Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ASIAN HISTORY ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 28 June 2017

The Uyghur Empire (744–840)

Abstract and Keywords

The Uyghurs (Chinese Huihe迴 紇, Huihu回鶻) were a pastoral nomadic people living in the region of the Selenga and Orkhon river valleys in modern Mongolia; they spoke a Turkic language. The empire that they created on the steppe lasted for nearly a century (744–840) and played an important role, both politically and culturally, in East Asia. Centered on the Mongolian Plateau, the Uyghur Empire at its height controlled numerous other peoples within a territory that included lands to the north in the modern regions of Tuva and Buryatia, as well as some parts of the northern Tarim Basin and eastern Inner Mongolia.1 During its eventful history, the Uyghur Empire sent cavalry to help the Tang Dynasty put down the An Lushan rebellion, maintained strong political and economic ties with China, fought with the Tibetan Empire for control of important international trade routes, built cities on the steppe, celebrated its rulers’ achievements in stone stelae, and—uniquely in the world—adopted Manichaeism as its state religion. After their empire collapsed, the Uyghurs developed new polities in Gansu and the Tarim Basin that continued to exercise influence in Inner Asia.

Keywords: Uyghurs, Mongolia, Inner Asia, nomads, steppe, Tang Dynasty, Sogdians, Tibetan Empire, An Lushan

The Early Uyghurs

Prior to the establishment of their empire, the Uyghurs appear in Chinese historical records under various names. In the 6th century, they became subjects of the First Türk (Chinese Tujue突厥‎) Empire (552–630) and enjoyed some prestige within that empire’s administration, governing the “wild regions of the north” in the name of their Türk overlords. In 627, when the First Türk Empire was weakening, the Uyghurs revolted together with two other subject peoples, the Bayïrqu (Chinese Bayegu 拔也古‎, with variants) and the Xueyantuo 薛延陀‎. After the political collapse of the Türks a few years later, the Uyghurs continued to exercise their power and, especially after defeating the Xueyantuo in 646, dominated many of the other peoples of the region. This victory had been achieved with support from the Tang Dynasty (618–907) in China, marking an early connection between the Uyghurs and the Tang Empire. From this position of strength, the Uyghurs continued to collaborate with the Tang in a partnership that included coordinated military campaigns against various groups of Türks as well as the Goguryeo kingdom of northern Korea/southern Manchuria in the 650s. Indeed, until the Türks succeeded in throwing off Tang control and established the Second Türk Empire (682–744), the Mongolian Plateau and its environs were largely dominated by the Uyghurs for nearly four decades.2 Not enough is known of this period, however, to say much about the Uyghurs and their political organization during this interregnum. It does appear that they accepted close ties with the Tang, even being named the Hanhai Prefecture within the Tang administrative organization, although the precise level of Tang influence over the Uyghurs is uncertain, and there was at least one conflict between them in 661.

After the Türks reestablished control over Mongolia in 682, the Uyghurs were once again subordinated to them, although it appears that at least some Uyghurs moved into the region of Gansu under Tang administration for a time before being brought under Türk control. The power of the Second Türk Empire began to wane after about half a century, however, when the Türk ruler Bilge (Chinese Pijia毗迦‎) Qaghan (r. 716–734) was poisoned in 734. Even before that time, the Uyghurs and other peoples had begun to chafe under Türk rule. After Bilge’s predecessor and uncle Qapghan (Chinese Mochuo默啜‎) Qaghan (r. 691–716) was killed by subordinate peoples and his head sent to the Tang capital, the Uyghurs grew more independent and powerful. After Bilge’s death, his several successors ruled only briefly, and many Türk and Sogdian elites fled from the steppe to China to avoid the chaos. The Uyghurs allied with two other subject peoples, the Qarluq (Chinese Geluolu葛邏祿‎) and the Basmïl (Chinese Baximi拔悉密‎), to kill the last Türk qaghan. The Uyghurs and Qarluq then attacked the Basmïl. After they had been subjugated, the Uyghurs turned on their erstwhile allies and became the sole ruling power on the Mongolian steppe.

Foundation of the Uyghur Empire and Uyghur Involvement in Tang China

The Uyghur ruler who oversaw the foundation of the empire in 744 was a man of the Yaghlaqar (Chinese Yaoluoge藥羅葛‎) family known as Qullïg Boyla (Chinese Guli Peiluo骨力裴羅‎) Qaghan (r. 744–747).3 Although he died shortly after the empire’s establishment, his son, known in Chinese sources as Bayan Chor (Chinese Moyanchuo磨延啜‎, r. 747–759), ruled for more than a decade. Many scholars consider him the actual founder of the empire. The work of this father-and-son team helped to place the new empire on a solid footing. The Uyghur Shine-usu inscription describes Bayan Chor’s defeat of the Basmïl and Qarluq. Although the former became part of the Uyghur polity, a significant group of the latter moved westward into the region of Jungaria (around Lake Balkhash and Issïq-Köl) to maintain their autonomy. The same inscription notes that Bayan Chor built a capital city, the ruins of which can still be seen in the Orkhon River valley, as well as another city, known as Baybalïq, “Rich City” (Chinese Fuguicheng富貴成‎), built in the Selenga River valley by and/or for the Chinese and Sogdians within the empire. Although the sources of the period nowhere give us the capital city’s Uyghur name, a later source (the 13th-century Persian historian ‘Alā’ al-Dīn ‘Aṭa Malik Juwaynī) indicates that it was called Ordubalïq, “Royal Court City.”4

The early years of the empire were marked by frequent campaigns to neutralize other peoples and, in many cases, bring them under Uyghur control. The Uyghur Empire thus was from its inception a grouping of confederated tribes, including both willing and (at least initially) unwilling participants. Although many of these peoples spoke Turkic languages, it can be assumed that some did not. Furthermore, as the Shine-usu inscription attests, there were Sogdian and Chinese minorities living within the empire, some of whom achieved high status. At least one Chinese family was allowed to take the royal clan name of Yaghlaqar, replacing their traditional Chinese surname of Lü. As with most Inner Asian empires, the history of the Uyghur Empire indicates that subordinate peoples sometimes rebelled against Uyghur domination, causing both the power and the geographical extent of the empire to wax and wane.

