WE END OUR STORY WITH A STATUE. A BRONZE STATUE, TO BE exact. For about a decade from the mid-1920s, it stood in a handsome two-story pavilion in Urumchi’s West Park, a monument to traditional Chinese architecture. Standing on the second floor was a metallic bust of Governor Yang Zengxin, visible to all passersby.
In 1927, Nicholas Roerich, the peripatetic Russian philosopher and painter, passed through Xinjiang and recorded a dim view of its aesthetic qualities. “The collection of funds for the erection of the monument was conducted throughout the whole district by forced subscriptions,” he observed. “And as a gift ‘from the grateful population’ appeared an ugly copper figure with gilded epaulettes and stars.” Han visitors from outside the province were hardly more charitable. In the words of Xu Bingxu, the Chinese codirector of the Sino-Swedish Scientific Expedition, “The dress and appearance of the statue do not resemble the governor, and gold foil has been plastered all over the face. What a truly odd sight. Though the governor has many merits to his name as a result of his lengthy tenure in Xinjiang, a gesture such as this, which exults in a vain celebration of his achievements, seems most unnecessary.”1
After Sheng Shicai rose to power, he removed the statue from the park and placed it in a locked storage closet in the Bureau of Transportation. Ten years later, Nationalist governor Wu Zhongxin stumbled upon it by accident. “I unlocked the door and went in to take a look. Inside I found a bronze statue of Yang Zengxin, about the height of a man. It exudes a majestic air. Gazing upon the statue, I was moved to reflect upon how Yang governed Xinjiang for seventeen years, keeping our borders peaceful and secure. His achievements cannot be denied. Therefore, I knelt and bowed three times, to show my respect.”2 The trail of documentation peters out after 1944. One suspects the statue remained in the same storage closet during the early Communist years, gathering dust much as it had during the era of Sheng Shicai. In all likelihood, it probably met with an ignominious end during a Maoist campaign: if not melted down to forge useless pig iron during the Great Leap Forward, it was probably trotted out for a public inferno during the Cultural Revolution, when so many other symbols of the imperial and republican past were destroyed.
Much like the legacy and fate of the Qing empire during the twentieth century, the bronze statue of Governor Yang Zengxin has long constituted an enigma. Historians know what distant observers have said about it—almost all of it negative—but we do not know much about its underlying substance. The reason men like Nicholas Roerich and Xu Bingxu, both outsiders to Xinjiang and its political scene, could not withhold criticism of Yang’s statue was because they both expected something else. When the former deplored its gilded epaulettes and stars and the latter fixated on the vain indulgences of an acting official, they were comparing the statue unfavorably with idealized Western and Chinese aesthetic sensibilities. For Roerich, the addition of European military insignia on Yang’s chest and shoulders sullied the architectural grandeur of the traditional Chinese pavilion in which the statue stood. For Xu, the very notion that an official should consent to the erection of a “living shrine” while occupying the office celebrated by that shrine smacked of cronyism and corruption. In traditional Chinese political culture, a renowned official may live to see the commissioning of a shrine celebrating his achievements after his departure from the jurisdiction in which it is built, or he may become a highly stylized and fabled “city-god” several centuries after his death. But he is not supposed to occupy the same office for seventeen years and then oversee the construction of his own living shrine while still holding that post.3
Yang’s statue did not conform to either man’s ideal. Instead, it was an amalgam of two political and cultural traditions fused into a Sino-Western hybrid whose true utility could be comprehended only by someone familiar with the politics of difference in Xinjiang. To put it another way, we might say that Governor Yang’s statue was the physical embodiment of the historical argument this book has set forth: that is, the legacy of the Qing empire in twentieth-century Xinjiang obliged Han rulers in Urumchi to forge a new foundation of political difference. Drawing upon domestic precedents and foreign models, this new approach transformed modern China into a community of anxious administrators forced to deal with the legacy of multiethnic states in an era of national determination and decolonization. In crossing the 1911 and 1949 divides, we have identified important continuities in Chinese rule in Xinjiang. The strategies used by Han officials in Urumchi were neither especially Manchu nor Nationalist nor Communist, but were instead techniques drawn from the imperial repertoires of finance, defense, administration, and ideology that circulated freely among China’s contemporary imperial rivals. Even when developments on the ground lagged far behind political discourse, Chinese officials still made it a point of pride to keep pace with the latest trends in ethnopolitical posturing.
