By: Jonathan Karam Skaff (Shippensburg University of Pennsylvania)
| Cross-cultural contact is a fundamental issue in world history. Throughout history peaceful and violent encounters between cultures and civilizations have been prime causes of social, economic, political, and material changes. Although scholarship on cultural contact has garnered increasing attention in recent years, the growing body of literature has bfeen dominated by studies on European colonialism.1 In order to place European expansion in broader historical context, there is a need for more studies of intercultural relations in premodern times. This paper will attempt to take a step in this direction by examining cultural contacts along the premodern China–Inner Asia frontier and then placing these findings in a comparative framework. | 1 |
| This paper fits within a growing body of historical, anthropological, and archaeological scholarship that views frontiers as dynamic zones of political, cultural, and economic interaction. This recent research challenges ideas about the frontier and cultural contacts that dominated the academy for much of the twentieth century and ultimately were based upon assumptions derived from nineteenth-century European nationalism and imperialism. In the 1800s circumscribed territorial boundaries of newly created western European nation-states led to the tendency to view all frontiers as “natural” geographical, political, cultural, and sometimes economic dividing lines. Moreover, nationalism bred the assumption that the cultures restricted within the boundaries or countries were static and homogenous. In the case of expanding frontiers, such as in colonial America or Africa, nineteenth-century imperialism implied that cultural change could occur, but only when “advanced” peoples engulfed and civilized “barbarians” (Sahlins 1989, 1–9; Whittaker 1994, 2–8; Lightfoot and Martinez 1995, 473;Power and Standen 1999, 1–9). Although there were some early dissenters from these views, such as Febvre and Lattimore, who championed the idea of culturally complex zonal frontiers, for the most part their perceptive observations were not as influential as they should have been (Febvre 1924, 296–315; Lattimore 1951, 238–251). For example, in 1982 Eric Wolf complained that “[t]he habit of treating named entities such as Iroquois, Greece, Persia, or the United States as fixed entities opposed to one another by stable internal architecture and external boundaries interferes with our ability to understand their mutual encounter and confrontation” (p.7). | 2 |
| It was not until the 1980s and especially the 1990s that more dynamic models of frontiers and cultural interaction became generally accepted. According to the revisionists, most frontiers are permeable zones of fluid interaction and exchange. As regions distant from centers of governmental control and ideological orthodoxy, people with different political and social affiliations have greater freedom to come together. These tendencies were even more pronounced in the pre-modern era when communications were primitive and states had lower capacities to control their populations.2 Moreover, anthropological studies of ethnic identity now generally agree that ethnicity is mutable and recognize that when different cultural groups meet in contact zones, they are prone to mutual influence, even in cases where one culture is militarily and politically dominant (Barth 1969; Keyes 1981; G. Bentley 1987; Pratt 1992, 6–7). | 3 |
| In keeping with this new focus of scholarship, this first part of this paper will use a case study of the early seventh-century China–Inner Asia frontier to argue that it was a highly dynamic zone of interaction —with overlapping political, economic, and cultural spheres—where personal relations and self-interest, rather than ideologies of loyalty, generally determined political allegiances. Typically, this period in Chinese history has attracted attention from historians who have sought to explain the rapid demise of the short-lived Sui dynasty (587–618) and the military and civil consolidation of the much longer lasting Tang (618–907). Rather than being concerned with the rise and fall of dynasties, this paper investigates heretofore obscure events on China’s northern frontier involving interactions between nomads, warlords, and refugees. | 4 |
| The second part of this paper adopts a comparative framework that attempts to place the Sui-Tang borderlands in the broader context of premodern frontier history in order to contribute to the theoretical literature on frontiers. It will endeavor to move beyond generalizations about the fluidity of borderlands by analyzing the different factors that affected bonds of political solidarity between people in frontier areas and core polities. Although borderlands and the people who lived on them often have been considered peripheral or marginal to the study of history, this paper will demonstrate that they can illuminate our understanding of premodern identities and loyalties. It will argue that on the premodern frontier political allegiances to the center generally were weak and not defined according to identity, except under certain restricted conditions. | 5 |
| The Sui-Tang Frontier Zone | |
| Before going into the details of this study, it will be helpful to the nonspecialist to provide some historical and geographical background on China and Inner Asia. After the fall of the Han dynasty (202 b.c.e.–220 c.e.), China underwent three and a half centuries of political disunity. Toward the end of the sixth century, China was divided into two rival kingdoms: the Zhou in the north and the Chen in the south. In 581 Yang Jian, a high official of the Zhou, usurped the throne from a child emperor and founded the Sui dynasty. Ruling as Emperor Wendi (r.581–605), in 589 he conquered the Chen and reunited all of the Chinese heartland under effective central rule for the first time in almost four centuries. However, Wendi’s work rapidly came undone under his son, Emperor Yangdi (r.605–617), in the late 610s, as rebellions broke out throughout the empire. One of the rebel leaders, a former Sui frontier official named Li Yuan, gained control of the capital in 617 and declared the founding of his own Tang Dynasty in the following year. Li (r.618–626), who is best known by his posthumous temple name of Gaozu, gradually consolidated power over the next several years in a series of battles against Sui loyalists and other warlords. The Tang house would remain in control of China for almost three centuries (Bingham 1941; Wright 1978; Wechsler 1979; Wright 1979). | 6 |
| During the Sui and early Tang dynasties, China’s major external rival was the Eastern Turk tribal confederation that had united the pastoral nomadic tribes living on the steppe and deserts of Inner Asia to the north of China proper. The Turks and other pastoral nomads conventionally have been considered to be antithetical to the Chinese. While the Chinese heartland had a subsistence economy based on intensive sedentary agriculture, nomads lived off of the products of the livestock they raised in periodic migrations on the grasslands of Inner Asia. The latter’s ability to raise large numbers of horses on the steppe allowed them to develop cavalry armies that could threaten the more populous and wealthier China.3 | 7 |
| The idea that these two societies were starkly opposed was enshrined in traditional Chinese histories, which are the main sources for the first part of this study.4 Composed by elite scholars living in the dynastic capitals, these histories paint an idealized portrait of a superior Chinese state and society that contrasted with uncivilized “barbarians” living beyond the borders (Yang 1968, 27; Sinor 1978). Traditional Chinese historians were not unusual in engaging in this type of stereotyping. Romans used similar topoi when writing about neighboring peoples (Whittaker 1994, 131–133; Olster 1996, 93–97). Moreover, the distinguished French linguist Emile Benveniste has written that this was typical of premodern writing: “Every name of ethnic character, in ancient times, was differentiating and oppositional. There was present in the name which a people assumed the intention, manifest or not, of distinguishing itself from the neighboring peoples, of affirming the superiority derived from a common, intelligible language. Hence the ethnic group often constituted an antithetical duality with the opposed ethnic group” (Benveniste 1969, 1:368; quoted in Armstrong 1982, 5). The elite members of Chinese society who produced Sinocentric rhetoric were engaged in a project of defining “the distinguishing marks and boundaries” of Chinese culture (Duara 1995, 60), which implicitly was assumed to be coterminous with the borders of the Chinese empire. | 8 |
| Although it is not unusual, the Sinocentric ideology and concomitant stereotyping in the premodern Chinese sources have influenced modern historians’ understanding of China’s relations with the outside world. Scholars have been inclined to accept these claims about traditional Chinese culture because they confirmed the abovementioned assumptions about the essentiality of ethnicity that grew out of nineteenth-century nationalism and imperialism. A related idea, also developing partly from the bias of the traditional sources and partly from modern geographical studies, is the notion that there was a natural ecological frontier that neatly divided an agricultural Chinese “culture island” from pastoral nomads of the Inner Asian steppe, especially those in modern Mongolia. Belief in an ecological border led to the related assumption that pastoral nomads and Chinese were culturally incompatible (Fairbank 1968, 5; Barfield 1989, 16; Jagchid and Symons 1989, 24). Those who have held this view of an ecological and cultural boundary between China and Inner Asia ignored Owen Lattimore’s work originally published in 1940 in which he described China’s northern frontier as a “marginal zone” where these distinctive ecological niches blended imperceptibly into each other. He believed that farmers and nomads could intermingle there and potentially change their cultural affiliations.5 More recently, Peter Perdue has developed a useful model for understanding the relationship between China and Inner Asia. He views China proper and Mongolia as “core” areas where Chinese peasants and pastoral nomads never met. However, the peripheries of these civilizations came together where modern Inner Mongolia borders Gansu, Ningxia, Shaanxi, Shanxi, and Hebei Provinces. In this extensive region, stretching around two thousand kilometers from west to east, steppe and agricultural ecological zones flow imperceptibly into one another. In this “contact zone” people met and created new cultural formations (Perdue forthcoming). | 9 |
| Empirical evidence drawn from Sui and Tang records supports the idea that the Chinese–Inner Asian frontier was a zone with ample opportunities for political and cultural interaction. During the Sui-Tang transition, when China lacked a strong centralized government, these borderlands were a fluid region where people’s physical locations and loyalties could shift. Although ethnic differences were recognized, they were not primary determinants of political affiliation. Moreover, Sinocentric ideology did not have a strong grip on ethnically Chinese frontier inhabitants. Despite what traditional histories and modern ideals of nationalism might lead us to expect, contact on the margin led to political and cultural interchange. | 10 |
| The Frontier as a Zone of Interaction and Interchange | |
