IMPERIAL REPERTOIRES IN REPUBLICAN XINJIANG
IN HONING THEIR TOOLS OF GOVERNANCE IN XINJIANG, HAN officials looked to both domestic precedent and foreign models. Their source for investigating the former is readily apparent: for the first two decades after the 1911 revolution, nearly all Han officials in Xinjiang were veterans of the Qing state, claiming extensive experience in the old imperial bureaucracy. Most of them were intimately familiar with Qing repertoires of rule and had an extensive literature of former imperial precedents at their fingertips. But what about their knowledge of contemporary European and Japanese models? The journey from Urumchi to Beijing, undertaken on camel and horseback well into the 1930s, could take upward of three months to complete. Clear into the 1940s, Xinjiang was notorious for the extent of the information blockade imposed upon its inhabitants by a series of Han warlords. The telegraph system was in perennial shambles, earning the derogatory nickname of “camelgraph” from the Russians. As a result, the conventional view of Han officials in Xinjiang has been that they were of a rather parochial and reactionary mindset, scarcely interested in the outside world. Writing about Governor Yang in 1917, Xie Bin, an envoy from the central government, concluded that “his mind is too steeped in the old ways of thinking and his convictions are too deeply imprinted. He has served as an official in the northwest for too long and knows nothing of intellectual currents in the outside world. He will prove unable to row his boat with the tides of progress.”1
Xie may have been surprised to learn just how informed about the outside world both Yang and his former Qing colleagues in Urumchi actually were. In 1909, Liankui, the Manchu governor-general for both Gansu and Xinjiang and Yang’s superior at the time, was tasked with overseeing provincial elections and the formation of a parliament for Xinjiang. After noting the miniscule proportion of Han in his province and the almost total lack of educated Han gentry, Liankui proceeded to regale the central government with his extensive knowledge of ethnopolitics in other contemporary empires. “In governing their dependent territories,” he wrote, “both Eastern and Western states employ special methods. The British in India and the French in Vietnam, for example, both maintain an autocratic form of government.” After citing Herbert Spencer’s The Study of Sociology, Liankui outlined three types of government in European empires: settlement colonies subject to a monarch, dependencies with representative organs but no cabinet, and those with both representative organs and a cabinet. In those dependencies where “white people are many but natives are few,” the “degree of civilization” made it possible for “the state to grant them self-governing powers.” Thus, according to Liankui, until Xinjiang made significant strides in either developing education among the non-Han peoples or resettling droves of educated Han to the northwest, the province was not prepared to undertake elections.2
As was made evident in the discourse on Hokkaido and Taiwan, the once prevalent view of Chinese officials on the frontier as doddering relics of an intellectual and cultural backwater is almost entirely baseless. While they may have striven to limit the contact their subjects had with the outside world, the Han governor and his ranking officials in Urumchi were fully enmeshed in the intellectual and political currents of a cosmopolitan imperial elite. Wang Shunan (1851–1936), the provincial treasurer of Xinjiang during the final years of the Qing dynasty, provides us with an excellent example. Born into a literati family in Hebei, Wang worked his way up through the traditional examination system before serving as a magistrate in Sichuan for eight years. After coming to the attention of Zhang Zhidong, one of the most vigorous reform officials of the late Qing, Wang worked on various modernization projects in the Yangzi delta, coming into frequent contact with Chinese students returned from Europe. An ardent admirer of Japan, Wang was sent in 1898 to escort Western-style munitions to suppress a Muslim rebellion in Gansu, just southeast of Xinjiang. Wang would spend much of the next two decades in the far northwest, obtaining a transfer to Xinjiang in 1906.3
Just as important as his political career, however, were Wang’s scholarly labors. During his time in Gansu, Wang wrote five books concerning the history of various European countries, with the goal of identifying the source of their wealth and power. By the time of his death, Wang had authored or edited histories of Greek philosophy and of great wars in European history, a massive gazetteer of Xinjiang, the official history of the Qing, and a history of Russia under Peter the Great. Among his voluminous publications, some are notable as the first-ever treatment of their subject in China. More germane to our purposes here, most of Wang’s works on European history and culture were written during his time in Gansu and Xinjiang and published in Lanzhou, far from the traditional centers of Chinese scholarship. When the famous French sinologist Paul Pelliot passed through Urumchi in 1907, he was delighted to find so many cosmopolitan savants among Wang’s entourage. Wang, “a man of great learning” and a “very esteemed scholar,” asked Pelliot to lend him astronomical instruments for some scientific experiments. After noting a few places in need of revisions in Wang’s history of Peter the Great, Pelliot was “overwhelmed” by local officials hoping to pick his brain. One asked Pelliot to write several pages summarizing the last two centuries of developments in European philosophy, while another asked him to pen an article describing the financial conditions of loans and interest rates in Europe, with an eye toward eliminating the usurious practices of Hindu moneylenders in Kashgar.4
Wang is so important because he presided over a patron-disciple relationship with Yang Zengxin, the future Republican governor of Xinjiang (1912–28). Indeed, over a relationship that would ultimately span more than three decades, Wang secured Yang’s transfer to Xinjiang from Gansu, contributed several prefaces to Yang’s Records from the Studio of Rectification, authored the epitaph for the tombstone of Yang’s father, published a book of poetry in praise of Yang, and carried out a host of political assignments on Yang’s behalf in Beijing. In return, Yang financially supported Wang and his entire family for nearly twenty years after the latter’s departure from Xinjiang. Thus it seems safe to say that Yang, the governor of Xinjiang for nearly two decades after the 1911 revolution, was able to partake amply of the global knowledge economy of imperial governance as distilled in his patron’s writings. Nor were Wang and Yang unique among Chinese officials in Xinjiang. The large volume of travel writings left by Western archaeologists and explorers who visited Xinjiang during these decades suggests that many—though by no means all—of Yang’s subordinates were similarly eager to learn about the latest developments abroad. In 1906, when Finnish traveler and future statesman Gustaf Mannerheim passed through the southwestern oasis of Kashgar, he noted how “during our visits to the Chinese authorities they were interested in informing themselves about the political situation in Russia,” and that they “expressed the conviction that H.M’s government will not be long-lived and that their mighty neighbour is moving towards a republican form of government.”5
Their interest in Russia was not misplaced. As with so many other areas of modern Chinese history, developments in the Russian empire often constituted a preview of social and political vicissitudes that were soon to shake China. Nowhere does this realization come through more clearly than in the response of Han officials in Xinjiang to the evolving imperial repertoires deployed by their Russian counterparts. Russian officials in and around Xinjiang first showed Han officials how to turn their newfound imperial liabilities into nationalized assets. Though political elites throughout China were closely attuned to strategies of difference practiced in the British, French, and Japanese empires, those practiced in the Russian empire would ultimately determine the geopolitical and ethnocultural configurations on display in today’s People’s Republic. Due to Russia’s geographical proximity, superior military technology, transportation networks, and economic resources, Han officials posted along the northern and western borderlands ignored their neighbor at their own peril. In fact, more often than not, whenever Han rulers saw fit to import Russian political innovations into their own non-Han jurisdictions, they did so as part of a defensive strategy designed to counterbalance an offensive version of the same tactic first introduced by the Russians.