The foregoing chapters have presented a narrative of modern China as an empire in transition whose political elites were obliged to meet similarly evolving threats with similarly evolving imperial repertoires. We began with Governor Yang and his selective revival of key conservative platforms from the high Qing era. Yang’s revival of the discourse and structures of ethno-elitism, in which local Turkic begs and Chinggisid Muslim and Mongol princes retained their hereditary or divine rights to rule parts of Xinjiang, was a repudiation of premature attempts by the late Qing state to integrate Xinjiang into the rest of the country. When the Western discourse of the nation-state infiltrated Xinjiang during the Russian civil war and eventually served as the pretext to establish Outer Mongolia as an independent state, Yang’s response was twofold. First, he developed a unique strain within the discourse of national humiliation to suggest that anyone who failed to treat Xinjiang differently from the rest of China would bear responsibility for its becoming “the next Outer Mongolia.” Next, he responded to the collapse and reconstitution of imperial authority throughout Eurasia by attempting to insulate Xinjiang from any and all nationalist platforms, both those from the inner provinces, which valorized the “Yellow Race,” and those from Soviet Central Asia, which promised to enfranchise the non-Han commoners of Xinjiang. He did this by refashioning the diplomatic institution of the Western consulate in his own image, directing his hard-won consulates in Soviet Central Asia to ensure that anyone inspired by Soviet affirmative action be barred reentry to the province.
When the continued inability of inner Chinese warlords to reunite the country led to a geopolitical crisis in Xinjiang—the assassination of Governor Yang in 1928 by agents of another warlord—Yang’s successor, Jin Shuren, embarked on a wholesale repudiation of the politics of difference. His centralization and modernization initiatives very much resembled the late Qing drive to integrate the non-Han borderlands, a prolonged campaign again carried out in response to a geopolitical crisis. In Jin’s case, however, the main difference was his entire lack of financial, political, or military support from the Chinese metropole. The late Qing state, however weak it may have seemed in the face of foreign aggression, had once supplied such support to its borderland officials, albeit in ever decreasing amounts. In quick succession, Jin managed to alienate the Incarnate Lama of Karashahr, the Soviets, and the Turkic peoples of Hami, both noble and commoner. When former retainers of Shah Maqsut rallied Turkic peasants against Jin’s administration, they did so by invoking the idea of a “Chanto”—that is, Uighur—nation-state, borrowing this name from Chinese descriptions of the turbans some men wore on their heads. Soon the entire province was at war, with a bewildering array of national platforms on the lips of every belligerent.
In 1934, a Han warlord by the name of Sheng Shicai emerged triumphant from the wreckage of Yang and Jin. To gain the support of his non-Han subjects, many of whom had shed their blood to prevent the rise of another Han like him, Sheng had to meet demands within his province for demonstrations of support for national determination. In response, Sheng adapted the Soviet discourse and institutional structures of ethnopopulism, in which progressive and ostensibly elected representatives of the masses are placed into positions of government and expected to mobilize their constituents toward a more equitable and prosperous future. Chinese Comintern agent Yu Xiusong described Sheng as having learned from the “only country in the entire world to have completely and correctly solved its nationality problems.” From a somewhat broader historical perspective, Sheng was simply continuing under newly ascendant Han stewardship the millennia-long tradition of ethnopolitical engineering and posturing typical of China’s many non-Han ruling elites in dynasties past.4 Sheng’s gambit worked, in spite of—or perhaps because of—his simultaneous importation of Stalinist police state tactics. These were eventually deployed to purge his administration of nearly all the non-Han peoples he had only so recently elevated into prominent positions of authority. Ultimately, however, Sheng’s “purchase” of Soviet affirmative action blueprints saddled him with massive debts that neither he nor his Chinese Communist adviser Mao Zemin could ever fully repay. After Sheng took advantage of the siege of Stalingrad to shirk his obligations to the Soviets and return to the Nationalist fold, Moscow decided to reclaim its debts by directing an ethnopopulist Uighur and Kazak “national” insurgency against the new Chinese administration in the province. Not surprisingly, this insurgency, later glossed as the East Turkestan Republic, took as its strategic targets only those three districts in which the Soviets had once maintained heavy interests in industrial minerals (Altay), pastoral stock (Tacheng), and oil (Ili).
