Are Ethnic Tensions on the Rise in China?

A ChinaFile Conversation

On December 31, President Xi Jinping appeared on CCTV and extended his “New Year’s wishes to Chinese of all ethnic groups.” On January 15, Beijing officials detained Ilham Tohti, a leading Uighur economist and subsequently accused him of “separtist offenses”; a fresh report shows arrests of Uighurs for “endangering state security” in Xinjiang rose sharply last year; and the number of Tibetans who have taken their own lives in public protest against Chinese rule has recently surpassed 120 since February 2009.


Is ethnic tension in China on the rise? The simple answer is that it depends. Yes, among the Tibetans and Uighurs, during the past few years we have observed several episodes of intense inter-ethnic communal violence. From the Lhasa riot in 2008 to the many cases of self-immolations among Tibetans, it does seem many Tibetans hold strong grudges towards the Chinese government and Han and also Hui Chinese presence in Tibetan areas. Likewise, in Xinjiang, from 2009 onwards we have witnessed several episodes of violent attacks by Uighur militants on the Chinese state apparatus. That said, among most other ethnic minority groups in China, we do not really observe such patterns of political tension. Even among the Mongols in Inner Mongolia, despite the existence of certain economic and political grievances, there have been no similar types of violence or pro-independence movements. Keep in mind that the Tibetans and Uighurs are only two of fifty-five ethnic minority groups in China. Even though they are internationally renowned, they are nonetheless not the most numerous. That means that there is a “silent majority” of ethnic minorities in China who do not employ confrontational strategies towards the Chinese state. In fact, one can argue that assimilation/sinification is perhaps the dominant trend throughout the nation.

Then again, how shall we understand whether there are rising ethnic tensions among the Tibetans and Uighurs? The term “rising” implies, chronologically, that tensions are higher now than in previous decades. Then perhaps the answer to the question “Are Ethnic Tensions on the Rise in China?” should be “no.” Obviously, the Tibetans had several major armed rebellions during the mid 1950s in Kham and later on in Lhasa, in 1959, which led to the exile of the Dalai Lama. Similarly, during the late 1980s ethnic tensions in Lhasa were such that major protests led to the imposition of martial law in 1989. If we compare today with those earlier periods, perhaps ethnic tension is not on the rise; it just stays at the same level at the most. The same can be said about Xinjiang, too. The 1990s were particularly volatile, with many clashes, bombings, and assassinations. Thus, quantitatively trying to measure any increase or decrease would have to take into consideration those past patterns of inter-ethnic relations in Tibet and Xinjiang.

At least when it comes to the Uighur, the situation is certainly getting worse. A Uighur friend messaged my partner last week, and she wished him a “happy new year.” “Not my New Year, thanks,” he wrote sourly. Like an increasing number of Uighur even in Beijing, he maintains a puritan distance from Chinese culture; he refuses to eat Chinese food, to speak Chinese unless absolutely necessary, or to have Chinese friends.

Dr. Han says that the Uighur and the Tibetans “employ” confrontational strategies with the state, and implicitly contrasts them with the good minorities who understand that they should just roll over and become Chinese. But to be Uighur, at least, is to be constantly confronted by the state. It’s to be told that your traditional meetings are forbidden, that your history is not your own, that your intellectuals are terrorist sympathizers, to have the veil forcibly ripped from your face, and to be made to keep the time of a capital 2,000 miles away. It intrudes even into small things; Uighur names are too long to fit onto Chinese ID cards, for instance, and so end up being written in butchered Sinified versions.

The violence we’re seeing is not driven by purely inter-communal tensions. Nor was it some kind of choice for the Tibetans and Uighur. It’s not ethnic violence, but anti-colonial violence. Unlike the other “minority peoples,” who have a long history within the Chinese state (whether as ordinary citizens, like the Hui, or as marginalized peoples, like the Miao), the Tibetans and Uighur have their own long-standing sense of national identity. (So do the Mongolians, but they also have a nation of their own—which China did all it could to prevent happening, but which bleeds off most of the anger.) For the Tibetans and Uighurs, the Chinese are invaders and colonizers, and always will be.

The only way to begin to resolve this would be for the Chinese state to acknowledge its own imperialism, to cease the colonial project and the attempt to obliterate independent histories, and to allow genuine cultural and political autonomy. But those ideas are utterly anathema, both institutionally and ideologically, in today’s China.