The concept of imperial repertoires is essential to understanding the specific types of institutions and strategies of difference employed in support of late imperial and early Republican rule in Xinjiang. These can be thought of not as “a bag of tricks dipped into at random nor a preset formula for rule,” but rather the evolving tools of governance that imperial elites could envision on the basis of past precedents, cultural habits, geopolitical context, and rival innovations.6 Flexible in nature, they tended to evolve from a pragmatic impulse to pursue the path of least resistance in securing the loyalties of diverse constituencies. At any given moment during the twentieth century, the changing repertoires of Chinese rule in Xinjiang are best viewed as a sort of administrative cabinet of “best practices,” an imperial portfolio of governing tactics whose contents were culled from domestic precedent and contemporary foreign models. Successful Chinese rule in Xinjiang during the twentieth century depended upon up-to-date knowledge and innovative application of portable repertoires of differentiated rule then circulating through the various empires of Eurasia.
A number of specific precedents would have been familiar to Governor Yang as he assumed his new post in 1912, less than a year after the Wuchang uprising: territorial accommodation, dependent intermediaries, supranational civic ideology, deflection of ethnic tensions, and narratives of legitimacy.
The institutionalization of difference in units of territorial administration has long been a hallmark of empire. For most of the imperial era in China, the driving force behind such demarcations was the “northern hybrid state.” These political entities combined the military advantages of an Inner Asian conquest elite with the cultural, economic, and administrative resources of the Han heartland. The largest and most successful empires in continental East Asia invariably evinced strong northern and northwestern associations.7 Whenever pastoral peoples from the “northern zone” managed to incorporate the sedentary agricultural communities of the south into their state, however, they continued to treat the Han heartland as a distinct economic, cultural, religious, and political unit. While the “inner provinces” (neisheng) were intended to provide most of the wealth, labor, administrative knowledge, and cultural resources necessary to run a massive empire, Mongolia, Tibet, Manchuria, and Xinjiang were chiefly envisioned as strategic bulwarks unique to the geopolitical concerns of an Inner Asian conquest dynasty. As such, they were not expected to finance their own administrative and military expenses, and instead drew massive subsidies of silver—known as “shared funds” (xiexiang)—from the wealthy interior. Their special status was captured in the name of the government office tasked with their administration during the Qing: the Court for Managing the External (Lifanyuan). Whether envisioned as a “dependent territory” (shudi) or an “outer dependency” (tulergi golo in Manchu), the point was that Xinjiang and other non-Han territories were different, and this difference should be recognized in the territorial institutions through which they were governed.8
In practice, this meant that a political map of Qing Xinjiang would have evinced composite layers of jurisdictional units, which claimed varying degrees of autonomy and ties with Beijing. Some of these geographical units—Ili, Tarbagatai, and Khobdo among them—were governed by Manchu and Mongol bannermen dispatched from Beijing. Others, such as the Muslim khanates of southern and eastern Xinjiang, were hereditary fiefdoms granted by the Manchu court to the descendants of the indigenous Turkic supporters of the initial Qing conquest of Xinjiang in the 1750s. Muslim princes retained control over the taxproducing resources of their districts, the most prominent of which were located in Hami, Turpan, Kucha, and Khotan. Though cultural and social links persisted among these Muslim khanates, the relationship was ultimately oriented toward Beijing, where each prince and his entourage were expected to participate in a periodic round of pilgrimage to present tribute to the emperor. The end result very much resembled a so-called hub-and-spoke patronage network, “where each spoke was attached to the center but was less directly related to the others.” Ideally, horizontal intercourse among the constituent parts—such as marriages—could be undertaken only through the center (Beijing).9
During the first decade of the republic, this patchwork legacy of territorial and administrative layering induced seemingly unending political headaches for Governor Yang Zengxin. Faced with numerous geopolitical crises in and around these semiautonomous regions, Yang found it almost impossible to force their officials to do his bidding. Making matters even more difficult for Yang was the fact that his rivals enjoyed their own lines of communication with Beijing, thereby allowing them to bypass the governor’s censors in Urumchi. It should come as no surprise, then, to learn that Yang made it a high priority to eliminate these autonomous jurisdictions. From 1915 to 1921, the governor, aided by geopolitical crises occasioned by the Russian civil war, successfully lobbied for the abolition of the Ili general, Tarbagatai councilor, and Altay minister. In fact, the present-day borders of Xinjiang are largely a result of Yang’s efforts to create an administratively homogenous provincial unit free from internal challenges to his rule.
Yet a crucial caveat is in order. Whereas Yang was only too happy to aggrandize the authority of those (mostly) Han officials sent by Beijing to fill military posts once reserved for the Inner Asian conquest elite, he did not apply the same model of aggrandizement to the indigenous non-Han nobility of the province. Partly this was for the simple reason that the Qing court had already done away with most of them. By the time Yang became governor, the Muslim princes of Turfan, Kucha, and Khotan existed in name only, having been divested of their economic and military prerogatives during the reforming zeal of the late Qing. Yet the prince of Hami, Shah Maqsut, still lorded over his khanate in both name and substance. Yang Zengxin, over a period of nearly two decades, never saw fit to continue the late Qing trend of depriving the indigenous Muslim nobility of Xinjiang of their hereditary privileges by eliminating the last Muslim prince of Hami. In fact, he often decried the integrationist thrust of late Qing reforms in Xinjiang, lamenting how Xinjiang had become a “colony” (zhimindi) of the inner provinces.
The cumulative picture to emerge from Yang’s efforts, then, is as follows. Regarding as a mistake the designation of Xinjiang as a province in 1884, Yang moved to restore a foundation of institutionalized difference in Xinjiang, which he regarded as a land distinct from the inner provinces. To do this, he needed to eliminate those semiautonomous offices traditionally filled by appointees from Beijing, for these extended into Xinjiang a volatile continuity with the centralization and integrationist efforts of the late Qing state. In their place, Yang staffed his provincial bureaucracy with veteran Qing officials of the northwest and reserved criticism for those who ventured to govern the non-Han borderlands without prior experience in the field. In effect, he was recreating the principles of territorial accommodation once prevalent in Xinjiang prior to the last decades of the Qing, complete with a new occupational caste modeled on the Inner Asian conquest elite once assigned sole responsibility for the non-Han borderlands. These men were Han officials who had spent most, if not all, of their careers in the northwest.
Without a doubt, Yang manipulated this legacy of territorial difference for his own purposes. Yet part of the reason he managed to stay in power for seventeen years is precisely because he recognized the very real conditions of difference bequeathed him by his Qing predecessors. According to Yang, Xinjiang, though still a province in name, must be treated differently from the Han heartland. Otherwise, as he was fond of telling anyone who would listen, it just might become “the next Outer Mongolia.” In 1955, the Chinese Communists, in repudiating Xinjiang’s provincial status and designating it the Uighur Autonomous Region—effectively restoring the early Qing distinction between “inner” and “outer” domains—gave formal sanction to what Yang had long acknowledged in practice.