The Nationalist response was to raise the stakes of ethnopolitical patronage in wartime and postwar Xinjiang. Those within the party who admired an ethnopopulist approach—a camp that included Chiang Kai-shek—decided to meet Soviet sponsorship of Uighurs and Kazaks by elevating their own Uighur and Kazak figures to positions of visible authority in Xinjiang. Combined with the fruits of those who championed an ethno-elitist approach—chief among which was Governor Wu Zhongxin (1944–46)—the Nationalist cabinet of non-Han personages in Xinjiang during the 1940s effectively negated the novelty of Soviet-sponsored figures such as Ali Han Tore and Ahmetjan Qasimi. Never again would so much political agency be granted to the prince of the Kazaks (Ailin), the khan-wang of the Torgut Mongols (Manchuqjab), a Kazak batur (Osman), or ethnopopulists such as Masud Sabri, Mohammed Emin Bugra, Isa Yusuf Alptekin, and Burhan Shahidi. And it goes without saying that never again would the Chinese central government suffer Uighur officials in its political wing to publish journals that criticized the “tyrannical Chinese butchers” or advertised Xinjiang as the home of the “Turkestan nation,” as General Zhang Zhizhong had once allowed. The reason for these bold Nationalist innovations was simple: in the postwar atmosphere of increasing demands for decolonization, the Nationalists had to prove their commitment to the emerging global discourse of “development” and “nurturing” of subaltern peoples the world over or risk tarnishing their narrative of ethnopolitical legitimacy in Xinjiang.
After 1949, it fell to the Chinese Communists to deal with this legacy of ethnopolitical “inflation” in Xinjiang. They did so by introducing one final adaptation to the Soviet affirmative action model: ethnocultural autonomy would be territorialized (unlike Sheng’s Austro-Marxist interpretation), but it would not be institutionalized at the level of a republic, complete with the right of secession. The realpolitik calculus behind this innovation is evident in a speech given by Wang Enmao, party secretary of Xinjiang during the 1950s, to newly arrived Han cadres. To judge from the content of his speech, these men were far more familiar with Soviet ethnopopulist discourse—and its institutional embodiments just across the border—than with Chinese Communist adaptations. Noting widespread unease among Han cadres over the stated goal of making Uighurs the “masters of their own house” (dangjia zuozhu), Wang put his colleagues at ease: “The [Han] mistakenly believe that after the implementation of ethnic regional autonomy, they can go back home to the inner provinces. Some think the implementation of ethnic regional autonomy means that the responsibilities of ethnic cadres will increase, while those of Han cadres will decrease. What none of them realizes is that even after ethnic regional autonomy is put into practice, Han cadres still will not be allowed to go home. Our job of assisting the various nationalities of Xinjiang is a long-term mission.”5
Elsewhere, Zhou Enlai justified the unprecedented waves of Han migration in the same language of “development” that the Nationalists before him had used. As for the Communist obligation to “nurture” the non-Han peoples of Xinjiang, Zhou could point to the institutionalization of ethnocultural and political privileges for fifty-five official non-Han nationalities, about half of which had only been “identified” by Communist ethnographers during the first five years of the new state.6