| One interesting phenomenon that illustrates a tendency to break down the division between mobile nomads and sedentary Chinese was the establishment of Turk cities in what today is western Inner Mongolia. This began around the year 600 while the founding Sui emperor Wendi was on the throne and the dynasty’s rule was still strong. Taking advantage of factional infighting—typical of decentralized nomadic politics —among the Turk ruling family, the Sui attracted Qimin Qaghan and his followers to move south from Mongolia across the Gobi desert into Inner Mongolia.6 The Sui strategy was to have these Turk allies assist in defending the frontier against their brethren still in Mongolia. Some of these Turks lived the traditional pastoral nomadic lifestyle in the grasslands of the northern Ordos region, but what is more interesting is that the Sui also built two walled towns, Dali and Jinhe, for the Turks to the northeast of the Ordos. In addition, the Sui stationed twenty thousand troops at these cities to assist in defending the frontier and probably to monitor Qimin and his followers. Just for the construction of Dali, fifty thousand Sui subjects were mobilized (SS 84:1873; ZZTJ 178:5568–5569, 5571–5572;Pan 1997, 105–107). Unfortunately, surviving sources do not provide information about life in these towns. However, we can assume that the concept of urban living was not necessarily alien to the Turks in the early seventh century. Modern archaeological investigations show that agriculture and city building were more prevalent in premodern Mongolia and Inner Mongolia than previously believed (Di Cosmo 1994; Di Cosmo 2002, 251, nn.155–156; Minyaev n.d.) Nonetheless, the involvement of the Sui government must have increased Turk urbanization to a higher than usual scale.7 Moreover, the closer proximity of the Turk political leadership to China proper would have important implications because it was going to precipitate a higher degree of political and cultural interaction during the Sui-Tang dynastic transition. | 11 |
| During the civil wars that broke out over a decade later at the end of the Sui, many people fled China for the relative stability that the Turks could offer in Inner Mongolia. A number of sources mention that at least ten thousand Chinese (Zhongguoren) had fled to the Turks by 618, and they were settled in the cities built during the Sui (SS 84: 1876; ZZTJ 188:5878; JTS 194a:5154; XTS 215a:6029). By 631 their number had grown to eighty thousand. Of this larger number, some of the Chinese appear to have been captured in Turk raids, but the sources are not clear about how many.8 Most of these people were commoners, but elite Sui loyalists were among their number, including the wife and grandson of the recently killed Emperor Yangdi. In a move that demonstrated newly reversed power relations, the reigning qaghan, Chuluo, appointed Yangdi’s grandson as a subordinate king of Sui. Chuluo organized the Chinese commoners, apparently as an agricultural populace, under the jurisdiction of officials who used the Sui legal code (SS 84:1876; ZZTJ 188:5878; JTS 194a:5154; XTS 215a:6029). The imposition of Sui law implies that the refugees had to pay Chinese tax rates, which would have gone into Chuluo’s coffers.9 It appears that Xieli Qaghan, who succeeded Chuluo, went even further and tried to administer his nomadic followers based on sedentary models. One of the Chinese who had fled to the Turks, Zhao Deyan, and a number of Sogdians (Hu), originally from the oases of West Turkestan, became highly influential in Turk government.10 | 12 |
| This Turk polity with its openness to cultural interchange began to deteriorate in the late 620s. By then the Tang had consolidated power in China proper and Xieli Qaghan’s fortunes were in decline. Supposedly, many Turk tribesmen were disenchanted with new laws and taxes based upon Sogdian and Chinese models, which were more onerous than nomadic customary practices. Moreover, in 627 there were heavy snows that killed livestock and made agriculture difficult. Reports filtered into the Tang court that the Chinese and some Turks had risen in rebellion with the former finding refuge in the mountains.11 Taking advantage of the situation, the Tang decisively defeated the Turks in 630.12 However, not all Chinese were willing to return to the Tang. Some fled west to Gaochang (Turfan) in East Turkestan where the local king refused the Tang emperor Taizong’s (r.626–649) requests to send them back (JTS 198:5294; XTS 221a:622; Chavannes 1900, 104). Although the sources do not state why this group escaped to Gaochang, one can surmise that they may have been Sui loyalists who feared returning to the new dynasty. | 13 |
| From this case, we can see that in these borderland people’s political allegiances were determined mainly by self-interest. For ethnically Chinese and Turk commoners, loyalty depended upon a political leader’s capacity to ensure stable economic and social conditions. When bad times came, they sought to escape or overturn a ruler. For the social elite, loyalty was contingent upon a ruler’s ability to provide political patronage. In neither case does it appear that ethnic affinity was a primary determinant of political allegiance. Likewise, the evidence does not support the putative superiority of Chinese culture posited by traditional Sinocentric ideology. Although the Turk leadership south of the Gobi Desert was influenced by the new experience of ruling over sedentary and nomadic peoples, we cannot conclude that Chinese culture had an overriding attraction. Xieli Qaghan adopted some Chinese methods of governance, but also took advice from Sogdians from West Turkestan. Moreover, his nomadic followers did not find all of these changes to be appealing, whether their origins lay in China or West Asia. | 14 |
| Frontier Warlords | |
| We can learn more about the complexity of loyalties on the frontier if we look at the careers of some Chinese warlords who made alliances with the Turks during the civil war at the end of the Sui. A number of claimants for power sought aid from the Turks and accepted Turkic titles. The History of the Sui claims that seven of them proclaimed themselves emperors of China while simultaneously acknowledging themselves as subjects of the Turks and accepting the Turkic title of qaghan.13 Chinese historians have debated whether the Tang dynastic founder, Gaozu, formally acknowledged himself as a vassal to the Turks, but undoubtedly he initially acted deferentially toward them (Pan 1997, 169–171; Graff 2002, 34–35). He had to behave in this manner because, like the other warlords, he was in need of assistance. For example, soon after revolting against the Sui, he accepted Turk horses and cavalrymen for his army.14 He and other contestants for power also probably had hopes, which ultimately proved to be futile, of gaining an exclusive alliance with the Turks against their rivals. On the other hand, the Turk qaghan Chuluo and his successor, Xieli, seem to have done their best to keep the rival claimants to power in China divided and weak. For example, as the Tang became the dominant force in the civil war—consolidating power in the economically and strategically important capital region by 619 and subsequently conquering China’s fertile northeastern plain in 621—Tang-Turk relations turned increasingly cold and the Turks actively supported the Tang’s rivals.15 The Turks evidently hoped to remain the preeminent power in East Asia by preventing the Tang from fully unifying China (Eisenberg 1997, 313; Graff 2002, 35). | 15 |
| We can gain a greater appreciation about relations between the Turks and the frontier warlords during the civil war by looking more closely at the careers of three of them: Liang Shidu, Gao Kaidao, and Yuan Junzhang. Liang was from a prominent family in Shuofang commandery in the middle of the Ordos region, which today is divided between northern Shaanxi Province and Inner Mongolia. During the Sui it was a frontier commandery beyond which lay nomadic tribes loyal to the Turks. After civil war broke out in 617, Liang gathered a band of followers and led a successful revolt against the local Sui official. He subsequently captured three neighboring commanderies, then proclaimed himself emperor and carried out traditional Chinese Confucian rites to legitimize his rule. To bolster his power, he allied with the Turk qaghan Shibi, who sent him a wolf ‘s head standard and invested him with the dual titles of qaghan, which was the Turkic appellation for a ruler, and Son of Heaven, which was applied to Chinese emperors (JTS 56:2280; XTS 87:3730; ZZTJ 183:5718, 5724). Although qaghan means “independent ruler,” in practice the Turk confederation might have several subordinate qaghans at any one time (Drompp 1991). From the perspective of the Turks, Liang Shidu probably was considered one of these subordinate qaghans. | 16 |
| Despite this promising beginning, by 620 Liang had been unable to appreciably expand his power base. This left him in a vulnerable position because his relatively arid territory in the marginal zone on the edge of the steppe was not densely populated nor particularly wealthy. He obviously feared the increasingly powerful Tang based in more fertile and populous southern Shaanxi. Seeing that he stood no chance of defeating the Tang alone, Liang encouraged his ally, Chuluo Qaghan, who had succeeded Shibi, to follow the example of the founder of the earlier non-Chinese Northern Wei dynasty (386–534) by invading and setting up rule over China. Chuluo took his advice and devised a four-pronged attack against China that included Liang’s forces. However, Chuluo died before the invasion was launched and the plan was dropped (JTS 56:2280; XTS 87:3730; ZZTJ 188:5895–5896). Thereafter, Liang held out as a borderland warlord until 628, when Turk power began to wane. At this point the Tang besieged Liang’s base at Shuofang. As food dwindled, Liang’s followers became rebellious. Eventually, his cousin Liang Luoren killed him and turned the city over to the Tang (JTS 56:2281; XTS 87:3731; ZZTJ 192:6050). | 17 |
| The life of Liang Shidu illustrates that although he was an ethnically Chinese member of the local elite who concerned himself with some Confucian norms, he did not feel that non-Chinese should be excluded as political patrons. Being from a border area, he probably had prior experience dealing with nomads. We do not have evidence that he was influenced by Sinocentric notions of cultural superiority, but even if he was, he did not allow them to override his pragmatic political calculations. His loyalty was not directed to an abstract concept of “China,” but rather toward a suzerain who could benefit him, regardless of ethnicity. His followers likely had similar attitudes because they abandoned him and the Turks only after it became apparent that their nomadic patrons no longer could shield them from the Tang. | 18 |
| The life of another rebel, Gao Kaidao, demonstrates that Chinese from more central areas could move into the borderlands and also uninhibitedly interact with the Turks. Gao was from Cangzhou on the coast of modern Hebei, south of Tianjin. Unlike Liang, he was of humble origins and probably illiterate because his biography states that he was a salt boiler in his youth. At the end of the Sui he joined a local revolt and rose up through the ranks. After the head of his band was killed, Gao took over command of one hundred surviving men. Initially, he mostly fought and allied with other Chinese warlords, rather than nomads. In these dealings he demonstrated strong Machiavellian tendencies. For example, after pretending to join the forces of the rebel monk Gao Tansheng—who had declared himself emperor, taken a nun as empress, and established a kingdom based on Buddhist law—Gao Kaidao killed the monk-emperor and took control of his five thousand followers. Through this sort of trickery and natural leadership ability, by 618 he had built his forces to the point where he had several thousand horses and ten thousand soldiers, including a personal guard of several hundred men who presumably were most loyal to him. Most germane to the present study, during his rise to prominence, he moved his base of operations into the China—Inner Asia frontier zone north of modern Tianjin. He declared himself king of Yan and established a calendar and an official bureaucracy, which indicates acceptance of Chinese norms of government (JTS 55:2256–2257; XTS 86:3714–3715; ZZTJ 188:5892). | 19 |
| Initially, Gao refrained from dealing with the Turks. In 620 he looked toward the central lands when he gave his fealty to the Tang. The dynasty recognized his de facto independence by leaving him in place with the title of Yu Prefecture commander-in-chief, and awarding him use of the imperial surname, Li, along with other impressive-sounding titles. This was a common Tang practice at the start of the dynasty when it was too weak to enforce centralized control over outlying prefectures (Somers 1990, 374–378). Gao had gained the trust of the Tang by giving aid to another of their semi-independent commanders-in-chief, Luo Yi of You Prefecture (modern Beijing), whom Gao had saved from an attack by the warlord Dou Jiande. Despite the appearance of loyalty to the dynasty, in 621 or 622 Gao proved once again that he was purely an opportunist when he pretended to offer aid to Luo during a famine at You Prefecture. After Gao promised to procure grain supplies, Luo sent three thousand soldiers, several hundred carts, and over one thousand horses and donkeys to pick up the grain. Taking advantage of the situation, Gao detained all of the men, carts, and animals; broke relations with the Tang; and reestablished the independent kingdom of Yan. To bolster his position, he allied with the Turks (JTS 55:2256–2257; XTS 86:3714; ZZTJ 188:5892, 189: 5939–5940). For the next couple of years, Gao cooperated with other Chinese rebels, the Turks, and their subject tribe, the Xi, in plundering Tang territory (JTS 55:2257; XTS 86:3714; ZZTJ 190:5948–5949, 5966, 5968–5969, 5971–5972). Gao obviously was willing to enter into any alliance that was advantageous to himself. | 20 |