Closely associated with the aforementioned structures of territorial accommodation, dependent intermediaries generally belonged to one of four categories: indigenous elites from a conquered society (e.g., the Muslim princes of Xinjiang); members of the ruling caste intentionally displaced from their homeland (e.g., Manchu, Mongol, and Han bannermen); learned members of a previously marginal group who improved their lot by serving the new power (e.g., Jews, Armenians, Germans, Tatars, and Jesuits); and members of a stigmatized group suddenly placed in a position of authority (e.g., slaves and eunuchs).10 What these groups had in common, however, was utter dependence on a transcendent authority, who in turn entrusted them with the most sensitive and strategic of tasks. Though certain types of dependent intermediaries were more prevalent in Han-dominated polities than in the northern hybrid states—the Ming emperors, for instance, employed at least five times as many eunuchs as did their Manchu successors—Qing emperors habitually made use of representatives from each of these four categories.
Due to the strategic nature of the posting, these dependent intermediaries played a particularly important role along the non-Han borderlands. In fact, until the late nineteenth century, the only post a successful Han graduate of the civil service examination system could hope to obtain was one located within the inner provinces. Mongols and Manchus, however, could serve in both the heartland and the non-Han borderlands. Thus, when Wang Shunan and Yang Zengxin were transferred to Xinjiang in 1906 and 1907, respectively, both were representatives of the first generation of Han officials to govern the non-Han borderlands in over a thousand years, all the way back to the Tang dynasty. Yet they quickly learned the tools of the trade. During the four decades of the republic in Xinjiang, Chinese governors would make full use of Tatar expatriates, Manchu bannermen (the Solon and Sibe), White Russian soldiers, the Muslim prince of Hami, Mongol and Kazak princes and chiefs, Turkic begs, and Hui (Chinese-speaking Muslims) soldiers.
Prior to the reassertion of central government control in Xinjiang during the 1940s, the goal of Han governors of the republican era was to redirect the loyalties of these dependent intermediaries away from their traditional orientation toward Beijing and recalibrate the relationship back toward the administration in Urumchi. Unlike the approach to rival Han generals stationed throughout the province, the goal was not to eliminate Xinjiang’s dependent intermediaries. In the final analysis, Yang and his successors, in dealing with matters of religious, cultural, economic, or political import, aimed to position themselves as chief arbiters over the affairs of their dependent intermediaries, whom they generally recognized as continuing to play a critical role in mediating Han rule among their subjects. So long as they continued to acknowledge their vertical relationship within the hub-and-spoke network of the Chinese administration in Urumchi, their livelihoods were typically safeguarded.
When Shah Maqsut, the Muslim prince of Hami, attempted to break free of such constraints, he met with stiff resistance from Governor Yang. In 1914, the prince requested permission to make his scheduled pilgrimage to Beijing via the Trans-Siberian Railway in Russia, bypassing the costly and time-consuming camel paths through inner China. Yang begged the republican administration in Beijing not to set what he considered to be a dangerous precedent. “If the prince enters Beijing through Russia and traverses thousands of miles in a foreign land,” he wrote, “he will see with his own eyes the material prosperity and superior transportation of the outside world…. It will then be very difficult for us to prevent him from becoming disaffected at heart.”11 When the central government seconded Yang’s stance, Maqsut was forced to trod the well-worn caravan routes of the inner provinces. In overruling Shah Maqsut’s request, the early republican government, not yet plagued by warlordism, was acting in a manner familiar to contemporary European empires. In 1919, when the paramount chief of Basutoland visited Britain for an audience with King George V, he was refused permission to proceed to Rome out of fear that he “might be unduly impressed by the pomp and state of reception at the Vatican, and might form the conclusion that the Pope was more important than the King.”12 Though both states had a vested interest in sponsoring initiatives that reinforced ties of politically conservative difference, such difference was beneficial to the state only if it was oriented toward its own metropole.
The Hami prince got his revenge in Beijing by circulating accusations of obstruction and sabotage against Governor Yang, who in turn directed his own brother, Yang Zengbing, to refute these accusations in the capital.13 Whatever the truth of Shah Maqsut’s grievances, however, it is important to note that Yang never saw fit to complete the course of late Qing reforms that had already divested Maqsut’s fellow Muslim princes of their khanates. Quite the contrary: no matter how much trouble Shah Maqsut created for the Han administration in Urumchi, Yang seems never seriously to have considered abolishing his khanate. And yet the troubles Maqsut brought about were indeed serious. “The prince of Hami has treated his subjects so cruelly,” Yang observed in 1915, “that in 1907 and 1912 the people twice rose against him in great rebellions.” When the British archaeologist Aurel Stein passed through Hami in 1907, he too marveled at what the local prince could get away with. Describing Maqsut as “a Muhammadan local chief whom the Chinese have found politic to keep in power,” Stein noted the ways he “squeezes his people far more than the most rapacious Amban [Qing official] would,” resulting in “a little riot some weeks before my arrival, which, owing to the Wang’s [prince’s] possessing a supply of Mauser rifles, ended quickly with a good deal of needless bloodshed.”14
Stein did not know the full story, however. A rebellion stemming from the misrule of Shah Maqsut did not constitute a piddling affair from which the Han governor in Urumchi could remain aloof. In the end, the Chinese administration in Urumchi took chief responsibility for costly military operations and the restoration of peace and order within the Hami prince’s jurisdiction. After the 1912 uprising, Yang even went so far as to invite the chief rebel, Timur, to take up a minor military post in Urumchi. When Shah Maqsut reneged on his promise to reduce the corvée labor imposed on his Turkic subjects, Timur began to plot his return to Hami. Yang, apprised of the imminent renewal of hostilities, finally made his loyalties clear. Timur was arrested and executed, while Shah Maqsut continued to lord over his khanate for another two decades, leaving this world much as he had come into it. Had Yang been looking for an excuse to rid himself of a troublesome Muslim peer, this was it.15 In 1930, when Yang’s successor, Jin Shuren, seized on Shah Maqsut’s death as a pretext to finally abolish the khanate, a massive rebellion—one of the pivotal events of chapters 2 and 3—eventually broke out. Even then, envoys sent by the new Nationalist government in Nanjing managed to track down Maqsut’s son and promise him the restoration of his father’s khanate should the Nationalists emerge triumphant in Urumchi.