The Uyghurs resembled their steppe predecessors in a number of ways. As pastoralists, they derived their livelihood from their herds (primarily horses and sheep, as well as goats, camels, and cattle) and engaged in seasonal nomadization, supplemented by hunting and some agriculture. They spoke a language that was identical to that of the Türks and wrote it in a runiform or “runic” script that the Türks had developed. The Uyghur rulers engaged in military campaigns to extend their power and celebrated their achievements through the creation of stone stelae with laudatory inscriptions. Their officials carried the same titles as had those of the earlier empire. In 744, the Uyghur ruler took the supreme title of qaghan (Chinese kehan可汗‎), used by the rulers of the Türks as well as those of many other Inner Asian polities, after defeating the Basmïl; this title was recognized by the Tang Dynasty in 746. In their use of imperial symbols the Uyghurs even employed wolf’s-head standards that harkened back to the Türks’ own foundation myth. To their neighbors, the Uyghurs would have seemed very much like the Türks; indeed, the Uyghur Empire has been called a “Third Türk Empire,” although this is problematic because the Uyghurs clearly saw themselves as having an identity distinct from the Türks. Furthermore, the similarities that have been noted were in some ways overshadowed by important political events that made the contours of Uyghur history and life ultimately quite different from those of the Türks.

The Rebellions of An Lushan and Pugu Huaien

These events began just over a decade after the Uyghurs had established their empire, when China was thrown into chaos by the rebellion of the frontier general An Lushan who declared the founding of a rival dynasty and set out to overthrow the Tang in 755. Although better known for its dramatic and lasting effects on the Tang Dynasty, the An Lushan rebellion had a significant impact on the Uyghur Empire as well. Finding itself in a desperate situation, the Tang court sought foreign assistance to help put down the rebellion; the Uyghurs responded to this call by sending troops, which played a pivotal role in the conflict. After a brief skirmish with the Tongra (Chinese Tongluo同羅‎), a Turkic people who had allied with An Lushan in 756, Bayan Chor deputed his eldest son to lead some 4,000 cavalry to China in the summer of 757. These troops helped turn the tide, allowing Tang forces to recapture the dynasty’s two capitals, Chang’an and Luoyang, and led to the rebellion’s ultimate collapse. But the Uyghur collaboration with the Tang court was not an easy one. After taking the western capital of Chang’an, the Uyghurs had to be offered additional inducements to continue on to Luoyang, the eastern capital. The Uyghurs were allowed to plunder Luoyang for three days.

Bayan Chor’s death in 759 did not stop the Uyghurs’ involvement in China’s civil conflict, which continued under his successor, Bügü (or Bögü, Chinese Mouyu牟羽‎) Qaghan (r. 759–779). Bügü was at least as difficult an ally as his father had been. When the rebel leader Shi Chaoyi sent news to the Uyghurs of the death of the Tang emperor Suzong (r. 756–762) in May 762 and encouraged Bügü to attack the Tang, the qaghan seriously considered this option. Although a new emperor (Daizong, r. 762–779) was soon enthroned in China, it was not until a Turkic general in Tang service, Pugu Huaien, who was also Bügü’s father-in-law, intervened that Bügü was convinced to return to the earlier alliance, helping the Tang to finish the destruction of the rebel forces by 763. Once again, the cost of Uyghur support was paid by the populace of Luoyang, which was plundered a second time.

The Uyghurs’ decision to respond positively to the Tang Dynasty’s request for help was to prove of fundamental importance in the development of their own empire. Their reasons for doing so are nowhere made explicit, but it has been argued that the Uyghurs were eager to destroy elements of the Türks, who had fled to China when their state collapsed.5 Possibly, the Uyghurs saw this as an opportunity to obtain leverage over China and exercise further power in this way, much as the Eastern Türks had done when the Sui Dynasty (581–618) had collapsed and been replaced by the Tang.

Uyghur involvement in the An Lushan rebellion had many consequences. First, the Uyghur qaghans enjoyed a level of prestige vis-à-vis China that earlier steppe rulers typically had not. This arrangement was parlayed into two particularly advantageous factors, one political and the other economic. The first was the marriage of three Tang imperial princesses to different Uyghur rulers, indicating an unprecedented degree of Chinese respect for the qaghans. The second was the profitable trade between China and the Uyghurs in which the former were required to pay high prices in silk for the latter’s horses. Although the Uyghurs were regularly granted gifts of silk by the Tang court, the trade network that Uyghur assistance had assured was far more significant. The wealth engendered by this trade helped the Uyghur rulers to build cities and defenses within their realm and enjoy a high standard of living that was bolstered by this unusual level of commerce.

The Uyghurs soon were once again involved in Chinese politics when General Pugu Huaien, who had helped put down the An Lushan rebellion, himself turned on the Tang Dynasty; through his marital connections to the Uyghur court, he immediately sought Uyghur assistance for his rebellion. Turning their backs on the Tang, the Uyghurs sent troops to help him in 764 under the leadership of their high official Ton Bagha (Chinese Dun Mohe頓莫賀‎); Bügü Qaghan remained in Mongolia. Pugu Huaien enjoyed the support of both Uyghur and Tibetan troops, but his death in September 765 caused the Uyghurs ultimately to do another about-face and link with the Tang general Guo Ziyi to repel the Tibetans, which they did successfully. The Uyghur connection to the Tang Empire thus was restored.6 Their brief support of Pugu Huaien may have had more to do with the qaghan’s marital relationship to Pugu Huaien than to any particular anti-Tang sentiment, or perhaps reflects a Uyghur desire to increase their influence in China.