| However, in 624 Gao began to regret his earlier break with the Tang, which had defeated most rivals for power by this time. Although Gao sensed that it would have been most advantageous to reestablish fealty to the Tang, he feared being punished for his past treachery. Consequently, he continued his alliance with the Turks for his own protection. Trouble was brewing, however, because his officers and soldiers, who were mostly from Shandong to the south, were homesick and increasingly restless. These problems had manifested themselves earlier in 623 when two of his garrisons surrendered to the Tang. Finally in 624 the opportunistic Gao himself fell victim to treachery when his “beloved” general Zhang Junli staged a coup. Although Gao’s personal guard rushed to protect him, they surrendered when they discovered that Zhang’s men had surreptitiously disarmed them. After Gao was surrounded by Zhang’s personal forces, Gao killed himself and Zhang subsequently submitted to the Tang (JTS 55:2257; XTS 86:3714–3715; ZZTJ 190:5970, 5977). | 21 |
| Gao Kaidao’s life demonstrates the fickle nature of loyalties during the dynastic transition. Since he was uneducated and came of age in the unstable environment of rebellion, it is not surprising that Gao’s career shows little attachment to the niceties of Confucian ideals of loyalty. To him, allegiance at best was granted for personal gain, and at worst was bait used to trap an unsuspecting victim. In this perfidious game he did not discriminate between Chinese and nomad, who both were potential benefactors or dupes. Ultimately, though, he was deceived by a favored follower who played by the same set of treacherous rules. His adherents’ allegiances likewise were contingent or not held deeply. Although Gao’s guard corps of several hundred men were personally loyal to him, they did not defend him to death, and most other adherents appear to have joined Gao for personal enrichment or protection, or out of coercion. When Gao failed to establish a secure state, their most powerful attachment proved to be local. Rather than wanting to rejoin “China,” or to fight to the end for Gao, they merely desired to return to their homes in Shandong. | 22 |
| The final frontier rebel whose life can give us insight into loyalties and relations with the Turks is Yuan Junzhang. Unlike Gao Kaidao, but similar to Liang Shidu, Yuan was a native of the frontier zone and appears to have had great familiarity with the Turks from the start of his career as a warlord. The record of deliberations by Yuan and those around him is particularly rich probably because, unlike many other warlords who were killed in the civil war, Yuan ultimately negotiated a surrender to the Tang and was appointed to an official position. These circumstances provided contemporary historians with greater access to stories about him. Like Gao, Yuan was illiterate and of obscure background, rising to prominence in the ranks of the warlord Liu Wuzhou in northern Shanxi. After a major defeat at the hands of the Tang in 620, Liu fell out with his erstwhile allies, the Turks, and they killed him. The Turks then placed Yuan in charge of Liu’s remaining forces in Shuozhou in northern Shanxi (JTS 55:2255; XTS 92:3804–3805; ZZTJ 188:5882–5883). We can see that Yuan truly was subordinate to the Turks because Yushe Shad, a member of the Turk ruling clan, and a garrison of at least two hundred Turks monitored his activities.16 Incidentally, the Turks probably supervised other frontier warlords in a similar way.17 | 23 |
| Throughout his time in power, Yuan appears to have aspired to be a regional ruler in the frontier zone between China and the steppe, using the support of the Turks to retain his independence from the Tang. In policy discussions before Liu Wuzhou launched his ill-fated attack against the Tang, Yuan had promoted a strategy of regional independence by warning Liu to remain in northern Shanxi and not challenge the Tang: The Tang rose up with the troops of one prefecture and pacified the capital region. The prefectures and counties [of central China] have submitted [to them]. Those who face them are dispersed in the wind. This certainly is the Mandate of Heaven [to rule China], how can it be called the work of men? Moreover from Bingzhou [Taiyuan] to the south, the terrain is treacherous. If we move troops deep into their territory, I fear that later it will not be possible to advance further. It would be better to unite with the Turks and have foreign relations with the Tang, who have claimed the emperorship. This suffices as the best plan.18 | 24 |
| Although Liu did not follow this advice, Yuan later strove to put it into practice when he gained power. Assuming that the exact wording of the speech did not spring from the imagination of a Confucian-in flu-enced historian, Yuan apparently recognized basic Confucian concepts, such as the legitimization ideology of the Mandate of Heaven to rule China, but his actions were not guided by them.19 His plan envisioned a world in which the ideal was not a unified empire containing all those who were ethnically Chinese. While admitting that only Tang rulers had the legitimacy to take the title of emperor because of their control of the bulk of central lands, Yuan’s plan left room for borderland states with Chinese-speaking rulers. To guarantee independence, Yuan proposed a subordinate alliance to the supposedly ethnically alien Turks. Yuan did not promote a culturally unified China or loyalty toward the abstract concept of a Chinese state. Instead his ideal was regional independence under the suzerainty of a non-Chinese people, the Turks. Presumably the attraction of Turk overlordship was that their local leaders enjoyed greater freedom of action and lighter tax levies than those in the more centralized Chinese administrative system.20 | 25 |
| Yuan’s plan appears to have gone smoothly until 623. In that year the Tang frontier official Liu Shirang began to implement a successful scheme to counter Yuan and the Turks. In a memorial to Emperor Gaozu, Liu pointed out that Shuozhou was strategically important because the Turks used it as a stopping place to rest and resupply their forces prior to raids on Tang territory. He proposed a policy that would use the carrot and the stick to bring Shuozhou into Tang hands. Raids on the countryside around Shuozhou would seek to destroy crops, while generous rewards would be offered to those who surrendered to the Tang. Evidently, these measures were effective because the local people began to feel discontent toward Yuan and his forces (JTS 69:2522; XTS 94:3835; ZZTJ 190:5968). | 26 |
| In response to these growing problems, one of Yuan’s generals, Gao Manzheng, proposed a xenophobic strategy that made full use of Sinocentric stereotypes of foreigners. The histories quote him as saying, “The [Turk] barbarians lack propriety. At their root they are not human. How can you face the north and serve them? It would be better to kill all of the Turks [in the garrison] and return [allegiance] to the Tang Dynasty” (JTS 55:2255; XTS 92:3805). | 27 |
| Despite Gao’s appeal to xenophobic sentiment and increasing military pressure from the Tang, Yuan declined to follow this advice. Shortly thereafter, Gao carried out a coup, killing Yuan’s son and two hundred Turks and then gave fealty to the Tang. Although Yuan escaped to the Turks, Gao received Tang recognition as commander of Shuozhou and probably other generous rewards in return for his services. Nevertheless, later that year Yuan recaptured Shuozhou with assistance from his Turk allies and killed Gao and twenty local notables who had supported him (JTS 55:2255; XTS 92:3805; ZZTJ 190:5967–5969, 5973). Yuan’s strategy of alliance with the Turks ultimately was vindicated, and Gao’s plan to unite with the Tang was repudiated, because the Tang could not project enough power to keep Shuozhou out of Turk hands. | 28 |
| By 627 Yuan reached a point where he was forced to reevaluate his strategy. Earlier, the Turks had negotiated the transfer of Shuozhou to the Tang, forcing Yuan and his followers to move farther north to Heng’an (modern Datong). Also at this time, the Tang had further consolidated their power in China proper and potentially had the ability to challenge the Turks in northern Shanxi. Moreover, some of Yuan’s Chinese followers were becoming disenchanted and slipping away to the south(JTS 55:2255; XTS 92:3805; ZZTJ 190:5973). The sources do not explain why some adherents were deserting Yuan. It is likely that they fled Datong because of homesickness or economic hardship. Both factors were at play in 627. As is mentioned below, the city was experiencing problems with its food supplies. Furthermore, when Yuan moved his forces north to Datong, he must have taken followers from his original base at Shuozhou, who later may have developed a desire to return home. As seen above, this factor had a strong influence on the behavior of Gao Kaidao’s followers. | 29 |
| As Yuan tried to determine a course of action, he was wooed by representatives from the Turks and Tang, both of whom wanted his allegiance, but Yuan had a difficult time coming to a decision. At a policy meeting with his followers, his son Yuan Xiaozheng favored submitting to Tang rule because of problems with food supplies. However, a native of Datong, Guo Ziwei, gave a speech that convinced Yuan to maintain the status quo in foreign relations: “Heng’an [Datong] formerly was a royal capital [of the Northern Wei dynasty]. The lay of the land provides an impregnable defensive position. The Turks, who now are powerful, are our close neighbor [literally: lips and teeth]. Occupying this solidly fortified city provides an opportunity to observe the developments in All under Heaven [the World]. Why should we desire to surrender to others?”21 | 30 |
| Later, when it became apparent that Turk power was collapsing, Yuan finally submitted to the Tang. He ended his career as a Tang military official in the central province of Hubei, which, from the dynasty’s perspective, was safely removed from his original base of power (JTS 55:2255; XTS 92:3805; ZZTJ 190:5973). | 31 |
| For the most part these policy debates in Yuan Junzhang’s inner circle during the 620s placed greater emphasis on calculations of self-interest rather than cultural affinity. The alliance with the Turks was maintained as long as their ability to project military power in northern Shanxi was greater than the Tang’s. Turk overlordship presumably also had the benefit of affording more political independence to local leaders than Chinese centralized administration. There were two arguments for uniting with the Tang. One was based on ideals of ethnic affinity, and the other on pragmatic problems with food supplies. Although Gao Manzheng promoted cultural unity with Tang China, he made these arguments only when Shuozhou was feeling severe military and economic distress. He presumably was further enticed by expectation of a reward upon surrendering to the Tang. On the other hand, local allegiances appear to have played an important role. Guo Ziwei was happy to have regional independence in Datong under Turk suzerainty, despite the hardships. Other followers of Yuan were slipping away to the south, presumably at least in part because they wanted to return to their original homes. Although the ideal of Chinese cultural unification may have appealed to some, it did not have the overriding allure that Sinocentric ideology might lead us to expect. For example, even though Turk power played a major role in the decision to maintain an alliance, those in Yuan’s camp who favored this position neither complained about Turk barbarity nor expressed nostalgia for an unified Chinese empire. For those who approved of an alliance with the Turks, cultural affinity was a nonissue in political relations. And even when Gao Manzheng made Sinocentric arguments to oppose the Turks, pragmatic considerations probably were at the forefront of his political calculations. For him, cultural affinity became a viable issue only when the military and economic situation could add weight to his arguments. | 32 |