To Han officials in Xinjiang, the political value of retaining what many outsiders regarded as anachronistic relics of feudal administration extended far beyond a handful of khanates. Throughout the southern oases, Yang saw fit to keep in place the system of local Turkic begs, village headmen who assumed responsibility for the affairs of their communities. Formerly a hereditary post under the Qing, the begs under Governor Yang were a diverse group: some indeed inherited their positions, while others rose to prominence through economic or religious stature. One thing is clear, however. Despite the egregious corruption carried out under the cloak of their rank, Yang had absolutely no interest in undercutting beg privileges. Xie Bin, while acknowledging that Governor Yang was a “brilliant and level-headed administrator, perhaps foremost throughout the nation,” nonetheless criticized him for “having made virtually no attempt to eliminate the manifold abuses and exploitative practices of these begs.”16
In addition to the Muslim nobility and local Turkic notables such as begs, Yang Zengxin also presided over a vast hierarchy of Mongol and Kazak princes and khans who looked to the governor to secure their interests. The case of a merchant debt owed to the khan of the Torgut Mongols in Karashahr in 1919 will illustrate this point. Having sold seven hundred head of cattle to a Russian merchant who later defaulted on his payments, the khan looked to Chinese officials to pursue his debt. Yang’s minister of foreign affairs, Zhang Shaobo, pestered the Russian consul in Urumchi, who wanted nothing to do with the case. “This was originally just a private business transaction between the khan and said Russian merchant,” replied the consul, then besieged by the Russian civil war. “Seeing as my consulate did not purchase the cattle, this matter does not concern us.” The temptation to lobby—and to be seen lobbying—on the khan’s behalf, however, was simply too strong to pass up. Zhang continued to force the issue, demanding that the beleaguered czarist consul produce the debt. Most importantly, Zhang also made sure that copies of his protests were forwarded to Karashahr, so that the Mongol khan knew exactly who was looking out for his welfare.17
Last but not least were the services offered by “people who had earlier been marginal and could see advantages in serving the victorious power.”18 In twentieth-century Xinjiang, this designation—which in other Eurasian empires often denoted Jews, Armenians, and Germans—referred to Tatars and White Russians. Members of both these groups would find themselves violently displaced by the Russian civil war, and both possessed invaluable skills and resources to offer Han rulers in Xinjiang. By far the most famous Tatar under Governor Yang—as well as under a staggering eight administrations, until his death in 1989—was a man by the name of Burhan Shahidi. Originally a merchant and fur trader, Burhan, fluent in Uighur, Chinese, and Russian, held countless job descriptions during his seven-decade career. He makes a cameo appearance in the diary of Eleanor Lattimore, wife of the noted Inner Asian historian Owen Lattimore. During her time in Turfan, Eleanor notes how she and Owen were called upon by “a Tatar gentleman whom Owen had met in Urumchi, some sort of an agent of the governor who happened to be in Turfan on business, and who took it upon himself to be our host, found us our cool room to live in and invited us to an all day picnic at Grape Valley.”19 Six years earlier, Burhan had witnessed his life and livelihood upended by the Russian civil war. Commenting on his application to adopt Chinese citizenship—surely a novel experience for a Han official at this time—Yang noted Burhan’s “considerable family assets, his respectable bearing, and the high esteem he holds among Russian merchant circles.” Bringing this Tatar under Yang’s wing, the governor wrote, could only bring “numerous advantages for us.”20
Men like Burhan were attractive to Chinese officials precisely because they had nowhere else to go. Thus, much like the eunuchs of yesteryear, they could be entrusted with the most sensitive of tasks. Burhan was not a Uighur, yet he spoke their language and could be accepted as one of them. He was not a Russian, yet he could move seamlessly through Russian social and political circles, and then share with Han officials his knowledge of developments across the border. Whereas the White Russians, bereft of an alternative sanctuary, offered modern military armaments, organization, and battlefield tactics, someone like Burhan offered invaluable intellectual and administrative services honed abroad. As a result of his unique background, Burhan was well equipped to adapt to one of the most significant developments to affect the politics of difference during the twentieth century. Prior to the Russian civil war and the unwelcome arrival of White and Red partisans across China’s northern borderlands, the foundation of ethnopolitical difference both in China and elsewhere had been built upon a conservative alliance of ethnic and religious elites. This discourse, reflecting as it did the interests of Manchu and Mongol nobles, their Han civil servants, and a largely hereditary non-Han elite, will be referred to throughout this study as “ethno-elitism.”
Yang Zengxin, though obliged to pay lip service to Confucian platitudes of patrimonial concern for “the people,” relied heavily upon the high Qing discourse of ethno-elitism. The most concrete emblem of Yang’s adherence to this platform is to be found in his Xinjiang Provincial Assembly, established as a concession to republican forms, if not substance. The job of this assembly, stocked with prominent Mongol princes and Muslim royalty, was to lend legitimacy to the words of a Han governor in Xinjiang by issuing sensitive or otherwise controversial decrees in its own name rather than that of the governor. When Jin Shuren, Yang’s successor as governor (1928–33), encountered resistance from the Nationalist government in Nanjing regarding the assumption of his predecessor’s titles, he trotted out the same Mongol and Muslim nobles to speak on his behalf. “It is said that the authorities are extremely anxious regarding the lack of a response from the central government,” noted Xu Bingxu, the Chinese codirector of the Sino-Swedish Scientific Expedition then resident in Urumchi. “So today they sent a telegram to Nanjing under the names of the Mongol and Muslim princes requesting the confirmation of [Jin’s] position.”21 The strategic calculus on display here is quite clear: if no dissent is evident among the Mongol and Muslim elite enjoying the patronage of the Han governor in Xinjiang—and the Provincial Assembly took great care to project just such an image—how could anyone from outside the province justify disrupting the status quo?
The answer lies in the realization that Yang’s Provincial Assembly, like nearly every one of his tools of governance, could undermine only competing platforms of ethno-elitist difference, such as those put forth by czarist, British, or Japanese rivals. It had no answer for an entirely new discourse of difference that was soon to sweep across the political landscape of the former Russian empire, a form that we refer to in this study as “ethnopopulism.” This new version of political legitimization shared with ethno-elitism the goal of justifying the rule of alien elites in an unfamiliar ethnocultural milieu. Where it differed was in its identification of the non-Han (in China) or non-Russian (in the Soviet Union) masses as the rhetorical pivot upon which alien rule would continue to be justified. Instead of claiming to secure and protect the interests of ethnic or religious elites—who in turn usually invoked some sort of divine or hereditary right to their position—the discourse of ethnopopulism claimed to champion the ethnopolitical enfranchisement of the indigenous common people, now portrayed as downtrodden and oppressed by corrupt and feudal ethnoreligious elites. The ethnopopulist platform, first institutionalized along the borders of Xinjiang by the Bolsheviks in the 1920s, attempted to cultivate a new narrative of political legitimacy. It did so via promises to help shepherd “backward” minority peoples toward their very own unit of territorial and political autonomy, be it a sizable republic within a federation or a smaller region within a republic.
That the ethnopopulist discourse of the Bolsheviks was related to the broader discourse of the nation-state is not in doubt. Yet there was a clear difference. Whereas the discourse of the nation-state was used by the Allied powers after World War I to dismember the German, Austro-Hungarian, and Ottoman empires along putative “national” lines, Soviet ethnopopulism was designed to “disarm” the national ideal by granting the forms of nationhood before potential rivals could do the same. In essence, the goal was to reconfigure the politics of difference away from its conservative origins and toward a progressive future, one defined in opposition to—and intended to defuse—the nationalist threat. Another way of looking at Soviet innovations to the politics of difference is to say that at the very moment when the idea of the “nation” was deployed as an offensive weapon for the first time in systematic and coercive fashion—the Allied dismantling of the Central Powers—the Bolsheviks managed to devise a “defensive” strategy designed to combat the possibility of just such a tool being used against their own multiethnic state.