At the same time that they were engaged in China, the Uyghurs were expanding their power in other directions. In 754–756, a western campaign advanced into the region around the Tarim Basin. At approximately the same time (ca. 756–759), the Khitans (Chinese Qidan契丹‎) to the southeast submitted to the Uyghurs, who appointed overseers for them. The Uyghurs also campaigned to the northwest, defeating a large Kirghiz (Chinese Xiajiasi黠戛斯‎, with variants) force in 758. The Kirghiz moved further to the northwest and the Uyghurs constructed at least eighteen defensive fortifications, the remains of which can still be seen, along their northern frontier in modern Tuva. This complex of Uyghur constructions in Tuva includes a walled compound built on an island in the lake called Tere-khol’ in southern Tuva. This site, known as Por-Bajin (Por-Bazhyn), has been dated to the late 8th century, probably ca. 770–790.7 The Uyghur settlement at Por-Bajin shows a strong influence of Tang building materials and techniques, which are found in other Uyghur constructions as well. From the available textual and archaeological evidence, it seems that the Uyghurs employed a syncretic approach to their construction projects, including both Chinese and Sogdian influences.8

The Uyghurs and Manichaeism

At about this time, Bügü Qaghan began to patronize the Manichaean religion. Manichaeism—which has an elaborate dualist theology and demands vegetarianism for its “elect”—seems a peculiar choice for a steppe empire in which meat and animal products played such an important role in the nomads’ diet. According to the Chinese text of the Karabalghasun inscription, while he was in Luoyang Bügü Qaghan encountered Sogdian Manichaean clerics who were well-versed in Manichaean doctrine. Modern scholars have assumed that it was this encounter, which they properly connect to the qaghan’s 762–763 campaign, which led to Bügü Qaghan’s conversion. But what seems odd about this interpretation, which is nowhere supported directly by the available sources, is the fact that Sogdians lived in the Türk and Uyghur Empires long before this happened. The qaghan was certainly aware of the Sogdians within his own realm as well as their culture, which included Manichaeism. In fact, the Uyghur contingent that he led to help defeat the rebels in China in 762 included at least two generals with names indicating their Sogdian background: An Ke and Shi Diting. Both played important roles in the qaghan’s campaign into China.9

Bügü Qaghan’s conversion to Manichaeism thus may have been earlier than 762–763, or perhaps at least his encounter with the religion began at an earlier date.10 As for his reasons for choosing Manichaeism, he may have sought a unifying force that would help bind the diverse peoples of his empire together. Aware that the Second Türk Empire had fallen to the revolt of subject peoples (including the Uyghurs themselves), he also may have sought a new type of religious legitimacy to help combat such centrifugal forces. A further benefit was that Manichaeism was not a Chinese religion, so its adoption would not suggest subservience to China. Or it may simply be that the clerics Bügü encountered in Luoyang were particularly learned and persuasive; the Karabalghasun inscription indeed supports this interpretation. Although the date of his conversion thus remains uncertain, clearly, Bügü Qaghan adopted this syncretic Iranian faith. Evidence from an Uyghur source indicates that the religion was not universally welcomed within the empire; Bügü agreed to redouble his efforts to protect and promote the faith.11

Uyghur rulers were not content simply to act as patrons of Manichaeism within their own realm; in 807 the Uyghur qaghan put pressure on the Tang court to establish Manichaean temples within the Tang Empire. The Tang government had previously restricted the practice of this religion to foreigners, but things changed with Uyghur patronage. In 768, Manichaeans received the court’s permission to build temples in Southern China; sources also indicate the presence of such temples in Chang’an. In 807, the Tang official Bai Juyi wrote a letter to the Uyghur Qaghan Qutlugh (Chinese Guduolu骨咄祿‎) agreeing to the building of Manichaean temples in Luoyang and the northern city of Taiyuan.12 Ultimately, it is difficult to determine how deeply Manichaeism penetrated to the general population of the Uyghurs and other peoples within the empire. The religion apparently did not last long on the steppe after the collapse of the empire in 840—an event which led to its persecution in China as well. Despite Tang efforts to uproot the Manichaean church in China, however, it managed to survive there.

Despite the Uyghur rulers’ general support of Manichaeism, some evidence indicates that the Uyghurs continued to employ the services of shamans. For example, Chinese sources state that in 765, during the rebellion of Pugu Huaien, the Uyghurs consulted shamans who prophesied that the Uyghurs would not do battle with the Tang and would return home after meeting a “great man,” Guo Ziyi. Soon thereafter, when the Uyghurs had switched sides and were fighting the Tibetan forces allied with Pugu Huaien, Uyghur shamans employed weather magic to create wind and snow that helped defeat the Tibetans. Many years later, the Arab traveler Tamīm ibn Baḥr, writing around 821, also noted the presence of “rain stones” that were employed by the Uyghur ruler to engage in weather magic.

The Continuing Uyghur–Tang Connection

After the collapse of Pugu Huaien’s rebellion, the Uyghurs re-established good relations with the Tang Dynasty. This began with an embassy to Daizong’s court in 765. Of the six persons comprising the leadership of this embassy only one, a “chieftain” named Shi Yena, is mentioned; his surname reveals his Sogdian background. The revived trade that the Uyghurs enjoyed with the Tang Empire enriched the Uyghurs dramatically, allowing them the wealth necessary not only to engage in the building of urban centers but also to maintain the loyalty of their subjects and support the state-sponsored Manichaean church and its clergy. The Tang Dynasty’s weakness after the An Lushan debacle, coupled with the Uyghurs’ ability to assert the moral high ground through their actions that helped to preserve the dynasty, lubricated this trade. Tang officials grumbled about the horse–silk trade and complained repeatedly about the poor quality of the horses they were obliged to purchase. As for the Uyghur Empire, this connection led to many Uyghurs and Sogdians travelling to, and even living in, the largest cities of China where they engaged in trade (including the purchase of tea) and moneylending. On the Uyghur side, furs from the forested regions of their empire were also an important part of their trade with China and other nations. Finally, it is worth noting that the Chinese were aware of Uyghur spies who relayed information on Tang political events back to Ordubalïq. It seems likely that these spies were Uyghur and Sogdian merchants and moneylenders living in Tang cities.