| To sum up the above discussion, the China–Inner Asia borderlands from the early 600s to early 630s exhibited a great deal of cultural interchange and experimentation. The frontier truly appears to have been a permeable zone of ecological transition that permitted people to move in both directions along the borderlands in terms of both their physical locations and their political allegiances. When the Sui was strong and stable, Turks migrated closer to China and experimented with sedentary living. Later, when Sui China fell into chaos, Chinese moved in the opposite direction and submitted to the Turks. The discontinuity between political and cultural spheres of influence in the borderlands is demonstrated by Yuan Junzhang’s use of the expression “Mandate of Heaven” to describe the Tang’s rule over China. If he really used this turn of phrase and it is not a Confucian historian’s embellishment, culturally he was within China’s mental universe, using the expression as a rhetorical shorthand that his listeners would readily apprehend to describe changed power relations in China. Politically, however, he wanted to stay outside of China because he refused to recognize the Tang’s legitimacy to rule over northern Shanxi. Likewise, Liang Shidu’s use of a wolf ‘s head standard illustrates that Turkic symbols also had power in the frontier zone. In this hybrid cultural environment, symbolic categories from different traditions converged and might develop new nuances in meaning. | 33 |
| Political loyalties exhibited great flexibility in the frontier zone. The strongest solidarities present were local or personal, but either could be forsaken for private gain or protection. Personal loyalties were formed when convenient and rejected when they became a hindrance. For example, trusted followers eliminated two of the three aforementioned warlords when it was to their advantage. The tension between local and personal loyalties is demonstrated by the fact that some people were willing to uproot themselves from their towns and villages to follow a powerful warlord, but the pull of home became strong again when the war leader’s star was no longer rising. In addition, Sinocentric discourses contained in traditional literary works had a weak hold over the people of the frontier. For example, Gao Manzheng’s denigration of the Turks was only a smokescreen to mask his true intentions, and in any event did not persuade his master, Yuan Junzhang. Another example is Yuan’s delaying submission to the Tang for almost a decade after saying that “Heaven” had given the dynasty the “mandate” to rule China. This pillar of Confucian ideology did not influence his allegiance, which remained with the Turks as long as they could offer him protection. | 34 |
| Consequently, in these borderlands political loyalty was not defined by concepts of ethnic solidarity, cultural affinity, or absolute personal loyalty, but rather by pragmatic calculations of self-interest. Charles Peterson has called such allegiances “contractual,” since they were conditional on both parties fulfilling obligations. Peterson associates these values with Inner Asians, but obviously they were shared by Chinese in the borderlands (Peterson 1970–1971, 445). Because these were informal and unnegotiated “social contracts,” the parties involved had a great deal of latitude in judging whether the other was upholding commitments. Some, such as the Machiavellian Gao Kaidao, even entered into these unwritten agreements in bad faith. The negligible influence of core Confucian ideals—which during the Tang, as will be discussed below, encourage absolute loyalty to ruler or state—is not surprising if we consider that the power of the government to propagate ideology was at its weakest on the margins of empire, and contemporary literacy rates must have been very low.22 Even though local attachments had a stronger hold than Confucian ideals on inhabitants of the frontier, these too could be rejected when expedient. These pragmatic tendencies must have been magnified by the often violent and unpredictable nature of life on the frontier that would have encouraged self-preservation at the expense of idealism. This suggests that in the borderlands of the Sui-Tang transition, despite the imagined Chinese cultural unity of the dynastic histories and our ingrained assumptions of modern nationalism, group identity was not the dominant determinant of political loyalty. Rather, when given the opportunity to choose allegiances—as occurred during periods of dynastic transition, especially on the margins of empire—people entered into a “social contract” with someone, regardless of personal or ethnic affiliation, who could offer economic stability, protection, and sometimes patronage. | 35 |
| Loyalties in Comparative Perspective | |
| Although the above discussion suggests some important reasons for the fickleness of political allegiances in frontier areas, we need to adopt a wider perspective in order to draw firmer conclusions about the nature of solidarities in premodern borderlands. The dynamic interplay of geography, society, state power, and ideology could have effects on the loyalties of frontier inhabitants. In order to provide a more nuanced understanding of how variations in these factors could affect allegiances in borderland regions, and how the frontier in turn could influence perceptions of identity, this section will use a comparative approach that draws upon other cases from premodern history. | 36 |
| Looking at other frontiers, we find that, like the Sui-Tang borderlands, cultural, religious, or ethnic solidarities often did not inhibit cross-border relations, but governmental strength and policies could influence the identification of frontier inhabitants with the core polity. An example is the Roman empire, where elite writers depicted their frontiers as clear dividing lines between Romans and barbarians, but in reality borderlands exhibited confused identities and allegiances in which “ethnic distinctions, far from increasing, became blurred as contacts grew more frequent” (Whittaker 1994, 122, 131–133). This was a result of polices that, intentionally and unintentionally, encouraged integration of borderland natives into the Roman world. Many non-Romans purposely were drawn to Roman service and exposed to Roman ways. For example, non-Roman auxiliaries served in the army and indigenous elites, who were given Roman citizenship, administered local populations. In addition, the need to supply Roman urban centers and military bases in the borderlands created economic relations that unintentionally encouraged cultural exchanges, resulting in the “creative amalgamation of indigenous traditions and new ideas” (Wells 1999, 188–199, 221, 229). These political and cultural exchanges mostly worked to the advantage of the empire so long as the central government remained strong. However, political weakness or instability could lead to the deterioration of allegiances to the empire on the periphery. An example is the rebellion of Firmus on the North African frontier in 373, which elite political factionalism at the center helped ignite. Even though the Roman sources depict this as a war of barbarians against Romans, the revolt exhibits a number of cross-cutting conflicts and alliances. Firmus, who was a member of an elite indigenous family with Roman citizenship, was supported by some of his brothers, but not others. Along with most natives of the region, two regiments of Roman troops also gave allegiance to Firmus (Elton 1996, 131–135; Sivan 1996, 141–143). Thus we can see that a government policy of open frontiers encouraged a mixed ethnic situation that generally worked to the advantage of the imperial center. A powerful state had political and economic resources at its disposal that could make it the object of more solid political allegiances, though not necessarily on the basis of ethnic solidarity. However, when the central government could not exercise sufficient control over the periphery, like during the rebellion of Firmus and the Sui-Tang transition, the potential for cross-cultural alliances could work to the disadvantage of the state. | 37 |
| Returning to Chinese history, we can see a similar phenomenon at work as the Han and Tang dynasties maintained policies of open frontiers. During the first six decades of the Han, when central power was weak, a perennial problem for the imperial government was generals and enfeoffed borderland kings who defected to the nomadic Inner Asian Xiongnu empire to protect their autonomy. By the reign of Wudi (140– 87 b.c.e.), Han China had solidified its power. Feudatories were eliminated or had their independence curbed through the adoption of aggressive policies, and successful attacks on the Xiongnu convinced many nomads in the borderlands to switch allegiance to the Han (Di Cosmo 2002, 230–241). In the case of the Tang at the height of its power in the late seventh and early eighth centuries, the dynasty gained a firmer grip over frontier inhabitants than in the early years of rule. Although the Tang never attempted to close off their Inner Asian frontier, the government used legal, military, diplomatic, ideological, and administrative means to regulate the borderlands to its advantage. Incidents of Tang subjects going over to nomadic enemies en masse are unknown after the early years of the dynasty because the government did a relatively good job of controlling the agricultural populace. Still, instances of elite “traitors” making common cause with enemies are known, especially during periods of central government weakness (Skaff 1998, ch.6; Skaff forthcoming a). Mobile nomads presented a greater problem of control than sedentary peasants, but, repeating a strategy used by the Han and Sui, the Tang used a number of inducements to attract the loyalty of nomadic groups. In practice, their allegiance to the dynasty was typical of what we have seen for the Sui-Tang transition, being contingent upon imperial favor and rewards. Blessed with mobility, disgruntled nomads had more opportunities than peasants to renounce fealty to the Tang and retreat into Inner Asia. Still, the dynasty’s extensive resources attracted ample numbers of Inner Asians, allowing nomadic cavalry to be employed as a major component of military strategy (Peterson 1970–1971; Skaff 1998, especially 271–293; Skaff forthcoming b). | 38 |
| Thus, although the Han and Tang, like Rome, made no effort to eliminate all political and cultural interchanges in the frontier zone, their policies ensured that the majority of exchanges were to the advantage of the imperial state, so long as central power and resources remained strong. A weakened government was prone to lose control of allegiances on the frontier, as during the Sui-Tang transition, because it no longer could guarantee stability or compel subservience. Consequently, allegiances of borderland peoples to strong empires were ensured through a combination of oversight and a “contract” guaranteeing rewards or stability. The degree of state power was a strong predictor of whether a government could continue to enforce obedience and maintain “contractual” loyalties by providing social and economic security. | 39 |
| Although a state’s potency clearly had an important effect on political loyalties in the frontier zone, the role of ideology deserves further scrutiny because its influence often has been overestimated. Postmodernist scholars have made strong claims about the power of ideological “discourses” to affect loyalties and identities. For example, Richard Terdiman argues that “discourses are the complexes of signs and practices which organize social existence and social reproduction. In their structured, material persistence, discourses are what give substance to membership in a social group … [and] mediate an internal sense of belonging, and outward sense of otherness” (1985, 54). We also have seen that modern historians often tacitly assume the influence of ideology by accepting stereotypes in traditional histories that were born out of premodern “discourses” of the barbaric other. However, this paper has questioned the potency of these discourses by demonstrating that Sinocentric rhetoric had a much weaker hold over the inhabitants of the frontier during the Sui-Tang transition than over the minds of traditional historians who wrote while safely ensconced in the dynastic capital. | 40 |