In some ways, these ongoing innovations in the politics of difference during the twentieth century in Xinjiang resemble the shifting criteria for political legitimacy in the Han heartland after the Communist takeover in 1949. The chief form of political capital in China prior to 1949—the “expert” knowledge of long-standing social and cultural elites—vacillated with a new form of political legitimacy after 1949, one that valorized humble origins, practical peasant knowledge, and unadorned revolutionary enthusiasm. The “red” versus “expert” pendulum continued to swing back and forth until the Cultural Revolution, when the ideal of the combined “red expert” finally emerged triumphant.22 Similarly, in the following narrative, we find a conservative ethno-elitist form of difference, invoked since the earliest empires, forced to compete with a progressive ethnopopulist form of difference imported from the Soviet Union during the Russian civil war. Deployed against one another off and on throughout the 1930s and 1940s, the progressive platform of ethnopopulism would emerge the victor after 1949, but not before it created a new class of dependent intermediaries scarcely distinguishable in class and occupation from their conservative predecessors. It did not take long before the ethnopopulist intermediaries of the Communist era, men like Burhan Shahidi and Saypiddin Azizi, mastered the art of speaking for the people while partaking of the same elite power structure that ensured their separation from those very people.
SUPRANATIONAL CIVIC IDEOLOGY
All large and diverse states need to develop a cosmopolitan civic ideology capable of bridging those very differences that its own governing practices often serve to accentuate. In other words, the foremost goal of any metropole is to hold itself up as a sponsor of difference—whether ethno-elitist or ethnopopulist—while simultaneously orienting that difference toward its own center of political gravity. The tool used to facilitate this reorientation process is a supranational civic ideology that stresses real and imagined commonalities among the various subjects of the state, who otherwise may have little in common. Though the Qing court went to great lengths to institutionalize Manchu and Mongol privilege and to circumscribe areas of perceived Han dominance, its Manchu emperors made it clear from the start that their possession of the Heavenly Mandate—and their concomitant right to rule China—did not stem from any inherent birthright. Instead, it arose from their possession of “virtue” (de), an abstract political concept unbeholden to ethnic or national prisms. “The empire is not an individual’s private empire,” proclaimed the regent of the first emperor. “Whosoever possesses virtue, holds it.”23
The revolutionary ferment of the late Qing dynasty, often expressed in the uncompromising discourse of militant Han nationalism, did much to undermine the supranational Confucian ideology that had long undergirded empires throughout East Asian history. In fact, we might say that, after 1911, the greatest threat to a successful reorientation of non-Han loyalties to an inner Chinese metropole lay in the widespread fear that the strident rhetoric of revolutionary Han nationalists might one day be translated into equally intolerant institutions of chauvinist Han rule. During the twentieth century, such fears were embodied via the widespread perception of a normative alignment of “nation” and “state.” Both the Republic of China and the People’s Republic of China chose to harness the Sinic compound zhonghua (lit. “central florescence”) to represent the “China” of their official state names. Derived from the shifting and vague cognates of zhongguo (“the central states” or “central plains,” i.e., “China”) and huaxia (the “florescent and grand” community of civilized people), Zhonghua is the ostensible embodiment of the ideals of enlightened inclusion, not xenophobic exclusion. After all, there was no historical precedent for zhonghua’s linguistic derivatives, zhongguo and huaxia, to serve as fixed markers of rigid ethnic identities. Quite to the contrary, they served as flexible and ever-shifting points of reference for vague categories of political and cultural identification.24
The Han-led revolutionary struggle against the Manchus, however, initiated a dramatic reconceptualization of these terms that has continued for more than a century now. In attempting to bend an imprecise premodern intellectual inheritance into the rigid classificatory parameters of a modern national history, post-Qing Chinese theorists have effectively equated both zhongguo (China) and huaxia (the civilized culture sphere) with the ethnic—and virtually genetic—inheritance of a newly defined “Han” people. Prior to the twentieth century, such a strict and overt alignment of political and ethnic identity had never before been articulated for any group defined as “Han.”25 As a result, any Han official who found himself in power along the non-Han borderlands of the twentieth-century Chinese state had to deal with the common perception that he represented the very embodiment of chauvinist Han interests. One way to combat this perception was for Han officials to make a public show of protecting the livelihoods and customs of the non-Han peoples against the perceived assimilationist juggernaut of the majority Han. For Governor Yang Zengxin, cut from the ideological cloth of a loyal Qing civil servant, this was easily accomplished. Until his death in 1928, suppression of Han nationalism was the order of the day. In 1919, when a pamphlet of military songs exhorting the Han people to rise up and “wipe clean our national humiliation” was brought to Yang’s attention, he ordered its immediate confiscation and destruction. The pride its authors took in their “yellow skin” was incompatible with the universalist teachings of the Confucian canon upon which he had consistently based his own narrative of legitimacy.26
The passing of Yang and the old imperial order in Xinjiang, however, coincided with the appearance of an innovative rhetorical tactic first peddled by Chiang Kai-shek’s newly established Nationalist government in Nanjing. To one degree or another, each successive administration to occupy the Chinese metropole would attempt to defuse hostile charges of “Han chauvinism”—the kryptonite of any cosmopolitan civil discourse—by engaging in highly visible theatrics involving the public scapegoating of a Han villain associated with Xinjiang. Though such scapegoats were preferably Han men buried safely in the distant past, the most desperate administrations were willing to skewer anyone. Thus in the 1930s, the Nationalist government, shut out of Xinjiang entirely, pounced on the opportunity to arraign fleeing Han governor Jin Shuren on formal charges of exploiting and otherwise oppressing the non-Han peoples of Xinjiang. The trial, covered widely in the media and associated with expatriate Uighur intellectuals operating under Nanjing’s umbrella, gave the Nationalists a stage on which to advertise their ethnopolitical enlightenment with respect to the non-Han borderlands. Just one decade later, they would push the envelope considerably farther by condoning Uighur periodicals that took “the bloodthirsty swords of the tyrannical Chinese butchers” and criticisms of then-governor Wu Zhongxin as their topics of discussion. Much like the early Soviet state, which found it expedient from time to time to make a grand show of criticizing Russian “chauvinism,” the Chinese Communists similarly initiated periods of overt criticism directed at the majority Han. Common to each case, however, was the strategic desire to shore up the integrity of a supranational civic ideology by undercutting the novelty of rival charges of mainstream ethnic chauvinism.