Within the Uyghur Empire, the wealth that trade engendered, and the stability that the empire generally enjoyed, caused people to engage in trade with other regions. It also helped promote agriculture, which was also necessary for the support of Manichaean elites. Modern archaeological techniques have recently affirmed Tamīm ibn Baḥr’s description of Ordubalïq as a city surrounded by cultivated fields.13 Uyghurs who had traded their nomadic ways for a more sedentary life resided in the empire’s urban centers. It is, however, difficult to measure such transformations in any meaningful way. While evidence of settled populations, agricultural activity, handicraft production, and urban markets is clear, it is difficult to ascertain how many persons were identified with such things or what their identities were. The sedentary component of the empire may have been largely comprised of Sogdian and Chinese populations, but it is likely that at least some Uyghurs were drawn to the settled life of the towns. And despite the urban court that the Uyghur rulers established, they themselves maintained a symbolic “nomadic” life in a magnificent golden tent, noted in Chinese accounts as well as that of Tamīm ibn Baḥr, which became an important marker of royal power.

The Uyghur-Tang connection was sufficiently important that when Bügü Qaghan contemplated an attack on China, he was killed in a coup led by Ton Bagha, who seized the Uyghur throne. This event was connected to the increased influence of Sogdians at the Uyghur court, who had encouraged the attack. Ton Bagha was determined to avoid damaging the lucrative trade with the Tang Empire, so he engineered his coup in 779 that not only led to the death of Bügü Qaghan but also of many of his Sogdian advisors. Bügü Qaghan seems to have been testing the waters; Chinese sources note that there had been an Uyghur raid on Tang territory in 778. Although some scholars have regarded Ton Bagha’s coup as not only pro-Tang but also anti-Sogdian and anti-Manichaean, it is striking that both were soon once again influential within the empire. However, the coup was likely aimed at stopping Bügü’s plans to invade China and eliminating those persons, Sogdians or otherwise, who had supported such an action. Ton Bagha Qaghan remained on friendly terms with the Tang; a brief Uyghur involvement in support of Zhu Tao’s anti-Tang rebellion in 784 appears to have been the action of a regional leader and not directed from, or sanctioned by, the Uyghur court.14

As noted previously, part of the Uyghur-Tang connection was the marriage of three Tang imperial princesses to Uyghur rulers; this was a sign of the great favor the qaghans enjoyed in the eyes of the Tang emperors. Unlike other Chinese princesses sent to wed foreign rulers during the Tang era and earlier, these women were the daughters of emperors—not the collateral relatives, etc., typically given the title “princess” and then sent to engage in diplomatic marriages with foreign rulers. The first of these imperial daughters was the Ningguo Princess, daughter of the emperor Suzong, who wed Bayan Chor in 758; she died in 791 after more than three decades living in the Uyghur Empire. The second was the Xian’an Princess, daughter of the emperor Dezong (r. 779–805), who married Ton Bagha in 788; she lived among the Uyghurs until her death in 808. The third imperial marriage was that of the Taihe Princess, daughter of the Tang emperor Xianzong (r. 805–820); her brother, the emperor Muzong (r. 820–824), agreed to betroth her to the Uyghur ruler in 821; she remained among the Uyghurs until the empire’s collapse.15 In addition to these imperial princesses, two daughters of Pugu Huaien were married to Bügü Qaghan; he wed the first before becoming qaghan. Many Inner Asian peoples maintained a practice in which widows were wed to male relatives of their former husbands, particularly sons—so long as those were not the women’s own biological sons. This was certainly the case for the Xian’an Princess, who enjoyed the honor of being qatun not only during the reign of Ton Bagha Qaghan but also the next three Uyghur rulers. There is insufficient information in this regard for the Taihe Princess.

As another show of favor, the Tang court regularly granted official titles to the Uyghur qaghans. These included long formulaic titles in Old Turkic, which are often so similar that it is difficult to distinguish one qaghan from another. These Turkic titles were certainly created and taken by the Uyghur rulers themselves. The Tang court affirmed these and added to them laudatory titles of two or four characters in Chinese. These grand titles were intended to enhance Uyghur prestige throughout the region, showing the favor they enjoyed from the Tang court and its recognition that the Uyghurs were the paramount power in North Asia.

Relations with Tibet

Uyghur relations with the Tibetan Empire, which had risen to power in the first half of the 7th century, were often hostile, and trade played an important role in the continuing conflict. With the outbreak of the An Lushan rebellion, the Tibetans had seized the opportunity to push deep into Tang territory, occupying parts of Qinghai and Gansu and even briefly seizing Chang’an for about two weeks late in 763. Although the capital was restored to the Tang government by Guo Ziyi and his forces, Tibetan incursions persisted for more than a decade after that.

