup of the empire. There was quite a lot of Tibetan Buddhist activity in Shazhou (Dunhuang) even after the Chinese recapture of the city in 848;³¹ Tibetan monks were to be found in the armies of the two feuding ministers Lön Gungzher and Zhang Bibi, at the same time.³² In addition, it is well known that the refugee monks from Central Tibet, including eventually Lhalung Pelgyi Dorjé, the assassin of Langdarma, settled in Amdo after attempting to preach to the Uighurs (presumably those in Ganzhou); this presupposes no open hostility to Buddhism in the area. The Tibetan successor-states in Liangzhou and neighboring areas were pro-Buddhist. When the Tanguts finally occupied this region they simply continued to support an already long-established Buddhist church. Furthermore, Tibetan monks were quite active at the court of the Song dynasty in China, where they assisted in the translation of several important Buddhist texts into Chinese. When the Mongols finally supplanted the Tanguts, they did not disturb the existing Buddhist establishment; on the contrary, they supported it as strongly as their predecessors had. A crucial fact of Tibetan-Mongol relations, one generally overlooked, is that Qubilai, the Mongol Great Khan and founder of the Yuan dynasty in China, was raised by a Tangut nursemaid³³ and grew up in Tangut at the court of his influential cousin Köden. There he met Pakpa Lama, and the Tibetan-Mongol cultural alliance was soon firmly in place. It is clear that the Mongol ruling class (at least) wholeheartedly accepted Tibetan Buddhism by the end of the thirteenth century.³⁴

Tibetan culture thus was enabled to expand uninterruptedly northeastwards during the Tangut period and on through the Mongol and Manchu periods, with the eventual result that the dominant high culture of northeastern Eurasia, including (besides Tibet, Mongolia, Manchuria and parts of Siberia, and the southeasternmost corner of Europe) also parts of Northwest China (mainly in Gansu) was Tibetan Buddhist culture, not Chinese culture. It is no accident that foreign accounts of eastern Eurasia written in the nineteenth century liked to refer to Tibetan as the “Latin” of what they called “High Asia”—meaning the vast area dominated by Tibetan Buddhist culture. Chinese and its attendant literary culture was, like Hindustani in India, the province of native scholars, with very few exceptions, whereas Tibetan was the common language of scholars from Tibet, Mongolia, Siberia, Manchuria, and China. One could hardly imagine this happening if the late Tibetan Empire, its successor states, and then the Tangut Empire—the latter a multinational state including Tibetans as one of its most important components—had not maintained a strong Buddhist bulwark against the powerful forces of Islamic expansion that were then eliminating Buddhism from East Turkistan.

In conclusion, the Tibetan Empire’s expansion into the Ordos and northwestern Chinese borderlands was merely the beginning of a much greater Tibetan Buddhist cultural empire that continued to spread from Tibet, eventually to dominate nearly the whole of northeastern Asia well into the twentieth century.

1. The details of this struggle, and of the topic of the present paper, are properly the subject of a thorough book-length monograph. Here, primary-source reference will be made only to Sima Guang, Zi zhi tong jian [Comprehensive Mirror to Aid in Governance], 10 vols. (Beijing: Zhonghua shuju, 1956; reprinted: Taipei, 1979): the single most important source for Tang-Inner Asian history, though the least utilized. The first Tang armies in Tibet’s northeast fell in the twelfth month of 756; see Sima, Comprehensive Mirror to Aid in Governance, 219:7011. For citations of other sources, and a more detailed historical narrative (which omits most of the material on the present subject) see my book, The Tibetan Empire in Central Asia: A History of the Struggle for Great Power Among Tibetans, Turks, Arabs, and Chinese During the Early Middle Ages (Princeton: Princeton University Press, 1993).

11. Ouyang and Song, New Tang Annals, 231–7442; for details, see the discussion in Beckwith, The Tibetan Empire in Central Asia. It is notable that the Uighurs were actually at war with China, or there were very hostile relations between the two nations, quite often during the period of the Uighur Empire’s existence, specifically, ca. 745–756, 764–765, 775–787; the Uighurs raided or threatened the border in 796, 813, 822, and 840. It was, nevertheless, deliberate Tang policy to cultivate the Uighurs, probably because the Turks, unlike the Tibetan Empire, were no real danger to a united China; they were never able to penetrate very far into the country, nor hold any territory; moreover, they were separated from China by the Gobi.

28. On international trade during this later period, see the valuable article by Yoshinobu Shiba, “Sung Foreign Trade: Its Scope and Organization,” in China Among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries, ed. Morris Rossabi (Berkeley: University of California Press, 1983), 89–115. It is important to realize that the fragmented polities of the area encouraged international trade, and made it possible by providing numerous alternative routes. It is notable that one important route for East-West trade passed through the Qingtang kingdom in northeastern Tibet. See Shiba, “Sung Foreign Trade,” 100–102.

29. Internecine conflict within the Uighur Empire was primarily responsible for the collapse. Perhaps the Kirghiz, former allies of Tibet, merely succeeded in taking advantage of a situation created by the Tibetan strategists, who were finally unable to do the job themselves. For subsequent events, particularly the fate of the Uighurs who fled to China, see the unpublished dissertation of Michael Drompp, “The Writings of Li Te-yü as Sources for the History of T’ang-Inner Asian Relations,” Ph.D. diss., Indiana University, 1986.

30. Already in the early eighth century under Mé Aktsom Tibetan Buddhists are said to have established a connection with the Buddhist centers on Wutai shan (Tib. Ri wo rtse lnga), a mountain sacred to Manjuśrī. According to an account in the Bazhé (Sba bzhed) connected to the material on the building of Samyé—and therefore, it would seem, basically reliable—Ba (Sba) Sangshi visited Wutai shan during a trip to China, and when he returned built a small temple, called Nang Lhakhang, “Inner Temple,” in the imperial palace precinct at Drakmar. The temple was built “in the shape of Wutai shan.” See C. Beckwith, “The Revolt of 755 in Tibet,” Wiener Studien zur Tibetologie und Buddhismuskunde [Contributions on Tibetan Language, History, and Culture] 10, vol. 1, ed. Ernst Steinkellner and Helmut Tauscher (Wien: Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien, 1983), 13. For further discussion of Sangshi and his journey, see Jeffrey Broughton, “Early Chan Schools in Tibet,” in Studies in Ch’an and Hua-yen, ed. Robert M. Gimello and Peter N. Gregory (Honolulu: University of Hawai‘i Press, 1983), 1–68, especially 5–7 and notes. In 824, during the reign of Relpachen, as is well known, Tibet formally requested the Tang government for a map of the holy mountain; the request was granted. See Paul Demiéville, Le concile de Lhasa [The Council of Lhasa] (Paris: Institut Hautes Etudes Chinoises, 1952), 188 n.1.

34. See C. Beckwith, “Tibetan Science at the Court of the Great Khans,” Journal of the Tibet Society 7 (1987): 5–11. It should not be overlooked that Tibetan Buddhism had two good chances to be established in the West as well, first through the strong patronage of the Mongol Ilkhans in Persia, and later through the patronage of the Zunghars, whose Kalmyk descendants living on the European shores of the Caspian Sea still follow Tibetan Buddhism. The History of Ilkhanid-Tibetan relations—especially the question of the Tibetan bakhshis at the Ilkhanid court, and their influence on non-Buddhist religious beliefs and practices—is a potential goldmine. Besides Tibetan material, there is much in Islamic sources, apparently some in Greek, and probably some in various other languages, waiting to be explored.