DEFLECTION OF ETHNIC TENSIONS
As much as possible, it is in the interest of imperial elites to devise ways of deflecting the association of indigenous bloodshed with rulers considered to be ethnically and culturally different from their subjects. During much of their conquest of the inner provinces, for instance, the Qing prudently delegated prosecution of major battles to loyal Han generals, who shed the blood of other Han on the government’s behalf. In early republican Xinjiang, Governor Yang Zengxin made it a guiding principle of his administration that Han commoners were best kept out of Xinjiang altogether, and that those who remained in the province should not be tasked with suppression of native discontent. Because this approach differed from that endorsed by late Qing reforms, which had fostered the establishment of Han-dominated New Armies in Urumchi and Ili, Yang justified his stance by reference to other successful empires of his day. “When the British conquered India,” he wrote in 1913, “they did so by making use of native Indian soldiers. When the French took over Vietnam, they did so by making use of native Vietnamese soldiers. Under no circumstances did they make exclusive use of outside troops [kebing].”27
In Xinjiang, Yang attempted to mimic the British and French approach to their empires by organizing an army composed largely of Hui soldiers whose families and livelihoods were rooted in the province. According to Yang, not only were these Chinese-speaking Muslims “fierce by nature and excellent fighters,” but they “all had property and family to which they could return,” thus making it easy to decommission them. “Had I conscripted itinerant Han, however, it would now be very difficult to get them to disperse.”28 As a result, entrusting military bloodshed to loyal Hui, whose historical enmities with the Uighurs ran deep, carried logistical and tactical advantages for a Han governor. In addition to enlisting the Hui, Yang continued to call upon the Sibe and Solon, two military castes ensconced in the Ili valley whose members were descended from Manchu bannermen who had assisted in the initial conquest of Xinjiang in the eighteenth century. Perhaps most important were the various Mongol banners stationed in and around Karashahr, Altay, and Tarbagatai, whose fearsome cavalry was a reminder of the prominent role these Mongols once held as privileged “partners” of the Qing conquest elite. In 1931–32, the refusal of the Torgut Mongols in Karashahr to assist Governor Jin Shuren in suppressing the rebellion in Hami ultimately hastened his downfall. Thereafter, the armed mobility of the Mongols seems largely to have been supplanted by White Russian soldiers, who played key roles in the military battles of the 1930s and 1940s.
As for the Uighurs and Kazaks, who together constituted more than 80 percent of the population of Xinjiang, neither were deemed fit for formal military service. The Kazaks were occasionally tapped for light paramilitary duties, such as that witnessed by Owen Lattimore in Tacheng. He noted their obligation to “beat up any night wanderer who is not in a respectable sleigh or provided with a lantern; a duty they can perform all the better for not being able to listen to excuses.” The Uighurs, however, referred to in Chinese documents of the day as “Turban Heads” (chantou), were best kept away from arms and munitions altogether. Yang’s official explanation was that they did “not possess knowledge or experience with regard to military service.” Though this was largely true, it was mostly a by-product of the Qing’s long-standing reluctance to arm their non-Han Muslim subjects rather than due to any innate characteristics of the Uighurs as a people.29
Thus far, the evidence seems to suggest that Yang did not maintain a Han army. He did. The difference, however, lay in the unique roles assigned to his Han soldiers. On the one hand, there were several crack contingents left over from the Qing era, formerly the pride of the New Armies or reformed secret society members. Into their arms Yang entrusted the most expensive and dangerous weapons, including the first machine guns ever fired by a Chinese soldier in Xinjiang. And yet, as any list of battlefield engagements in Xinjiang during Yang’s tenure as governor will show, these elite Han units were deployed mostly against outside threats—in other words, against foreigners (i.e., Russians) and warlord armies from inner China.30 The governor sent them to fight Russians and Mongols during a joint invasion of Khobdo in 1912, and he entrusted them with the governor’s precious machine guns during the siege of Gucheng in 1921 by the White Russian general Boris Annenkov. Aside from these elite but seldom-used Han units, Yang also kept a tatterdemalion throng of Han toughs drawn from what he referred to as the opium-smoking vagrant population. Numerous foreigners commented on their disheveled and miserable appearance. By all accounts, the units to which this Han rabble were assigned appear to have functioned as little more than a holding ground for social undesirables that Yang wished to keep off the streets. They seem rarely, if ever, to have been ordered into action, and it is difficult to imagine that the native populace took them as a credible threat. More often than not, they were mocked and lampooned as dregs of the Han race whom the governor thought best to place under organized watch rather than let them roam free through the non-Han streets of Xinjiang.
During the nearly two decades when Yang Zengxin was in power, the most prudent military approach to a non-Han region cut off from metropole support was to ensure that organized attacks on the indigenous peoples of Xinjiang were not directly associated with the Han officials who had countenanced such violence in the first place. Off the battlefield, Han governors also had to consider the deleterious effects of the inevitable influx of Han migrants from the inner provinces. Manchu and Mongol officials had already found the regulation of such migrants—most of whom came illegally—to be a difficult task in the late eighteenth and early nineteenth centuries. The integrationist policies of the late Qing exacerbated this problem by promoting the “reclamation” of “wastelands” in the decades following Zuo Zongtang’s reconquest of the region. By the time Yang Zengxin became governor, such migrants had become a threat to the stability of the province. “Over the past several decades,” he lamented in 1914, “Xinjiang has become the colony [zhimindi] of the inner provinces.” The following year the governor implored Beijing to stop sponsoring such aggressive policies. “What good can come from sending more vagrants beyond the pass?” he asked. “They bring with them nothing but trouble.”31 With the collapse of central authority in China following the death of President Yuan Shikai in 1916, Yang found it much easier to insulate Xinjiang from Han migrants. Indeed, though estimates vary, it seems likely that the Han population in Xinjiang did not exceed 10 percent of the population until well into the 1950s, and may even have been as low as 5 percent at times. Governor Yang, eager to downplay the Han presence, was wont to claim that the Han comprised less than one out of every hundred residents of Xinjiang.32
After a brief attempt in the early 1930s to resettle Han refugees from neighboring Gansu ended in disaster (see chapter 2), two more decades would pass before a central government emerged strong enough to enforce its authority in Xinjiang. When it did, however, one of the first items of business for the Communists—like the Nationalists before them—was to plan for a major influx of Han migrants, who were viewed as necessary to supply manpower for ambitious agricultural and industrialization projects (the subject of chapter 5). So long as rival powers such as the Soviet Union were strong and retained economic interests in the province, however, such plans were fraught with danger. As the Nationalist government would learn a decade before the Communists did, any blueprint for Han migration to Xinjiang offered a rhetorical tool—most often in the form of labels such as “colonialism”—with which rivals of the state could fan opposition to Chinese rule. Though it would not protect them from the application of such labels following the Sino-Soviet split—the Soviets immediately dusted off rhetorical salvos regarding “Han colonialism” that had proven so useful in the 1940s—the Chinese Communists did bring with them a more sophisticated rhetorical regime designed to justify the droves of Han migrants who arrived in Xinjiang after 1949. Much like Russian migrants to the non-Russian borderlands of the Soviet Union, Han migrants to Xinjiang were consistently presented as selfless “elder brother” guides who would bring prosperity to a poverty-stricken land. Whether anyone in Xinjiang actually believed this, from the historian’s perspective, it merely constituted the latest attempt by Han rulers in Xinjiang to deflect association with ethnic tensions expected to arise as a result of policies their detractors portrayed as a repudiation of the politics of difference.