| We need to investigate ideologies of loyalty further to understand the role they played in forming identities and political solidarities in premodern times. Were there conditions under which rhetoric propagated by governments or members of society could be brought to life in the minds of the majority of people so that they might feel subjective loyalty to their leaders or cultural affinity toward each other? The frontier can play an important role in helping to answer this question. Normally, people living in core areas of a polity were not challenged to accept ideals of loyalty. They were under firm governmental control and rarely would have had a chance to meet people from other cultures or even to contemplate switching political allegiances. On the other hand, opportunities to encounter people of other ethnicities and choose political loyalties were far more common in borderlands. The last part of this paper will investigate the role of frontiers in inducing premodern political ideologies of ethnic difference, and in turn the practical extent to which these belief systems could create feelings of political solidarity based upon cultural or religious affinity. | 41 |
| Social status was an important determinant of who was exposed to premodern political rhetoric. Ideologies of political loyalty or ethnic affinity certainly appear to have had a weak hold over lower classes. In the case of the fall of the Sui, we have seen that the majority of people who fled to the Turks were commoners seeking protection. This was not the only time in Chinese history that this phenomenon has been observed. For example, during another period of civil war in the tenth century, Chinese from the central plains fled north to the relative stability of the borderland Liao dynasty (907–1125) founded by the nomadic Khitans (Tao 1988, 26). This parallels the situation on the Roman and North American frontiers where people of lower status had a greater tendency to engage in cross-ethnic alliances because they were eager for financial gain or sought opportunities for upward mobility (Whittaker 1994, 188; Lightfoot and Martinez 1995, 486). Probably the most important reason why ideologies were not effective at creating a sense of political or ethnic solidarity among the common people, even in core areas of premodern polities, was because of the low levels of literacy and communications. Only a thin stratum of elites had the means to share ideas through books and letters over widely dispersed territory (Armstrong 1982, 6, 279). We can suppose that lower social classes, generally ignorant of ideology and whether on the frontier or not, normally did not see themselves as part of political solidarities extending outside of their local areas. Thus, belief systems were produced mainly by members of the social elite for consumption by their peers. | 42 |
| Still, we have seen from numerous cases above that there was not even a guarantee that literate elites would develop a strong allegiance to a polity or belief in ethnic solidarity. We can gain a sense of the contested nature of loyalty ideologies among elites and the role the frontier played in shaping ideals of political and cultural affinity by returning to Chinese history and then looking at two other cases drawn from Islamic frontiers. During the Tang, the dynasty encouraged an ideal of loyalty (zhong) that considered duty to the ruler (shang) and state (guo) to be more important than personal safety or family considerations. These ideas lacked any sense of cultural allegiance to a particularly “Chinese” state or ethnicity and instead involved purely political attachments between an individual and the ruler or state (THY 79: 1461; Standen 1997). Although non-Chinese enemies were berated as barbarians, if they surrendered to the Tang, which was common for borderland nomads, they were considered subjects in good standing as long as they served the dynasty loyally. The nonethnic basis of political solidarity is demonstrated by the 755 rebellion of the non-Chinese frontier general An Lushan. Even though some historians have attempted to portray it as a revolt of barbarians against the Chinese Tang, in reality the rebel and loyalist armies were ethnically mixed (Skaff 2000). The rebellion also exposed the elite’s lack of consensus about how to interpret loyalty ideals. After the insurrectionists captured the eastern capital of Luoyang and then the western capital of Chang’an, officials reacted differently. In Luoyang most entered the service of the rebels, but at Chang’an the majority fled rather than submit. After the Tang retook both capitals, the former group of officials found themselves on trial. In theory their actions could be justified by the idea that loyalty to the state took precedence over allegiance to a particular ruler, but in practice most seem to have claimed that they were forced to serve the rebels (Standen 1997). In cases of other areas captured by the insurrectionists, some officials fled or chose death over collaboration, but a substantial number served the rebels, apparently to ensure the safety of themselves and their families (Graff 1995, 12–13). | 43 |
| During subsequent decades, there were some new developments in loyalty ideals in reaction to the revolt. The dynasty, desperate to encourage allegiance because of increasing political decentralization, began to glorify those who had died for the sake of the Tang (Graff 1995). The perceived culpability of “barbarian” rebels also encouraged some new ideas about the relationship between ethnicity and political affiliation. Late Tang intellectuals, most notably Han Yu (768–824), adopted a stance against multiculturalism. Non-Chinese still were welcome to become part of the political life of the empire, but only if they assimilated to Chinese culture (Hartman 1986, 131–132, 158–159; Tao 1988, 40). Nonetheless, the shriller rhetoric does not appear to have been accepted uniformly. After the fall of the Tang, elite ideals of loyalty in North China remained nonexclusive during another chaotic period of civil war in the first half of the tenth century. For example, in those turbulent times the high official, Feng Dao (882–954), served six dynasties in succession that were founded by Chinese, Shatuo Turk, and Khitan rulers. He could operate in this way through the belief that his highest loyalty was to the state rather than individual rulers or dynasties. At the time, nothing was considered untoward about his serving multiple dynasties, including those with non-Chinese rulers (Wang 1962; Standen 1999). | 44 |
| Elite definitions of loyalty became increasingly exclusive during the Song dynasty (960–1276). Emphasis began to be placed on absolute political loyalty to a single dynasty and there was a growing tendency to encourage cultural allegiance to China. The context for these developments was that the Song was threatened by non-Chinese rulers who established militarily powerful, Chinese-influenced dynastic states in North China (Tao 1988, 40–42; Jay 1991, 93–98, 250–251). Ironically, both sides considered themselves to be the true inheritors of the Tang’s traditions with legitimate right to rule over China (Chan 1984, 36–42, 117–123; Franke 1987, 90–91; Tao 1988, 25–31). Significantly, the northern regimes displaced the Song from the ecological and cultural transition zone dividing agricultural China from nomadic Inner Asia, which would encourage changes in the outlook and habits of Song Chinese (Liu 1985). Frontier policies demonstrated continuities and significant changes. The Khitan Liao dynasty, which controlled the borderlands, retained the Tang tradition of open frontiers facing Inner Asia (Twitchett and Tietze 1994, 138). On the other hand, in 1004 the Liao and Song signed a treaty that created a more tightly monitored border between the two. In an attempt to restrict the frontier as a zone of exchange, both sides agreed to return “fugitives.” Although at first the border was ill-defined in some regions, it was clearly marked after a border dispute in 1074–1075. This was the first time in Chinese history that a linear frontier had been demarcated and regulated.23 Moreover, partly at the instigation of the Song state, Chinese elites began to have a greater awareness of ethnic identity and some felt uncomfortable about serving the Liao (Tao 1988, 34–52; Standen 1999, 73–76). Officials such as Feng Dao, who had exhibited mixed allegiances prior to the Song, came to be criticized by some, but not all, Song scholars. Unfortunately for his later reputation, the ones who criticized him became more influential with the passing of time (Wang 1962, 205–208). | 45 |
| After the Song lost even more of the north to the Jurchen Jin Dynasty (1115–1234) and the court was forced to move the capital south of the Yangtze River, once again there was an attempt to enforce a linear border, in this case along the Huai River. Xenophobia became a stronger, but still not all-encompassing component of the Song loyalist movement (Tillman 1979, 420–426; Jay 1991, 250–251). Not surprisingly, these attitudes were not prevalent in the Jurchen realm. In the multicultural milieu of North China, Jurchen rulers promoted loyalty to the state, demonstrating continuity with Tang traditions (Jay 1991, 254–255; Bol 1987, 483–493). Most northern Chinese appear to have accepted Jin rule. In 1206 when the Song minister Han Tuozhou organized an invasion of Jin territory, which a century earlier had been in the hands of the Song, the local populace did not rise up in support of the Song armies as Han and others had expected. Although Trauzettel has argued that the southern Song elites and the state successfully indoctrinated the populace in patriotic ideals and that these feelings reached a fervid pitch during the warfare of 1206, the facts do not bear out his claims (1975, 204–208). Tens of thousands of Song troops deserted, and the Song even suffered an elite defection when the Song governor-general of Sichuan, Wu Xi, gave his allegiance to the Jin. However, he soon fell victim to a coup by subordinates and Sichuan returned to the Song fold (Franke 1994, 247–248). | 46 |
| It is apparent that an ideology encouraging ethnic solidarity became more prevalent in Song China during a time of a perceived external threat. This development was not reciprocated in North China. Nonetheless, these intellectual changes remained contested among Song elites and did not trickle down to the great mass of commoners in society to create something akin to modern nationalism. As we have seen above, inhabitants of North China did not respond to the Song invasion of 1206. About a decade later, there were a number of rebellions against Jin rule in Shandong, which sometimes have been labeled as “patriotic” because many rebels aligned themselves with the Song. However, we can see that the insurrectionists had more pragmatic motivations, like warlords during the Sui-Tang transition, because many eventually switched allegiance to the alien Mongols after the latter invaded North China (Aubin 1987, 121–132; Franke 1994, 256–257). The inability of the lower classes in Song China to relate to ideas about ethnic solidarity has led Tillman to refer to this intellectual movement as “protonationalism” because it was an elite phenomenon (Tillman 1979). Yet even among the official class, these ideas were far from universally accepted. Not all scholars who wrote about ideals of political allegiance stressed ethnicity as a basis for solidarity (Jay 1991, 250–251). When ideals were tested in practice, some elites did not take ethnicity into consideration when making political alliances. The case of Wu Xi is mentioned above. Another example is the northern Song official Liu Yu (1079–1143?), who defected to the Jin after their invasion of North China (Franke 1994, 230–232). | 47 |