Farther west, the Tang Empire’s control over Gansu and the Tarim Basin had seriously diminished, also as a consequence of the rebellion; the Uyghurs and Tibetans both eyed this region with great interest. This interest was due not only to strategic reasons but also to trade going through the region, which could be controlled and taxed by whichever power managed to dominate it. The An Lushan rebellion (which came hard on the heels of a Tang defeat at the hands of an Arab coalition at Talas in 751) had caused the Tang Dynasty to withdraw most of its garrison troops from Gansu and the Tarim Basin, and the Tibetans quickly attempted to extend their power into that region. They also raided the Ordos region and occupied several cities there in 786. Efforts to establish a Tang–Tibetan treaty proved fruitless, and the threat of Tibetan power led the Chinese chief minister Li Mi to propose a Tang alliance with the Uyghurs and other powers to weaken the Tibetan threat. Although the emperor Dezong initially rejected the idea of an alliance with the Uyghurs, he ultimately acquiesced to this plan, which was cemented with a marriage alliance.16 In the end, it did not severely diminish the Tibetans’ ability to threaten the Tang Empire.

The Uyghurs also sought to exercise power in the Tarim region, leading to conflict with the Tibetans. Much of this conflict centered on the northern Tarim cities of Beshbalïq (Chinese Beiting北庭‎), which changed hands more than once, and Qocho (Chinese Gaochang高昌‎). A Tibetan victory in 790–791 marked the end of significant Chinese presence in the region for nearly a millennium.17 Although the Uyghurs managed to dislodge the Tibetans from some of their strongholds in the Northern Tarim in 792, the sparseness of records for these events makes it difficult to comment definitively on the level of Uyghur control or influence there at any particular time, but it does seem that they held Beshbalïq and other cities in the region from 792 until their state’s collapse. Conflicts between the Uyghurs and Tibetans continued regularly for several decades. There was, for example, a major Uyghur attack on Tibetan forces near Turfan in 813; just a few years later, the Tibetans threatened Ordubalïq in 816. Such conflicts were destabilizing and costly; peace was finally established in 822/823, suggesting that neither polity believed that further warfare would gain them much. A successful Tang–Tibetan treaty was established at the same time. The Uyghurs’ willingness to make peace with their old enemies is undoubtedly related to both the easing of relations between the Tang and Tibetan courts as well as the heating up of Kirghiz hostility toward the Uyghurs at about the same time.

The Empire’s Later Years and Collapse

After Ton Bagha Qaghan’s death there was a period of instability during which the next two qaghans sat on the throne only briefly before being assassinated. In 790, a new qaghan, known only by the Chinese name Achuo阿啜‎, succeeded to the throne and held it until his death in 795. During this period the empire weakened, and the Qarluqs in particular managed to exert their strength and seize some Uyghur territory. After Achuo died, the throne was taken by one of his ministers—a man not of the royal Yaghlaqar family, but of another Uyghur family known as Ediz (Chinese Adie 阿跌‎, with variants). This is very likely the Uyghur minister and military leader known by his title of il-ögesi (Chinese xieyujiasi頡于伽思‎) who dominated the imperial government during the short reigns of Ton Bagha’s two successors. When he became qaghan, he adopted the royal surname of Yaghlaqar, showing its importance to Uyghur legitimacy. This qaghan, after some failures prior to his ascension to the throne, finally dislodged the Tibetans from Beshbalïq in 791–792 and recovered much of the empire’s strength and territory, enjoying significant influence throughout much of the Tarim Basin and beyond.

The significance of Sogdians and Manichaeans during the period after Ton Bagha’s rule can be seen in a number of factors, including the text of the Karabalghasun inscription18 and the role of Sogdians in Uyghur embassies and trade missions to China.19 Other documents attest to the continuing importance of Manichaeism among Uyghur elites, some of whom had Sogdian Manichaean names. The Sogdians’ importance as merchants can also be seen through many examples. During the reign of Bügü Qaghan, when they were enjoying their first wave of influence among the Uyghurs, Sogdian merchants regularly traveled to China with Uyghur envoys, and many of them remained at the capital city of Chang’an where they became quite wealthy. After Ton Bagha’s 779 coup, these Sogdians worried about returning to the Uyghur realm because Ton Bagha had put many Sogdians to death. Staying in China proved no safer, however, as they were soon massacred—along with many Uyghurs, including Ton Bagha’s own uncle—on the order of a Tang military official named Zhang Guangsheng. Later, beginning in the early 9th century, Uyghur envoys to China regularly included Manichaean clerics, who also engaged in commerce while there. Chinese sources note that the Uyghur qaghans frequently involved them in state affairs. Manichaeans were important in the back-and-forth exchanges between the two empires; they sometimes led Uyghur embassies to China, and they were also entrusted by the Tang government to convey important messages to the Uyghur court.

Around 820, the Uyghur Empire began to weaken. The primary causes of the empire’s decline were political factionalism within the ruling elite and the revolt of subject peoples. As noted previously, the Turkic Kirghiz became increasingly bellicose at about this time, leading to some two decades of constant struggle. Bloody political factionalism occurred at the Uyghur court throughout the 830s, beginning with the assassination of Qasar (Chinese He-sa曷薩‎) Qaghan in 832. The sources do not indicate the causes of this factionalism, but it was severe enough to weaken and ultimately destroy the state. Qasar Qaghan was succeeded by his nephew, known only by the Chinese name Hu胡‎ (or Sa薩‎) Tegin Qaghan (r. 832–839), who could not restore order. He successfully foiled a plot against him being led by an Uyghur prince and a minister of Sogdian ancestry, An Yunhe, who was executed for his involvement in the coup. In 839, another Uyghur minister named Küräbir (Chinese Jueluowu掘羅勿‎) led Shatuo Türk troops to attack Hu Tegin’s camp, and he committed suicide. Although another Uyghur qaghan was then enthroned by Küräbir, the situation remained perilous; an Uyghur general named Külüg Bagha (Chinese Julu Mohe句錄莫賀‎), who had opposed Küräbir’s coup, fled to the Kirghiz. After a hard winter (839/840) in which heavy snows caused the deaths of many herd animals, leading to famine and disease among the Uyghurs, Külüg Bagha attacked the Uyghur court with a large force of Kighiz cavalry and killed the new qaghan and Küräbir. Ordubalïq was now in the hands of the Kirghiz, as was the Taihe Princess, whom they had captured during their assault.