NARRATIVES OF LEGITIMACY
Han rulers in twentieth-century Xinjiang devoted inordinate time and resources to crafting personal narratives of legitimacy. Unlike a supranational civic ideology, which deals with ostensibly universal concepts of an abstract nature and may prove equally applicable to competing actors over many centuries, narratives of legitimacy were personal and specific to the person in office. Why should Yang Zengxin be allowed to govern Xinjiang, as opposed to one of his many rivals, some of whom were Muslim? What gave the Communists the right to march into Xinjiang and displace the Nationalists? Why, indeed, is one person more fit to rule over an ethnoculturally alien population than another? Because such narratives of legitimacy invariably color much of the discourse here, it is important to understand several characteristics common to each generation.
For Yang Zengxin, the first Han governor of republican Xinjiang, a narrative of legitimacy needed to accomplish two things. First, in an age saturated with the discourse of national determination, it had to explain why a Han official should be permitted to rule a non-Han land. Second, it also needed to explain why Yang was superior to other Han and Hui officials, both those who had come before him and those plotting to replace him. Yang accomplished the first goal by making a clear distinction between himself and those Han governors of the late Qing who had immediately preceded him. “When Xinjiang first became a province,” he wrote in 1915, “the early governors … all excelled at administration, and their directives demonstrated familiarity with frontier conditions. Afterward, this all changed.”33 What Yang meant was that the earliest Han governors, posted to Xinjiang immediately after its provincialization in 1884, were cut from the same cloth as the Inner Asian conquest elite. These men justified their monopoly on power over the non-Han frontiers by pointing to their intimate familiarity with the northern and western peripheries. We might say that these early Han governors were, in Yang’s view, a sort of “honorary” Manchu or Mongol, whose special qualifications made them more fit to rule Xinjiang than were Han who only evinced knowledge of the inner provinces.
Clearly, Yang intended to include himself in this category, as is evident from frequent mentions of his long experience and office in the northwest. In fact, Yang had never held a post in the inner provinces. Born and raised in the far southwestern province of Yunnan, he was posted to the northwestern province of Gansu—home to a large population of Hui—immediately upon passing the imperial examination in 1889. Once, while noting the approximately twenty years he spent as a magistrate in Muslim-dominated counties of Gansu, Yang referred to that province as his “second home.” When he traveled to Beijing in 1907 for an audience with the Manchu court, the imperial censor classified Yang as a man uniquely suited to the peripheries, calling him “one of our more capable, important border officials.”34 In managing the process of official appointments in Xinjiang, Yang demonstrated a marked preference for veterans of the Qing state who had spent their entire careers in the northwest. Failing that, he often chose to arrange for the transfer to Xinjiang of Han officials familiar to him from his time in Gansu. Such men, in increasingly short supply as his career dragged on, were often permitted to retain their posts for many years at a time.35
More than anything else, Yang’s narrative of legitimacy was founded on the notion that he alone represented a return to the politics of difference once practiced by the Mongol and Manchu rulers of Xinjiang during the high Qing era. In his view, those who eschewed the premise that Xinjiang and other borderlands were different—and thus should be governed differently—were largely responsible for the alienation of Tibet and Outer Mongolia during the last decades of the Qing. As he wrote in 1923, the Han “cannot simply blame the Tibetans for the error of their ways; it was the degeneracy of Han officials that first pushed Tibet toward autonomy. The British simply took advantage of our mistakes. Similarly, the autonomy of Outer Mongolia cannot be blamed entirely on the Outer Mongols; it was the degeneracy of Han officials that first pushed the Outer Mongols toward autonomy. The Russians simply took advantage of our mistakes.”36 Clearly, according to Yang, without someone like him in power—a man deeply versed in the conditions of the non-Han periphery—Xinjiang would go the way of Tibet or Outer Mongolia, with little hope of reclamation. So long as the politics of difference was practiced in Xinjiang by a man of the borderlands, however, Yang was confident he could “ensure that indirect rule by the Han is far superior to self-rule by the Muslims and Turbans.”37
In his telegrams, public pronouncements, and official audiences, Yang constantly advertised his unique ability to understand and interact with his Muslim subjects. The validity of his claims rested largely on the fact that he had lived and served in some of the most difficult Muslim counties of Gansu for nearly two decades prior to his arrival in Xinjiang. During that time, he had forged deep and abiding ties with prominent local Muslim leaders and their families. In 1913, an envoy from the central government in Beijing traveled to Urumchi and recorded a few choice excerpts from his long conversations with the governor: “During his tenure [in Gansu, Yang] took the daughter of a Muslim surnamed Ma as his concubine. This Ma was a man of great talent and ability, well versed in the classics and history, proficient in the art of boxing, and an expert in swordplay. As a result, Yang enjoys harmonious relations with the Muslim people.”38 In practice, the logic of cause and effect was probably reversed here: it was unlikely that taking a Muslim concubine in itself had led to “harmonious relations” with the Muslim people. Rather, harmonious relations with the Muslims had most likely led to taking the concubine. Regardless, in arranging a strategic familial union with prominent Muslim families of the northwest, Yang was acting much as one of the most famous Qing emperors before him. In the late eighteenth century, the Qianlong emperor took the sister of a Muslim Turkic prince from newly conquered Xinjiang as one of his many concubines. In both cases, the abstract symbolism and political ramifications were clear to all.39
Yang rarely missed an opportunity to flaunt his detailed knowledge of Muslim scriptures and the complex relationships among the religious schools of the Hui and Uighur peoples under his rule. Not only that, but he frequently contrasted his intimate understanding with the ignorance of his predecessors and rivals. Anthony Garnaut notes that Yang had earned a reputation as a “capable manager” of Hui affairs in Gansu, something for which few late Qing officials won distinction.40 In analyzing Yang’s telegraphic directives to local officials and religious leaders in Xinjiang—subsequently published for the benefit of political observers on the eastern seaboard—one scholar concludes that Yang was “confident” in speaking on behalf of Islamic orthodoxy and evinced an “overriding interest” in “maintaining boundaries”—between natives and foreigners, among various menhuan and mosque communities, and between religious practice and other forms of social life.41 Toward the end of his career, Yang was perceived by educated Han visitors, such as Huang Wenbi, the Chinese archaeologist, as possessing a unique and singular ability to defuse tensions among his Muslim subjects: “Last year  rumors emerged among the Hui that the Han intended to slaughter each and every one of them, and the Hui religious leaders also announced this in their mosques. Fortunately, General Yang was able to adopt measures to defuse the situation, and nothing happened.”42