| The above discussion illustrates the contested nature of loyalty ideals among mid-imperial Chinese elites. One area of controversy revolved around whether higher loyalties were owed to the emperor or state. Another dispute involved whether ethnicity was a basis for political allegiance. In practice, some elites were moved by these ideals to give up their lives for a ruler, dynasty, or country, but others avoided becoming a sacrifice. Despite the lack of elite consensus, the most noticeable trend is the growing belief that political solidarity should be based upon ethnic affiliation. The Tang dynasty, with open but well-defended northern frontiers and a policy of accepting nomads into their armies, never encouraged ethnicity as a basis of loyalty. Later, as Song China found itself increasingly threatened and losing more territory, many elites, partly at the instigation of the state, began to define themselves in relation to the externally menacing “other” on the opposite side of a tense and tightly regulated border. The process of ethnic definition under these circumstances accords with the observations of historians and anthropologists who have noted that perceptions of self in relation to others play an important role in creating ethnic identity (Keyes 1981, 7, 27; Sahlins 1989, 9, 269–276). This phenomenon probably was encouraged by the Song’s removal from the China–Inner Asia ecological and cultural transition zone. On the other hand, older notions of ethnically blind loyalism to the state remained prevalent in North China, where the Inner Asian frontier stayed open and many ethnic Chinese served Khitan and then Jurchen masters in governmental systems that were partly influenced by Chinese norms. Ideas of ethnic solidarity did not take hold in the north possibly because of the unwillingness of the rulers and inhabitants of this multiethnic society to view Song Chinese as the “other” and also because of a perceived lack of threat from the Song military, which never had much long-term success in recapturing lost territory. Ironically, the Liao and Jin dynasties were not as alien as some xenophobic Song elites imagined, and in terms of society and government were actually far less foreign than the Turks and other pastoral nomads who had confronted Sui and Tang China. What had changed was that the frontier had moved south, away from the Inner Asian borderlands, and had become more tense and less permeable, and the Song’s perceived threat from the other side had increased. Thus, ethnically based ideals of political loyalty were not a primordial component of Chinese ideology, but arose in mid-imperial China as a result of highly contingent historical conditions. Still, these new political discourses were not universally accepted because of the inappropriateness of an ethnically exclusive ideology in multicultural North China, a lack of consensus among Song elites, and their inability to propagate these ideas among the great mass of people. | 48 |
We can gain further insight into the contingent nature of premodern political solidarities by looking at two contrasting cases along the frontiers of Islam in Iberia and southeastern Anatolia. The Iberian frontier zone from the eighth to eleventh centuries presents a familiar picture of mixed ethnicities, identities, and political affiliations. Although the Islamic sources paint an image of a clear division between Muslim holy warriors and “infidel” Christian kingdoms, the reality was far different. The Iberian Umayyad dynasty (756–1031), which ruled the southern half of the peninsula, had only a loose reign over the Arab, Berber, and indigenous convert aristocratic families who controlled the borderlands. The loyalties of the frontier aristocrats were constantly shifting as they engaged in relations with the Umayyads, Christian kingdoms, and each other. Sharing only an aversion to central control, self-interest was more important than ethnic or religious affiliation in determining political alliances (Moreno 1999, 48–52). The situation on this frontier should give pause to those who assume that an ideology of jih d, in its guise as holy war, has always been an essential part of Muslim political life. 24 Clearly, the limited power of the Iberian Umayyad state played a role in its inability to regulate the frontier and enforce political loyalties more effectively. | 49 |
| On the other side of the Mediterranean, the Arab-Byzantine frontier along the Taurus mountains developed a firmer fit between identities and political solidarities. Initially, after the Islamic conquests in the seventh century, this frontier had much in common with the one in Iberia. A number of political actors from both sides fled to the other as a result of internal conflicts and factionalism (Kaegi 1992, 248–249; Kaegi 1996, 91–92). In response, the Umayyad caliphate (661–750), based in Syria, created a wasteland along the frontier, including an atypical Muslim policy of removing most of the indigenous population to prevent cross-border contacts (Haldon and Kennedy 1980, 106–107, 110). Another familiar pattern emerged during the transition between the Umayyad and ‘Abbsid (749–1258) caliphates in the early 750s. Warlords arose with Muslim and Christian followers who preyed indiscriminately upon the population regardless of ethnic or religious affiliation (Bonner 1996, 48–49, 68). Once again, weak central control appears to be the most important factor in explaining the prevalence of cross-cutting alliances on this frontier. | 50 |
The situation began to evolve in a new direction during the first century of ‘Abbsid rule when the frontier emerged as a barrier and marker of difference. This process was facilitated by the topography along the Taurus mountains, which, though not a perfect impediment, made it easier for both sides to regulate contacts. The idea that this frontier’s purpose was to restrict access to the other side is reflected in its popular Muslim name, al-Thugh r, which literally means “throats” or “mountain passes,” and by extension began to refer to the frontier posts guarding the passes and then the entire frontier itself (Kaegi 1992, 240–243; Bonner 1996, 188). Nonetheless, geography alone cannot explain the emerging cultural division, because communications through the mountains always remained feasible and the two sides never developed a linear border. Heavy social and political intervention was needed to create a cultural divide. Perhaps the key element in this process was that the Arab-Byzantine frontier developed a special ideological significance for Islam because Byzantium was perceived to be its oldest and greatest enemy. Government and society played roles in turning this into the primary front of holy war. The ‘Abbsid caliph, Hrn al-Rash d (r.786–809), attempted to increase central control over this frontier by stationing professional soldiers there to create the most sophisticated defensive system in Islam and by promoting an ideology of caliph as leader of the holy war against the Byzantines. Annual campaigns against Byzantium yielded little in terms of territory or booty, but served a ritualistic function. The religious significance of the raids is demonstrated by the fact that in historical works the commander of the summer expedition is often listed side-by-side with that of the leader of the annual pilgrimage to Mecca. Although later caliphs did not consistently involve themselves in jihd, they continued a policy of supporting the professional military on this frontier (Haldon and Kennedy 1980, 106–115; Bonner 1996, 137–147). | 51 |
| Ultimately, however, forces within Muslim society were more influential in making holy war against the Byzantines a sustained reality. Religious scholar-ascetics and volunteer fighters from all parts of the Islamic world were attracted to the Byzantine frontier. They outnumbered the civilians in the formerly depopulated cities, and even came if their hometowns were in other borderland regions. For example, ‘Abdallah ibn al-Mubrak (d. 797), from Marv in eastern Iran and author of the earliest book on jihd, participated in the summer campaigns against the Byzantines every other year, but is not known to have fought on the eastern frontier, which was closer to his home. The volunteer warriors had wide social support for their efforts because they were maintained by pious donations from their places of origin. Scholar-ascetics, such as Ibnal-Mubrak, acted as spokesmen for the fighters, wrote some of the earliest and most influential treatises on jihd, and consequently were instrumental in developing the mature doctrine of holy war (Bonner 1992, 19–31; Bonner 1996, 119–125, 145–155). The end result of their efforts was a cultural divide between Islam and the Byzantines that grew increasingly difficult to cross. Although commercial contacts never ceased and there appeared to be some elements of common culture on both sides of the frontier, the ideological fervor of the mostly warrior borderland inhabitants made cross-frontier political alliances much more difficult to contemplate (Haldon and Kennedy 1980, 106–110;Bonner 1996, 137–139). | 52 |
| The hardening of the Arab-Byzantine borderlands must be considered a special case within Islamic history, just as the Song’s northern borders were atypical in Chinese history. What made the Byzantine frontier exceptional? One factor is geography, which provided a mountainous barrier that made cross-frontier contacts more controllable along a long zone. Aside from the physical division made possible by topography, perceptions of the Byzantines as the “other” were encouraged by political and social forces, with the latter being more important. The government removed most of the indigenous population and later transferred in professional soldiers organized to defend the frontiers and carry out holy war. After the reign of Hrn al-Rashd, even though ‘Abbsid rulers did not consistently pursue policies of jihd, society provided a constant impulse encouraging holy war against the Byzantines. Religious zealots supported warfare either by volunteering to serve on the frontier or by providing donations to support the fighters. Through the intervention of government and development of networks of self-selected individuals, borderland society became highly militarized, composed mainly of dedicated holy warriors. Perhaps what is most significant about this was the role of religion. As Armstrong has pointed out, organized religion was an important influence on identity in premodern times because networks of holy men could penetrate a society more deeply than most political administrations (Armstrong 1982, 238). On the Arab-Byzantine frontier, the intense activities of Muslim scholar-ascetics working among volunteer warriors already predisposed to their views would have helped to encourage political solidarity, based upon religious identification, that was hostile toward Christians on the other side. These efforts were facilitated by the earlier removal of the indigenous Christian population, which probably would have been more ambivalent to these efforts, even if they had converted. | 53 |
| Nonetheless, organized religion was not equally effective in all parts of the Muslim world in terms of creating political solidarity based upon religious identity. Thus, the Iberian borderlands were not greatly influenced by ideals of holy war, even though the Iberian Umayyad dynasty and its historians painted it as a promoter of these values. A relatively weak central government, lacking the aid of geographical barriers along its frontier, could not enforce political loyalties effectively among borderland aristocrats. More importantly, this frontier’s relative lack of prestige probably limited the number of zealous religious scholars and fighters who were attracted to it. Without an extensive socioreligious network encouraging jihd, borderland warriors were less likely to be motivated by ideology. In addition, frontier inhabitants probably were less receptive to negative ethnic stereotypes about the enemy because the indigenous population with its pre-existing cross-frontier contacts remained in place. Although the Iberian and Byzantine borderlands both were perceived as barriers between Islam and infidels, only the Byzantine frontier induced political solidarity based upon a common identity. This was primarily because Byzantium’s prestige as an enemy led to the removal of the indigenous population and attracted unusually dedicated and active socioreligious networks. | 54 |
| Conclusion | |
| Most premodern frontiers were not natural political and cultural boundaries, but rather were vaguely defined zones that encouraged dynamic interaction among a variety of peoples. Spheres of cultural, political, and economic influence fluidly blended into one another. In general, inhabitants of frontier zones had greater freedom to choose political affiliations than people living in core areas of government control. When left to their own designs, as often was the case in borderlands such as those of the Sui-Tang transition, people tended to forge political ties based on self-interest, personal connections, and local affiliation, with self-interest being most prominent. | 55 |