Although it can be assumed that some Uyghurs were killed or absorbed by the invading Kirghiz army, many fled from the Mongolian Plateau. A large number moved southwestward into the regions of Gansu and the Tarim Basin, where they were able to establish themselves and remain a significant cultural and political force. Some Uyghurs remained among the Khitans and one family, which had taken the Chinese name of Xiao, provided empresses for most Khitan rulers. Finally, two large groups of refugees fled to the Tang frontier. The Tang court played them against one another, eventually admitting one but attacking the other; the latter group was led by a new ruler, Ögä (Chinese Wujie烏介‎) Qaghan, who was, however, not universally accepted as sovereign. In the end, both groups were largely annihilated. Those who had hoped to restore Uyghur power were seen as a threat to the Tang, as were those who came into Tang territory with the plan to resettle in the frontier region. Neither ultimately bowed to Tang demands, so in the end most were killed, with only a few elites remaining. The Taihe Princess, who had been recaptured by the Uyghurs, was rescued. Also, in the winter of 842–843 a grand general of Sogdian background named Cao Mani (Chinese Cao Moni曹摩尼‎) sought asylum from the Tang government along with some 30,000 persons, including several other notables.20

The collapse of the Uyghur Empire relieved some pressures for the Tang Dynasty, particularly as the Kirghiz did not succeed in establishing an empire in the Uyghurs’ old realm that nestled against Tang territory; rather, they remained focused on their traditional homeland in the region of the upper Yenisei River, and their contacts with the Tang Dynasty were at best sporadic and not terribly significant. No powerful state emerged on the Mongolian Plateau in the years following the collapse of the Uyghur Empire.21 The Khitans shifted their allegiance to the Tang; after the collapse of the Tang polity early in the 10th century their independence (and power) increased. When the Khitans’ Liao Empire extended its influence into Mongolia in 924, they encountered little organized resistance (although some earlier historians have assumed that the Khitans must have encountered—and defeated—the Kirghiz, but this is incorrect). Indeed, the Mongolian steppe remained politically disorganized until the rise of the Mongols in the 13th century.

Within China, the collapse of the Uyghur Empire quickly led to the suppression of Manichaeism in 843 under the emperor Wuzong (r. 840–846) and his chief minister Li Deyu; Chinese rhetoric clearly links Chinese tolerance for Manichaeism to Uyghur power, and the persecution of the religion to the Uyghur Empire’s collapse. The success of the attack on Manichaeism led Wuzong to attack other “foreign religions,” including one of the most significant persecutions of Buddhism in Chinese history.

The weakening of the Tibetan Empire at about the same time further changed the political realities of Inner Asia. The Tibetans ceased to play a significant role beyond their own frontiers, again allowing the Tang Empire some respite and reducing the threat to the Uyghur refugees moving into the regions of Gansu and the Tarim Basin. Those Uyghurs who migrated to the west managed to develop a flourishing culture that remained important for centuries, ultimately being absorbed into the Mongol Empire of Chinggis Qan in the 13th century. Although their cultural legacy is profound, their political influence was relatively limited in comparison to that of the earlier Uyghur Empire.


Because only a handful of native sources are available from the Uyghur Empire, most of our information comes from nonnative sources (primarily but not exclusively Chinese), which often take a hostile viewpoint. The Uyghur texts (primarily stone inscriptions) that survive ameliorate this situation to a degree, but the inscriptions are often seriously damaged and quite difficult to read and interpret. Caution thus must be exercised in any historical analysis of the Uyghur Empire, given that it relies so heavily on antagonistic sources observing the empire from the outside and only a few fragmentary sources written by the Uyghurs themselves.

The most important of the Uyghur stelae is the trilingual Karabalghasun inscription, written in Old Turkic runiform script, Sogdian, and Chinese, which was erected some time during 809–821. The Chinese text has best survived the ravages of erosion. This inscription is particularly important because it is the only stone inscription that provides significant detail regarding the establishment of Manichaeism within the empire. The Sogdian text is in worse condition, but it still provides some interesting information. The Old Turkic text is the worst-preserved of the three.22

Other significant Uyghur inscriptions are monolingual, written in Old Turkic runiform script. The most important of these include the Shine-usu and Tes inscriptions, both of which date to ca. 759, and the Terkh or Terkhin inscription, also known as the Tariat inscription, which appears to be slightly earlier, ca. 753. All of these are particularly relevant to the reign of Bügü Qaghan. Finally, the Suci inscription, written ca. 840, is a brief text commemorating the life of one of the Kirghiz officers who helped destroy the Uyghur Empire.23 In addition to these stone inscriptions, a few documents have been found in the region of the Tarim Basin, especially Turfan, that relate to the Uyghur Empire. The most important of these is the so-called “Bügü Khan” text, which is also an important source on the introduction of Manichaeism within the empire.24

Chinese sources from the Tang era provide the largest body of information regarding the Uyghur Empire. The most significant include the two official Tang histories, Jiu Tang shu (completed ca. 945 by Liu Xu and others) and Xin Tang shu (completed significantly later, ca. 1060 by Ouyang Xiu and others, and containing a good deal of information not found in Jiu Tang shu). Each of these contains a chapter devoted to the Uyghurs, although there is much additional information on the Uyghurs scattered throughout.25 Also important are documents found in the literary collections of important Tang officials who lived during the era of the Uyghur Empire, such as Bai Juyi and Li Deyu.26 In addition, the magisterial chronicle by the Song scholar Sima Guang, Zizhi tongjian, completed in 1084, is indispensable, as are several other works that can be grouped together as collectanea: Tang huiyao, Tang da jiao lingji, Cefu yuangui, and others.