It is surely not a coincidence that Yang’s personal narrative of legitimacy, as a Han governor who was both a career northwestern official and an honorary Muslim “insider,” also served him well in justifying several cases of unsavory bloodshed sanctioned during his time in office. In 1916, when a group of republican loyalists from Yang’s home province of Yunnan plotted his assassination, Yang was quick to arrest and execute them. In explaining his actions—around which much mythical embellishment would later accrue—Yang pointed to the conspirators’ “ignorance of border affairs” as their chief retroactive sin, eliding altogether the issue of opposition to the governor’s monarchist sympathies. “Had Xia and Li managed to overthrow me,” Yang explained one year later, referring to his victims, “there is no guarantee that the Mongol and Muslim tribes would have submitted [to their leadership]. It would then have been very difficult to prevent a war of the races.” Therefore, “in the interest of maintaining our borderlands,” Yang had “no choice but to execute” the conspirators.43 He would deploy a similar pretext in 1924 when he decided to send an army to Kashgar to kill Ma Fuxing, a Hui general who Yang had learned was in intimate communication with warlords in Beijing. Rather than acknowledge the bald power politics that lay beneath his actions, Yang chose instead to highlight Ma’s oppression of the native Uighur populace, an ongoing tragedy ignored by the governor for nearly a decade. “Now that Ma Fuxing and his son have been eliminated,” Yang reasoned after the affair, “the outrage of the Turbans has been appeased, and there will be no further cause for any incidents.”44
Much of Yang’s narrative was premised on simple preservation of the status quo, in opposition to rivals who he alleged would disrupt it. Absent the silver subsidies that Beijing had once directed to Xinjiang from wealthier provinces in the interior, and lacking resources with which to maintain a respectable army, Yang’s strategy was basically to circle the wagons and then warn about what might happen if they somehow became uncircled. When his successor sparked a massive civil war interpreted by many of its participants as a war of the races, however, aspiring Han rulers of Xinjiang could no longer champion the existing order, nor could they look to the discredited past for models of narrative legitimacy. As a result, conditions were ripe for a major overhaul of existing ideological paradigms. The twelve-year tenure (1933–44) of Han warlord Sheng Shicai, a man with no experience whatsoever in the non-Han borderlands, witnessed the first steps toward importing a narrative of legitimacy first designed in the Soviet Union. The “affirmative action empire,” as historian Terry Martin characterizes it, provided a ready-made model for Han rulers to justify the retention of ethnoculturally dissimilar peoples and their lands. Treated in depth in chapters 3 and 4, the guiding tenets of Soviet affirmative action obliged Han officials to take a proactive role in shaping national identities.
In brief, this new narrative of legitimacy, successively adapted by Sheng, the Nationalists, and later the Communists, cast the Han-led state as an evolutionary sponsor of “less advanced” nationalities toward their fulfillment of “mature” nationhood and some form of political autonomy. A Nationalist official from the Ministry of the Interior put it best when, in 1945, he criticized the tendency in recent decades for Han rulers to shun policies of intervention among the non-Han masses of Xinjiang. “Most countries, in their governance of minority peoples,” he wrote, “intervene in all aspects of their affairs, leaving nothing untouched.” As a result of Yang’s hands-off ethno-elitist approach, however, the non-Han masses, increasingly aware of the ways the Soviets claimed to shepherd their non-Russian masses toward socialist modernity, had begun to “make demands on us to nurture them and get involved in their affairs.” Otherwise, “they will simply look to a foreign power to take our place and nurture them in our stead, thereby bringing about progress and modernization.” The conclusion was inescapable. “We must completely overhaul our past attitude of letting [non-Han peoples] rise and fall at their own volition, and strive with all our might to nurture their development.”45
Although this may sound like a blatant repudiation of the politics of difference, it was in fact simply a new innovation within a familiar imperial model. It was certainly true that Han rulers were now intent on “intervening” in the cultural lives of their non-Han subjects. But, as will become clear in the following chapters, such intervention was still carried out in accordance with the tenets of difference. In this case, it was embodied through an ethnopopulist narrative of political legitimacy. As Thomas Mullaney has shown, this process would ultimately culminate in the Communist “fathering” of fifty-six distinct nationalities, united in recognition of the metropole that had institutionalized their nationhood. In Xinjiang, this process had begun under Han warlord Sheng Shicai two decades before the arrival of the Communists. Both Sheng and the Nationalists had taken the unprecedented step of placing non-Han figures into government positions of conspicuous—yet often hollow—authority. Thus, when the Communists decided in 1949 to leave in place the Tatar politician Burhan Shahidi as governor, they were merely building upon the ethnopolitical foundation first constructed by the Soviets and imported across the border by Sheng Shicai and the Nationalists. In fact, the latter own the distinction of appointing the first-ever non-Han governor of Xinjiang.
As is evident in the above quote from a Nationalist official in 1945, by the time the Communists took over in Xinjiang, Han narratives of political legitimacy along the borderlands had become firmly dependent upon promises to keep pace with the latest global trends in ethnocultural engineering—or, as the ministry official put it, the “nurturing” of “minority peoples.” After 1949, the Communist state was also able to inject credibility into claims that they alone were capable of bringing economic prosperity to Xinjiang. Among the more novel additions to Han narratives of legitimacy about this time, however, were Communist proclamations that they were responsible for having rectified a century of “national humiliation” (guochi). Though the Nationalists were acutely aware of their “humiliation” and frequently promised its imminent reversal, the geopolitical conditions of the day did not allow them to succeed in this endeavor.
For the Communists, however, the discourse of national humiliation—and their purported role in negating it—would become an invaluable strategic tool along the non-Han borderlands. Without fail, on nearly every occasion after 1949 in which Uighurs or Tibetans have engaged in political dissent against the Chinese state, their motives are impugned in a way that Han protests in the heartland are not. As Yang Zengxin recognized nearly a century ago, it is only in those regions of China that Han elites “would not be able to recover” even if “we had an eternity” that Beijing can expect accusations of foreign meddling to sound plausible.
⋆ ⋆ ⋆
The ethnopolitical milieu into which the first Han governor of the Republic found himself thrust after the 1911 revolution was far from auspicious. British archaeologist Aurel Stein, who began his third expedition and second circuit of the province in 1913, the year after Yang’s hasty promotion, recorded a terse summary of the conditions he faced near Kashgar: “Kichik Beg, of Kashgar origin, deplores uncertain condition since there is no Emperor. Though all harvests since 1908, with exception of one were excellent & water abundant, no new lands opened. Want of labour felt. Great drain of labour to Farghana. High prices since a year or two attributed to uncertain political outlook.”46 And yet Yang Zengxin would ultimately remain in power for another sixteen years, all the while lacking a single credible military asset. Writing just four years after Stein, the central government envoy Xie Bin was surprised to find a seemingly anachronistic Chinese administration still ruling over such a vast non-Han land. “When the race of those who wear caps and sashes,” Xie observed, referring to Han officials, “rules over a people whose language is unintelligible and with whom no communication either via speech or writing is possible, it has no choice but to use interpreters to reach the people, while relying on local chiefs to implement their policies.”47 In the best of times, Han officials in Xinjiang already faced considerable impediments to the deployment of ethnopolitical policies not detrimental to the peace and stability of the region. To achieve their goals, these officials leveraged every new idea in the transnational global economy of imperial repertoires to buttress their rule. Governor Yang, however, did not rule during the best of times. How he dealt with the nationalist threat to Chinese rule in Xinjiang during the worst of times is the subject of the next chapter.