As evidenced by ubiquitous reports of ethnic tension and conflict on the factory floors of Urumchi, the Chinese Communists failed miserably to deliver on the promises of ethnopopulist affirmative action. At best, we might say that they delivered an emasculated version of ethno-elitism, one that masqueraded as ethnopopulism. That is to say, non-Han elites were still separated from “their” people and elevated into privileged positions of authority, but now without even the ability to wield that authority. All that mattered was that they stand in as a symbolic representation of the will of “their” people. Initially, however, the Communists had far less reason to worry about these ironic inversions than did the Nationalists. Unlike the Nationalists, they had both time and geopolitical circumstance on their side. They even had the assistance of a second Chinese metropole—the Nationalists in Taiwan—who assisted the mainland government in undermining any narrative of political legitimacy for Xinjiang that did not originate in Beijing or Taipei, both of which now evinced a more or less harmonious pursuit of common geopolitical and ethnopolitical goals. The end result was the creation of a “national empire” very similar to the state established by the Soviets just across the Eurasian border. Though no political elite or self-respecting Chinese nationalist today would employ such a term, marred as it is by the revolutionary discourse of an authoritarian—and most deplorable—form of power, this is precisely what has arisen from the ashes of the old monarchical empires: a vast state whose territorial and human heterogeneity must be nationalized to conform to new narratives of post-revolutionary political legitimacy.
The idea that an imperial entity could be defined, not undermined, by nationalism would have been familiar to the intellectuals of the late Qing era with which we began our study. In 1906, Ye Shanrong, editor of a family genealogy, envisioned the creation of a state capable of meeting the “tides of nationalism” (minzu chaoliu) that defined his age. Believing that his generation had already begun this process, Ye called his era “an age in which a national empire [minzu diguo] is being created.”7 It is not entirely clear what Ye meant by a “national empire,” a novel term in his day. It combines the word for “nationality” (minzu) with the characters for a “state” (guo) presided over by an “emperor” (di). In reference to a type of state, Ye was likely simply describing the political status quo of his day, in which a Manchu emperor sat on the throne. But his use of “nationality” (minzu) is more specific: it is the idea that people should be organized on the principle of the “nation-state” (guojia zhuyi de minzu) as a means to wealth and power in the twentieth century. For someone who did not exercise actual political control on the ground, this is probably about as far as the concept was liable to be explicated at the time. But for this study, concerned as it is with Han administrators responsible for preserving the heterogeneous human and geographical landscape of the former Qing empire, Ye’s formulation provides a striking premonition of just how the ideal of national determination would ultimately color the geopolitical calculus of Han administrators after the 1911 revolution.
The lands and peoples that compose the entity known as “China” are not today, nor have they ever been, a “nation-state.” But they do constitute a state of many nations, each of which, including the Han, bears the imprint of deliberate ethnic manufacture by the state. If we follow Frederick Cooper in ascribing to such polities the designation empire-state—in lieu of nation-state—then we have already approximated what some Soviet scholars refer to as Moscow’s “empire of nations.”8 But if the Soviet Union was an “empire of nations,” composed of potentially detachable parts, then it makes sense to call the modern Chinese state a “national empire,” defined by—but not divisible into—its fifty-six national components. Though anthropologists and historians of China have long recognized the hybrid nature of the modern Chinese state, few have attempted to trace in precise chronological detail the temporal and methodological contours of the evolution of this state throughout the late Qing, republican, and early Communist eras.9 By telling just such a narrative here, we can attempt to situate the ethnopolitical history of China both within long-standing East Asian mainland traditions and within Eurasian and Euro-American innovations of much more recent vintage.