Finally, some rare but useful sources are available in other languages. The most significant of these is the account from the late years of the empire left by Tamīm ibn Baḥr, who most likely visited the empire, including Ordubalïq, in 821.27 In Islamic sources the Uyghurs are referred to as Toghuzghuz, a term which derives from Old Turkic Toquz Oghuz (“Nine Oghuz”), a term for a union of peoples frequently associated with the Uyghurs; this term is also used regularly in Chinese sources where it appears as Jiu Xing 九姓‎, “Nine Surnames.”

While it is possible that new sources related to the Uyghur Empire will be found, it seems unlikely that these will be texts of any great impact. The further expansion of knowledge about the Uyghur Empire will most likely rely significantly on archaeological investigation, which still has much to reveal regarding various aspects of life in the Uyghur Empire.

Further Reading

Arden-Wong, Lyndon A. “The Eastern Uighur Khaganate: An Exploration of Inner Asian Architectural and Cultural Exchange.” PhD diss., Macquarie University, Sydney, Australia, 2014.Find this resource:

Beckwith, Christopher I. “The Impact of the Horse and Silk Trade on the Economies of T’ang China and the Uighur Empire: On the Importance of International Commerce in the Early Middle Ages.” Journal of the Economic and Social History of the Orient 34.2 (1991): 183–198.Find this resource:

Beckwith, Christopher I.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, NJ: Princeton University Press, 1987. Revised edition in 1993.Find this resource:

Clark, Larry V. “The Conversion of Bügü Khan to Manichaeism.” In Studia Manichaica IV.; Internationaler Kongreß zum Manichäismus, Berlin, 14–18. Juli 1997. Edited by Ronald E. Emmerick, Werner Sundermann, and Peter Zieme, 83–123. Berlin: Akademie Verlag, 2000.Find this resource:

Drompp, Michael R.Tang China and the Collapse of the Uighur Empire: A Documentary History. Leiden: Brill, 2005.Find this resource:

Guo, Pingliang and Liu, Ge. Huihu shi zhinan. Urumqi: Xinjiang Renmin Chubanshe, 1995.Find this resource:

Hayashi, Toshio. “Uigur Policies toward Tang China.” Memoirs of the Research Department of the Toyo Bunko 60 (2002): 87–116.Find this resource:

Kamalov, Ablet. Drevnie Uĭgury VIII–IX vv. Almaty: Izdatel’skiĭ Dom “Nash Mir.” 2001.Find this resource:

Kamalov, Ablet. “Material Culture of the Nomadic Uighurs of the Eighth-Ninth Centuries in Central Asia.” In Religion, Customary Law, and Nomadic Technology, Edited by Michael Gervers and Wayne Schleppe, 27–33. Toronto: Joint Centre for Asia Pacific Studies, 2000.Find this resource:

Kamalov, Ablet. “The Moghon Shine Usu Inscription as the Earliest Uighur Historical Annals.” Central Asiatic Journal 47.1 (2003): 77–90.Find this resource:

Mackerras, Colin. The Uighur Empire According to the T’ang Dynastic Histories: A Study in Sino-Uighur Relations, 744–840. Columbia: University of South Carolina Press, 1973.Find this resource:

Skaff, Jonathan K.Sui-Tang China and Its Turko-Mongol Neighbors: Culture, Power, and Connections, 580–800. New York: Oxford University Press, 2012.Find this resource:

Twitchett, Denis, ed. The Cambridge History of China, Vol. 3: Sui and T’ang China, 589–906, Part I. New York: Cambridge University Press, 1979.Find this resource:

Twitchett, Denis. “Tibet in Tang’s Grand Strategy.” In Warfare in Chinese History. Edited by Hans Van de Ven, 106–179. Leiden: Brill, 2000.Find this resource:


(2.) See Jonathan K. Skaff, Sui-Tang China and Its Turko-Mongol Neighbors: Culture, Power, and Connections, 580–800 (New York: Oxford University Press, 2012), 188–189.

(3.) The Uyghur rulers adopted formal titles that often employed very similar terminology, making those titles difficult to differentiate from one another. For that reason, this article employs the less formal names/titles that are found throughout various sources, particularly Chinese. For a list of the formal titles of the various Uyghur qaghans and their variants, see James R. Hamilton, Les Ouïghours à l’époque des Cinq Dynasties d’après les documents chinois (Paris: Imprimerie National, Presses Universitaires de France, 1955), 139–142, as well as Colin Mackerras, The Uighur Empire According to the T’ang Dynastic Histories: A Study in Sino-Uighur Relations, 744–840 (Columbia: University of South Carolina Press, 1973), 192–193. As for the titles used here, preference has been given to Turkic (rather than Chinese) forms whenever possible, even if some of those are tentative.

(4.) The site is often referred to by its modern Mongol name, Karabalghasun (with variants).

(5.) Kamalov, Ablet, “Turks and Uighurs during the Rebellion of An Lu-shan [and] Shih Ch’ao-yi (755–762),” Central Asiatic Journal 45.2 (2001): 243–253.

(6.) For further information on these events, see Charles A. Peterson, “P’u-ku Huai-en僕固懷恩‎ and the T’ang Court: The Limits of Loyalty,” Monumenta Serica 29 (1970–1971): 423–455.

(7.) See I. Arzhantseva et al., “Por-Bajin: An Enigmatic Site of the Uighurs in Southern Siberia,” The European Archaeologist 35 (2011): 6–11 as well as Lyndon A. Arden-Wong, “The Eastern Uighur Khaganate: An Exploration of Inner Asian Architectural and Cultural Exchange” (PhD diss., Macquarie University, Sydney, Australia, 2014): 234–237.

(8.) See Arden-Wong, “The Eastern Uighur Khaganate.”