| Nonetheless, a number of factors could increase the identification of people in frontier regions with core states and societies. Governments played an important role in encouraging or coercing the loyalty of people living in borderlands. Strong, centralized governments, such as Rome and Tang China at their heights, could more successfully supervise frontier inhabitants and maintain social stability, which would both compel obedience and encourage “contractual” loyalty to, or at least passive acceptance of, their regimes. Moreover, frontier policy could have an important influence on the nature of elite loyalty ideals. Empires that allowed their borderlands to be zones of exchange, such Rome and Tang China, did not attempt to enforce or encourage a cultural divide between the empire and outsiders. In Tang China loyalty was defined in purely political terms and ethnicity generally was not a barrier to participation in the political life of the realm. Although denigration of “barbarian” foreigners was common in elite circles, this rhetoric does not seem to have had much practical influence on governmental policy or the perceptions of people living in frontier society. | 56 |
| Other regimes, such as the Song or ‘Abbsid dynasties, tried to tightly regulate and defend parts of their borderlands, and more actively encouraged religiously or ethnically based ideologies hostile toward an adversary. In these cases the border, by causing separation, was an important element in the symbolic construction of identity in relation to the enemy. However, this was not a sufficient condition to create a group identity that could become the basis for political solidarity. In Song China the existence of an outside threat and lack of contact with Inner Asia encouraged elites to think about themselves in terms of cultural or ethnic affinity, but these ideals were not shared universally among the upper class, nor did they trickle down to the general populace on a large scale. Although some modern scholars, such as Duara, argue that the Song state cultivated “loyalty to the fatherland in peasant communities” via means such as orally transmitted myths, the evidence presented above does not support these claims. Still, Duara correctly perceives the possibility of national consciousness in premodern times (Duara 1995, 51–69). To create a more broadly shared sense of identity, the key ingredient was the existence of political or social networks that could carry the message. Even in the modern era, in which ethnic awareness is far more prevalent, it normally becomes a basis for political action only as a result of institutionalization and organization (G. Bentley 1987, 40–43). In most preindustrial societies, people could not be knit together effectively on a large scale because of an absence of mass communication, transportation, and educational systems. 25 Without these modern amenities, the concentrated and sustained effort needed to broadly transmit ideals of group identity rarely could be produced by governments, organized religions, or social networks. Islam’s Byzantine frontier was an exception. A stronger Muslim identity was maintained because of the long-term fiscal and human support that Islamic society voluntarily lavished on holy war against an enemy with special religious significance. Thus, even though Song China produced ideals of ethnic solidarity against an enemy across a fortified boundary, these were not popularized because government and elite society never developed a means to promote the message systematically to the great mass of people. | 57 |
| Consequently, we can conclude that the hazy frontiers, identities, and loyalties of the Sui-Tang transition were typical of the premodern era. A state’s most effective means of dispelling some of the fog and winning allegiances was effective centralized administration that could provide social stability, rewards, and oversight. However, the ideological goal of creating widespread identification of people in frontier regions with core governments and societies was a chimera that sometimes was pursued, but rarely became reality.BibliographyPrimary SourcesJTSLiu Xu, et al.[945] 1975. Jiu Tang shu [Old Tang history]. 16 vols. Beijing: Zhonghua shuju.SSWei Zheng.[656] 1973. Sui shu [History of the Sui].6 vols. Beijing: Zhonghua shuju.THYWang Pu.[961] 1991. Tang hui yao [Tang administrative documents]. 4 vols. Taibei: Taiwan shangwu yinshuguan.XTSOuyang Xiu. [1060] 1975. Xin Tang shu [New Tang history]. 20 vols. Beijing: Zhonghua shuju.ZGZYWu Jing (670–749). 1994. Zhenguan zhengyao [The Zhenguan reign’s essentials of government]. Edited by Wu Gui. Changsha: Yuelu shushe.ZZTJSima Guang.[1084] 1956. Zizhi tongjian [Comprehensive mirror for the aid of government].10 vols. Beijing: Zhonghua shuju. Secondary Works and Translations | 58 |
Anderson, Benedict.1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London, New York: Verso.Armstrong, John Alexander.1982. Nations before Nationalism. Chapel Hill: University of North Carolina Press.Aubin, Françoise.1987. “The Rebirth of Chinese Rule in Times of Trouble: North China in the Early Thirteenth Century.” In Foundations and Limits of State Power in China, edited by S. R. Schram. London: School of Oriental and African Studies, University of London.Barfield, Thomas. 1989. The Perilous Frontier: Nomadic Empires and China. Oxford, U.K. and Cambridge, Mass.: Blackwell.Barth, Fredrik, ed.1969. “Introduction.” In Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Boston: Little, Brown and Company.Bentley, G. Carter.1987. “Ethnicity and Practice.” Comparative Studies in Society and History 29:24–55.Bentley, Jerry H. 1993. Old World Encounters: Cross-Cultural Contacts and Exchanges in Pre-modern Times. New York: Oxford.———.1996. Shapes of World History in Twentieth-Century Scholarship. Edited by M. Adas, Essays in Global and Comparative History. Washington: American Historical Association.Benveniste, Emile. 1969. La vocabulaire des institutions Indo-Européennes. 2 vols. Paris: Éditions de Minuit.Bingham, Woodbridge.[1941] 1970. The Founding of the T’ang Dynasty. New York: Octagon Books.Bol, Peter K.1987. “Seeking Common Ground: Han Literati under Jurchen Rule.” Harvard Journal of Asiatic Studies 47:461–538.Bonner, Michael.1992. “Some Observations Concerning the Early Development of Jih d on the Arab-Byzantine Frontier.” Studia Islamica 75:5–31.———. 1996. Aristocratic Violence and Holy War: Studies in the Jihad and the Arab-Byzantine Frontier, American Oriental Series, 81. New Haven, Conn.: American Oriental Society.Chan, Hok-lam. 1984. Legitimation in Imperial China: Discussions under the Jurchen-Chin Dynasty (1115–1234). Seattle: University of Washington Press.Chang, Shelley Hsueh-lun. 1990. History and Legend. Ann Arbor: University of Michigan Press.Chavannes, Édouard. [1900] 1969. Documents sur les Tou-kiue (Turcs) Occidentaux, recueillis et commentés suivi de notes additionnelles. Taipei: Ch’eng Wen Publishing.Di Cosmo, Nicola.1994. “Ancient Inner Asian Nomads: Their Economic Basis and Its Significance in Chinese History.” The Journal of Asian Studies 53:1092–1126.———. 2002. Ancient China and Its Enemies: The Rise of Nomadic Power in East Asian History. Cambridge, U.K.: Cambridge University Press.Donnan, Hastings, and Thomas M. Wilson, eds. 1994. Border Approaches: Anthropological Perspectives on Frontiers. Lanham, Md.: University Press of America.Drompp, Michael R. 1991. “Supernumerary Sovereigns: Superfluidity and Mutability in the Elite Power Structure of the Early Türks.” In Rulers from the Steppe: State Formation on the Eurasian Periphery, edited by G. Seaman and D. Marks. Los Angeles: Ethnographics Press.Duara, Prasenjit.1995. Rescuing History from the Nation: Questioning Narratives of Modern China. Chicago: University of Chicago Press.Eisenberg, Andrew.1997. “Warfare and Political Stability in Medieval North Asian Regimes.” T’oung Pao 83:300–328.Elton, Hugh.1996. “Defining Romans, Barbarians, and the Roman Frontier.” In Shifting Frontiers in Late Antiquity, edited by R. W. Mathisen and H. S. Sivan. Brookfield, Vt.: Variorum.Fairbank, John King.1968. “A Preliminary Framework.” In The Chinese World Order, edited by J. K. Fairbank. Cambridge, Mass.: Harvard University Press.Febvre, Lucien.[1924] 1949. A Geographical Introduction to History. Translated by E. G. Mountford and J. H. Paxton. London: Routledge and Kegan Paul.Franke, Herbert.1987. “The Role of the State as a Structural Element in Poly-ethnic Societies.” In Foundations and Limits of State Power in China, edited by S. R. Schram. London: School of Oriental and African Studies, University of London.———. 1994. “The Chin Dynasty.” In The Cambridge History of China, Volume 6, Alien Regimes and Border States, 907–1368, edited by H. Franke and D. Twitchett. Cambridge, U.K.: Cambridge University Press.Gellner, Ernest. 1983. Nations and Nationalism. Ithaca: Cornell University Press.Graff, David A.1995. “Meritorious Cannibal: Chang Hsün’s Defense of Sui-yang and the Exaltation of Loyalty in an Age of Rebellion.” Asia Major 3rd Series, 8:1–18.———. 2002. “Strategy and Contingency in the Tang Defeat of the Eastern Turks, 629–30.” In Warfare in Inner Asian History (500–1800), edited by N. Di Cosmo. Leiden: Brill.Haldon, J. F., and H. Kennedy.1980. “The Arab-Byzantine Frontier in the Eighth and Ninth Centuries: Military Organization and Society in the Borderlands.” Recueil des travaux de l’Institut d’études byzantines 19:79–116.Hartman, Charles.1986. Han Yu and the Tang Search for Unity. Princeton, N.J.: Princeton University Press.Jagchid, Sechin, and Van Jay Symons. 1989. Peace, War, and Trade along the Great Wall: Nomadic-Chinese Interaction through Two Millennia. Bloom-ington: Indiana University Press.Jay, Jennifer W.1991. A Change in Dynasties: Loyalism in Thirteenth-Century China. Bellingham: Center for East Asian Studies, Western Washing-ton University.Kaegi, Walter E.1992. Byzantium and the Early Islamic Conquests. New York: Cambridge University Press.———. 1996. “Reconceptualizing Byzantium’s Eastern Frontiers in the Seventh Century.” In Shifting Frontiers in Late Antiquity, edited by R. W. Mathisen and H. S. Sivan. Brookfield, Vt.: Variorum.Keyes, Charles F.1981. “The Dialectics of Ethnic Change.” In Ethnic Change, edited by C. F. Keyes. Seattle: University of Washington Press.Lamouroux, Christian.1997. “Geography and Politics: The Song-Liao Border Dispute of 1074/75.” In China and Her Neighbors: Borders, Visions of the Other, Foreign Policy 10th to 19th Century, edited by S. Dabringhaus and R. Ptak. Wiesbaden: Harrassowitz Verlag.Lattimore, Owen. [1951] 1962. Inner Asian Frontiers of China. 2nd ed. Boston: Beacon Press.Lightfoot, Kent G., and Antoinette Martinez.1995. “Frontiers and Boundaries in Archaeological Perspective.” Annual Review of Anthropology 24:471–492.Liu, James T. C. 1985. “Polo and Cultural Change: From T’ang to Sung China.” Harvard Journal of Asiatic Studies 45:203–224.Mattingly, D. J., ed.1997. Dialogues in Roman Imperialism: Power, Discourse, and Discrepant Experience in the Roman Empire. Portsmouth, R.I.: Journal of Roman Archaeology, Supplementary Series Number 23.Minyaev, S. n.d. “Archaeology of the Hsiung-nu in Russia—New Discoveries and Some Problems” [Web article]. Silkroad Foundation. Available from <http://www.silk-road.com/artl/minyaev1.shtml>. Accessed 30 July 2001.Moreno, Eduardo Manzano. 1999. “The Creation of a Medieval Frontier: Islam and Christianity in the Iberian Peninsula, Eighth to Eleventh Centuries.” In Frontiers in Question: Eurasian Borderlands, 700–1700, edited by D. Power and N. Standen. New York: St. Martin’s Press.Olster, David.1996. “From Periphery to Center: The Transformation of Late Roman Self-Definition in the Seventh Century.” In Shifting Frontiers in Late Antiquity, edited by R. W. Mathisen and H. S. Sivan. Brookfield, Vt.: Variorum.Pan, Yihong.1997. Son of Heaven and Heavenly Qaghan: Sui-Tang China and its Neighbors. Bellingham: Western Washington University Press.Perdue, Peter C. Forthcoming. “From Turfan to Taiwan: Trade and War on Two Chinese Frontiers.” In Frontiers Through Space and Time, edited by B. Parker and L. Rodseth. Salt Lake City: University of Utah Press.Peterson, Charles A.1970–1971. “P’u-ku Huai-en and the T’ang Court: The Limits of Loyalty.” Monumenta Serica 29:423–455.Pohl, Walter, ed. 1998. Strategies of Distinction: The Construction of Ethnic Communities, 300–800, Transformation of the Roman World, Vol. 2. Lei-den: Brill.Power, Daniel, and Naomi Standen.1999. Frontiers in Question: Eurasian Borderlands, 700–1700. New York: St. Martin’s Press.Pratt, Mary Louise.1992. Imperial Eyes: Travel Writing and Transculturation. London and New York: Routledge.Pulleyblank, Edwin G.1952. “A Sogdian Colony in Inner Mongolia.” T’oung Pao 41:317–356.Sahlins, Peter.1989. Boundaries: The Making of France and Spain in the Pyrenees. Berkeley: University of California Press.Sinor, Denis.1978. “The Greed of the Northern Barbarians.” In Aspects of Altaic Civilizations II, edited by L.V. Clark and P. A. Draghi. Bloomington: Indiana University Press.Sivan, Hagith.1996. “Why Not Marry a Barbarian? Marital Frontiers in Late Antiquity.” In Shifting Frontiers in Late Antiquity, edited by R. W. Mathisen and H. S. Sivan. Brookfield, Vt.: Variorum.Skaff, Jonathan Karam.1998. “Straddling Steppe and Sown: Tang China’s Relations with the Nomads of Inner Asia (640–756).” Doctoral dissertation, University of Michigan.