As a type of state, empires did not all disappear during the twentieth century. Instead, as we have seen in Xinjiang, some of them simply redefined their institutional markers of difference from an ethno-elitist to an ethnopopulist model. In other words, they got a nationalist makeover. Before taking that fateful step, however, each empire made a point of canvassing the imperial repertoires of every one of its rivals, selecting for modified deployment those tactics viewed as most suitable to the unique conditions of its own imperial landscape. We might even go so far as to say that the hallmark of a successful empire during the twentieth century was the ability to deploy sufficient resources—and imperial repertoires innovative enough—to maintain a monopoly on the definition and sponsorship of categories of human difference, be they in ethno-elitist or ethnopopulist guise. A state that could continue to control the recognition and definition of its many peoples in nationalized forms was likely to subsist as some sort of a “national empire” or “empire of nations.” One that could not prevent its peoples from being recognized and defined in national terms by rival metropoles was likely to become a collection of putative “nation-states,” however imperfectly and artificially defined.
With the waning of Nationalist influence in world politics, the Chinese Communists have faced a host of new challenges in Xinjiang. After all, for nearly four decades, the Nationalists had managed from Taiwan to deflect foreign criticism of Chinese rule in Xinjiang by depriving dissenting Uighur ethnopopulists in Turkey of the support of “free China.” Beijing took full advantage of this extended reprieve, ensuring the complete integration of Xinjiang into China before Taipei lost what little remained of its political clout in the United States and Europe. Ever since the reform era, however, Chinese Communists have come face-to-face with a delayed reckoning on the failure of their ethnopopulist platform in Xinjiang during their first decade of rule. They have met this unwelcome increase in foreign scrutiny by once again adapting techniques of rule from other multiethnic states considered more “advanced” or “developed” than China, one of the few perks of a late-modernizing state. Hence, Beijing has learned from the United States the utility of discrediting non-Han dissent by suggesting transnational links to known terrorist groups, while learning from France how to build upon the precedent of banning Muslim veils. As with Nationalist and Qing officials before them, the Communists have justified such measures by pointing to similar conditions, experiences, and responses within those very multiethnic states that deign to criticize Beijing. Any further adaptations to the foreign model are then conveniently justified through the discourse of special “Chinese characteristics.”
Despite Beijing’s opportunistic use of such tactics, it is likely that most future challenges to Chinese rule in Xinjiang will be forced to make creative use not of national discourses and structures but rather of those that are transnational in design and scope. The reason is simple. The twenty-first century was the golden age of nationalist discourses. As such, the conservative ethno-elitist empires in existence at the beginning of the century have all been transformed by the use and abuse of the geopolitical tool of the nation-state. Whether they became national empires, disbanded empires, or dismembered empires, each state made its own contribution toward raising the stakes in a global ethnopolitical arms race. There are now few discourses derivative of the ideal of the nation-state for which a rhetorical and institutional antidote has not already been devised. It is thus difficult to imagine any successful challenge to Chinese rule along the non-Han borderlands based upon an ethnopopulist discourse, or indeed any discourse related to national determination. If political dissidents want to challenge Chinese control in Xinjiang, they will need to come up with an ideology for which Han officials do not already have an answer bred from a century of intimate experience on the Chinese borderlands. In practice, this may augur a return to transnational appeals of religion or hereditary descent characteristic of the premodern era, but this time shorn of ethno-elitist pretensions and discourse.
Regardless of what happens in the future, it is clear what happened in the twentieth century. Indeed, had Chinese administrators proven unable to keep pace with the evolution of imperial repertoires after 1911, it is likely that every Han nationalist’s worst nightmare would have come true. That Xinjiang did not become “the next Outer Mongolia” testifies to the pragmatic recognition of these administrators that China had not simply evolved—to paraphrase the words of intellectual historian Joseph Levenson—from a universal civilization to a nation in the world. If that had truly been the case, Han officials would not have been overly concerned with any part of the former empire outside of the inner provinces. On the contrary, even though most of them recognized that China was no longer the great “universal civilization” of yesteryear, they were far more bullish about what it had become than they are typically given credit for: an empire among empires, each of which was compelled to learn how to nationalize its empire before rival empires nationalized—and reoriented—it for them.