(9.) An and Shi (石‎) were two of the seven surnames commonly used by Sogdians in a Chinese context; the others were Cao, He, Kang, Mi, and Shi (史‎). Each of these was associated with a Sogdian city; see Edwin G. Pulleyblank, “A Sogdian Colony in Inner Mongolia,” T’oung Pao, 2nd series, 41.4/5 (1952): 320.

(10.) See Larry V. Clark, “The Conversion of Bügü Khan to Manichaeism,” in Studia Manichaica IV.; Internationaler Kongreß zum Manichäismus, Berlin, 14–18. Juli 1997, eds. Ronald E. Emmerick, Werner Sundermann, and Peter Zieme (Berlin: Akademie Verlag, 2000), 83–123.

(11.) See Willy Bang and Annemarie von Gabain, “Türkische Turfan-Texte II. Manichaica,” Sitzungsberichte der Preussische Akademie der Wissenschaften, Philosophische-historische Klasse, 1929, 411–430.

(12.) On Manichaeism in China under Uyghur patronage, see Samuel N. C. Lieu, Manichaeism in the Later Roman Empire and Medieval China: A Historical Survey (Manchester, U.K.: Manchester University Press, 1985), 194–198.

(13.) Jan Bemmann et al., “Bookmarkers in Archaeology—Land Use around the Uyghur Capital Karabalgasun, Orkhon Valley, Mongolia,” Praehistorische Zeitschrift 89.2 (2014): 337–370.

(14.) On the role of the Uyghurs in the Zhu Tao rebellion, see Mackerras, The Uighur Empire, 39–41.

(15.) See Yihong Pan, “Marriage Alliances and Chinese Princesses in International Politics from Han through T’ang,” Asia Major, 3rd series, 10.1–2 (1997): 95–131, as well as Michael R. Drompp, “From Qatun To Refugee: The Taihe Princess among the Uighurs,” in The Role of Women in the Altaic World: Permanent International Altaistic Conference 44th Meeting, Walberberg, 26–31 August 2001, ed. Veronika Veit (Wiesbaden: Harrassowitz Verlat, 2007), 57–68.

(16.) On Dezong’s hostile attitude towards the Uyghurs, see Martin Slobodník, “The Early Policy of Emperor Tang Dezong (779–805) towards Inner Asia,” Asian and African Studies 6.2 (1997): 184–196.

(17.) Twitchett, Denis, ed. The Cambridge History of China, Vol. 3, Sui and T’ang China, 589–906, Part I (Cambridge: Cambridge University Press, 1979), 610.

(18.) On this inscription see Mackerras, The Uighur Empire, 184–187.

(19.) For a complete list of all known missions, see Colin Mackerras, “Sino–Uighur Diplomatic and Trade Contacts (744 to 840),” Central Asiatic Journal 13 (1969): 215–240.

(20.) See Michael R. Drompp, Tang China and the Collapse of the Uighur Empire: A Documentary History (Leiden: Brill, 2005), 106.

(21.) Michael R. Drompp, “Breaking the Orkhon Tradition: Kirghiz Adherence to the Yenisei Region after A.D. 840,” Journal of the American Oriental Society 119.3 (1999): 390–403.

(22.) The Chinese inscription was translated, with commentary, by Gustav Schlegel in his “Die chinesische Inschrift auf dem uigurischen Denkmal in Kara Balgassun,” Mémoires de la Société Finno-Ougrienne 9 (1896). This translation contains many errors; a new translation is very much to be desired. An English translation of the Sogdian inscription may be found in Takao Moriyasu and Ayudai Ochir, eds., Mongorukoku genzon iseki, hibun chōsa kenkyū hōkoku (Osaka: The Society of Central Eurasian Studies, 1999), 209–227.

(23.) The Shine-usu and Suci inscriptions have been translated in Gustav Ramstedt, “Zwei uigurische Runeninschriften in der Nord-Mongolei,” Journal de la Société Finno-Ougrienne 30 (1913). An English translation of the former may be found in Moriyasu and Ochir, Mongorukoku genzon iseki, 177–195; on the latter, see also Louis Bazin, “L’Inscription kirghize de Sūǰi (Essai d’une nouvelle lecture),” in Documents et Archives provenant de l’Asie Centrale. Actes du Colloque Franco-Japonais Kyoto 4–8 Octobre 1988, ed. Akira Haneda (Kyoto: Association Franco-Japonaise des Études Orientales, 1990), 135–146. English translations of the Tes inscription are available in S. G. Klyashtorny, “The Tes Inscription of the Uighur Bögü Qaghan,” Acta Orientalia Scientiarum Hungaricae 39 (1985): 137–156 and in Moriyasu and Ochir, Mongorukoku genzon iseki, 158–167. English translations of the Terkh/Terkhin or Tariat inscription may be found in S. G. Klyashtorny, “TheTerkin Inscription,” Acta Orientalia Scientiarum Hungaricae 36 (1982): 335–366 and in Moriyasu and Ochir, Mongorukoku genzon iseki, 168–176. Translations of many of these inscriptions also exist in other languages, including Russian and Turkish.

(24.) For a German translation of this text, see Bang and von Gabain, “Türkische Turfan-Texte II. Manichaica.”

(25.) Most of the Jiu Tang shu and Xin Tang shu chapters on the Uyghurs are translated in Colin Mackerras, The Uighur Empire. This does not by any means exhaust the materials on the Uyghurs to be found in those two works.

(26.) The relevant writings of Li Deyu (a total of 69 documents) are translated in Drompp, Tang China and the Collapse of the Uighur Empire.

(27.) An English translation may be found in V. Minorsky, “Tamīm ibn Baḥr’s Journey to the Uyghurs,” Bulletin of the School of Oriental and African Studies 12.2 (1948): 275–305.