———. 2000. “Barbarians at the Gates? The Tang Frontier Military and the An Lushan Rebellion.” War and Society 18 (2):23–35.———.2002. “Western Turk Rule of Turkestan’s Oases in the Sixth through Eighth Centuries.” In The Turks, edited by H. Inalcik. Ankara: Yeni Türkiye.———. Forthcoming a. “Loyalties Divided: The Question of Ethnicity in the Tang-Türgish Conflict of 708–9.” In Battlefronts Real and Imagined: War, Border, and Identity in the Chinese Middle Period, edited by D. Wyatt. New York: Palgrave at St. Martin’s Press.———. Forthcoming b. “Tang Military Culture and Its Inner Asian Influences.” In Military Culture in Imperial China, edited by N. Di Cosmo. Cambridge, MA: Harvard University Press.Smith, John Masson.1978. “Turanian Nomadism and Iranian Politics.” Iranian Studies 11:57–81.Somers, Robert M.1990. “Time, Space, and Structure in the Consolidation of the T’ang Dynasty (a.d. 617–700).” In State and Society in Early Medieval China, edited by A. E. Dien. Stanford, Calif.: Stanford University Press. This is a revised version of an article that originally appeared in The Journal of Asian Studies 45 (1986): 971–994.Standen, Naomi.1997. “The Trials of the An Lushan Collaborators: Loyalty and Identity in Theory and Practice.” Paper delivered at the Association of Asian Studies Annual Meeting, Chicago.———. 1999. “(Re)Constructing the Frontiers of Tenth-Century North China.” In Frontiers in Question: Eurasian Borderlands, 700–1700, edited by D. Power and N. Standen. New York: Palgrave.Tao, Jing-shen.1988. Two Sons of Heaven: Studies in Sung-Liao Relations. Tucson: University of Arizona Press.Tekin, Talat. 1968. A Grammar of Orkhon Turkic. Bloomington: Research Center for the Language Sciences, Indiana University.Terdiman, Richard.1985. Discourse/Counter-Discourse: The Theory and Practice of Symbolic Resistance in Nineteenth-Century France. Ithaca: Cornell University Press.Tillman, Hoyt Cleveland. 1979. “Proto-Nationalism in Twelfth-Century China? The Case of Ch’en Liang.” Harvard Journal of Asiatic Studies 39:403–428.Trauzettel, Rolf. 1975. “Sung Patriotism as a First Step toward Chinese Nationalism.” In Crisis and Prosperity in Sung China, edited by J. W. Haeger. Tucson: University of Arizona Press.Twitchett, Denis.1961. “Chinese Biographical Writing.” In Historians of China Japan, edited by W. G. Beasley and E. G. Pulleyblank. London: Oxford University Press.———. 1962. “Problems of Chinese Biography.” In Confucian Personalities, edited by A. F. Wright and D. C. Twitchett. Stanford, Calif.: Stanford University Press.Twitchett, Denis, and Klaus-Peter Tietze.1994. “The Liao.” In The Cambridge History of China, Volume 6, Alien Regimes and Border States, 907–1368, edited by H. Franke and D. Twitchett. Cambridge, U.K.: Cambridge University Press.Tyan, E.1960. “Djihd.” In Encyclopedia of Islam. 2nd ed. Leiden: E. J. Brill.Wang, Gung-wu.1962. “Feng Tao: An Essay on Confucian Loyalty.” In Confucian Personalities, edited by A. Wright and D. Twitchett. Palo Alto, Calif.: Stanford University Press.Wechsler, Howard. 1979. “The Founding of the T’ang Dynasty: Kaotsu (Reign 618–26).” In The Cambridge History of China, Vol. 3, Sui and T’ang China, 589–906, Part I, edited by D. Twitchett. Cambridge, U.K.: Cambridge University Press.Wells, Peter S.1999. The Barbarians Speak: How the Conquered Peoples Shaped Roman Europe. Princeton, N.J.: Princeton University Press.White, Richard. 1991. The Middle Ground: Indians, Empires, and Republics in the Great Lakes Region, 1650–1815. New York: Cambridge University Press.Whittaker, C. R.1994. Frontiers of the Roman Empire: A Social and Economic Study.Baltimore: Johns Hopkins Press.Wolf, Eric R.1982. Europe and the People Without History. Berkeley: University of California Press.Wright, Arthur F.1978. The Sui Dynasty. New York: Alfred H. Knopf.———. 1979. “The Sui Dynasty (581–617).” In The Cambridge History of China, Vol. 3, Sui and T’ang China, 589–906, Part 1, edited by D. Twitchett. Cambridge, U.K.: Cambridge University Press.Yang, Lien-sheng.1961. “The Organization of Chinese Official Historiography: Principles and Methods of the Standard Histories from the T’ang the Ming Dynasty.” In Historians of China Japan, edited by W. G. Beasley and E. G. Pulleyblank. London: Oxford University Press.———. 1968. “Historical Notes on the Chinese World Order.” In The Chinese World Order, edited by J. K. Fairbank. Cambridge, Mass.: Harvard University Press. | |
Notes* I owe thanks to Albert Dien, Mark Elliott, David Graff, Valerie Hansen, Naomi Standen, and an anonymous reviewer for their helpful comments on earlier drafts of this article. The paper originally was presented in San Francisco at the 2002 American Historical Association annual meeting, which was organized around the theme of frontiers. I would like to express my gratitude toward the program committee for organizing a thought-provoking conference that allowed me to place my research on the Chinese frontier in a comparative context and inspired the world history focus of this article. Last but not least, I want to express appreciation to my colleagues in the Shippensburg University History/Philosophy Department for providing an intellectual environment that encourages thinking about world history.1. For a historiographical review of major scholarship on cultural contacts, see J. Bentley 1996, 22–24. Among premodern historians, those studying Rome have been the most interested in studying cross-cultural contacts, see Whittaker 1994; Mattingly 1997; Pohl 1998; and Wells 1999. Important comparative works are J. Bentley 1993 and Power and Standen 1999.2. Strong arguments for this revisionist position are Sahlins 1989; White 1991; Whittaker 1994; Donnan and Wilson 1994; Lightfoot and Martinez 1995; and Power and Stan-den 1999. This view predominated at the 2002 American Historical Association annual meeting.3. Major English-language works on China’s relations with Inner Asian pastoral nomads are Lattimore 1951; Barfield 1989; and Jagchid and Symons 1989.4. Contemporary Turkic sources do not exist for the period under study here. The earliest surviving Turkic written records are the eighth-century Orkhon inscriptions in Mongolia (Tekin 1968).5. Lattimore 1951, 238–251. One reason why Lattimore’s theory may have been ignored by Sinologists immersed in the Chinese textual tradition is that it was based more on his experiences in the Chinese frontier regions rather than on research in the traditional Chinese written sources.6. On the typical decentralization of nomadic politics, see Smith 1978, 60.7. Di Cosmo (2002, 251) points out that early settlements in Mongolia probably did not have “very large productive bases.”8. JTS 3:41; XTS 2:32; ZZTJ 193:6087. Eighty thousand Chinese were counted in 631 after being redeemed in the wake of the Tang defeat of the Turks. It is difficult to distinguish the number that went willingly to the Turks. The sources usually use the verbmo to refer to the movement of these people. It can mean to disappear, or be captured. Its use here may reflect the anti-Turk polemics at the Tang court or the bias of later Chinese historians. An example of polemical usage comes in the New Tang History, compiled four centuries after these events, which in describing the ten thousand Chinese who originally went to the Turks at the end of the Sui uses the termmo to refer to their movement (XTS 215a:6029). The other histories avoid this usage. For example, the History of the Sui, completed about two decades after these events, clearly states that these Chinese went of their free will when it says that they gave their allegiance (gui) to the Turks (SS 84:1876). Two other sources, composed several centuries after the events, are neutral, merely stating that they were in the north (ZZTJ 188:5878; JTS 194a:5154). In discussing the movement of people away from China after the founding of the Tang, use of the verbmo becomes the norm. This may represent polemical usage because it is based on records compiled by Tang historians, who would not have been able to admit that people fled the dynasty. Nonetheless, some Chinese definitely were seized against their will. For example, in 626 the Tang emperor Gaozu demanded the return of all captured Chinese households, including the Tang official Wen Yanbo, who had been taken in battle (ZZTJ 192:6021; JTS 194a:5158). The Turks also took captives in their annual raids on early Tang China (Graff 2002, 36). Unfortunately, it is impossible to determine percentage of Chinese taken against their will and it is doubtful that the contemporary Tang court had information about this either.9. Although one source mentions that the commoners were under the jurisdiction of the child king (JTS 194a:5154), we can assume that he was a puppet and did not exercise real authority over these settlers, many of whom must have predated his arrival.10. JTS 194a:5159; XTS 215a:6034; ZZTJ 192:6037; Pulleyblank 1952, 323–325.Pulley-blank identifies Hu as Sogdians, but this may refer more generally to other peoples from western Asia.11. JTS 68:2507; ZGZY 8:304. ZGZY dates this to 635. This has to be a mistake. See note 10.12. Graff (2002, 49–65) attributes the fall of the Eastern Turks to the natural disasters, political divisions, and successful Tang military tactics. On the other hand, Eisenberg (1997, 320–325) emphasizes Turk political fragmentation.13. History of the Sui claims that the rebels Xue Ju, Dou Jiande, Wang Shichong, Liu Wuzhou, Liang Shidu, Li Gui, and Gao Kaidao accepted the title of qaghan (SS 84:1876). Biographies in the Tang histories mention only Liu Wuzhou and Liang Shidu having done so (JTS 55:2253, 56:2280; XTS 87:3730; ZZTJ 183:5718, 5724; Pan 1997, 168–171 n.1).14. JTS 194a:5155; XTS 215a:6029. JTS claims one thousand horses and two thousand cavalrymen, but XTS numbers them as two thousand and five hundred, respectively.15. On the Tang’s consolidation of power during this period, see Wechsler 1979, 167; Eisenberg1997, 315;and Graff2002, 35.For Turk support of the rebels Gao Kaidao, Yuan Junzhang, and Liu Heita, see JTS 194a:5155–5156; XTS 215a:6030–6031; ZZTJ 190:5948, 5951, 5968–5969.16. JTS 55:2255; XTS 92:3805; ZZTJ 190:5967–5969. The exact number of Turk garrison troops is not recorded. The sources only mention that two hundred Turks later were killed during an uprising.17. The Western Turks had similar arrangements in oases that were subordinate to them; see Skaff 1998, 38–62, and Skaff 2002.18. Variations of this speech appear in JTS 55:2255; XTS 92:3804–3805; and ZZTJ 188:5882–5883. The translation is derived from the version in JTS.19. For the many possible distortions that can be introduced into traditional Chinese biographical writing, see Twitchett 1961; Twitchett 1962; and Yang 1961.20. See Skaff 1998, ch.2, for a comparison of Tang and Western Turk administration of oases in East Turkestan.21. JTS 55:2255; XTS 92:3805. The translation is based upon the JTS version of the speech.22. Shelley Chang (1990) has demonstrated in her study of China’s Ming dynasty (1368–1644) that even in core areas elite and popular values could diverge. Most germane to this study, Ming popular culture glorified friendship (a particular type of personal relations) as a greater virtue than elite Neo-Confucian ideals of absolute loyalty to sovereign. This could point to a long-term divergence between Chinese elite and popular values.23. Tao 1988, 10–11, 35–36, 69–78; Twitchett and Tietze 1994, 110; Lamouroux 1997; Standen 1999, 71–73. Standen notes that the Tang and Tibet earlier had demarcated a linear border, but in practice neither side respected it. Thus, the Liao and Song developed East Asia’s first truly regulated border.24. Jih d literally means “striving” for faith, and has been interpreted as both a spiritual struggle for religious perfection or holy war against enemies of Islam (Tyan 1960, 2:538; Bonner 1992, 19–24; Bonner 1996, 186).25. Communication, transportation, and education have been mentioned as key factors in the spread of nationalism since the nineteenth century. Gellner (1983) stresses public education, and Anderson (1983) places greater emphasis on print capitalism and transportation. | |
Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.