Ordinary Chinese conceive of themselves only as victims of imperialism, never as its perpetuators, and they deeply resent the idea that they might be the villains of this story. After all, aren’t they bringing civilization and modernity? (Like the British did to India, or the Japanese to Korea …) It’s not an impossible moral breakthrough for people to make—I remember one teenage girl coming back from a Tibetan visit and remarking “No wonder they hate us, there are police everywhere”—but to do so means going against everything they’ve been told since they were children.

Xinjiang has obviously only got worse over the last year. But one crucial threshold hasn’t been crossed: the violence has targeted symbols of the state like police stations, not the Han population as a whole. We saw what happened in 2009 when tensions exploded into inter-communal killings; I fear the outcome if that became the norm, rather than a horrific exception.

Asking about ethnicity and tension in China is one of those questions that mirrors the problems it tries to describe: it uses terms and categories that blur what those problems are and replicate the stresses that underlie them. The word “ethnic” is used these days in discussions about China because of an internal order by the Chinese authorities twenty years ago that from then on English-language texts in China must always use the word “ethnicity” instead of the only one used till then, the old Leninist term “nationality.” Chinese officials wanted, understandably, to get away from a word implying that these groups might be entitled to political demands or claims over territory. It hoped that the new term would suggest only cultural and economic differences. Many foreign writers and journalists now follow the same practice, perhaps without remembering the reasons why it was introduced and the issues that it veiled.

Both terms reflect the same convention: within China, these groups, which for forty years were “nationalities” but are now “ethnicities,” are treated by policy-makers as a single type of entity because they differ in some way from the majority of Chinese. There are, however, hundreds of other groups in China that could be classified in the same way, but for administrative reasons aren’t. Of those that are, some are very similar in culture to the ethnic Chinese, while others have histories and languages that are entirely different; some ruled themselves for centuries, while others have always lived under Chinese governments; while many have homelands in remote and little-known areas of China, the traditional areas of just three groups cover fifty-five percent of China’s landmass and are of immense strategic significance. Those three are the Uighurs, Tibetans and Mongolians. It makes sense for there to be differences in their relations with the larger state.

Put simply, some of these groups are subject to very restrictive policies while others are not: the ethnic approach has led to a policy quagmire. For the last three decades China tried to handle all these groups by promoting rapid development and, more recently, by boosting incomes. It also encouraged a limited amount of cultural expression. But when some groups that had long been promised more than economic progress restated demands related to their different histories and traditions, they were accused of “splittism, ” the Chinese term for separatism and the code word for a crackdown. The longer the distinctive history of that group, the more aggressive the state’s response. This in turn has led to violent protests, and the cycle of unrest has continued.

There are other factors, such as international involvement and growing nationalist emotion on all sides. But the most pressing question is whether Beijing officials can craft policies that respond to the different needs and histories of different constituencies. Treating them as a single block that needs only economic goods has so far not worked.

There is unfortunately little doubt that ethnic tensions have been on the rise in recent years in China’s two largest nominally autonomous ethnic regions, Tibet and Xinjiang.

Tibetan areas witnessed in 2008 the largest popular uprising in several decades. The protests spontaneously spread from Lhasa—where the still unexplained disappearance of security forces from the downtown area on March 14 led to multiple incidents of arson and violence—to the rest of the plateau, in particular the Eastern part, generally a much less contentious than the Tibet Autonomous Region (TAR). The harsh repression that followed the protests, documented in a July 2010 Human Rights Watch report triggered an historically unprecedented wave of self-immolations, with over 120 Tibetans setting themselves on fire in protest at Chinese policies since February 2009. Today, the security presence remains at a much higher level than before 2008, and the State pursues very ambitious social-control policies in the TAR.

Xinjiang experienced its worst episode of inter-ethnic violence in decades in July 2009, when the suppression of a peaceful protest by Uyghur students turned into a night of murderous rioting in which, officially, 197 people, 134 of them Han Chinese, died in the violence, and some 1,600 were injured. The real number of casualties might have been much higher, as high as 800. There, too, the repression was brutal and indiscriminate, and marked by an unusually high rate of disappearances at the hand of the security forces, as documented by Human Rights Watch in an October 2009 report.

Ethnic polarization in Xinjiang has since been further fueled by policies designed to accelerate the assimilation of the region through massive investments and state-led modernization, accompanied by ever-growing restrictions on religion and cultural expression imposed in the name of the anti-separatism struggle. The year 2013 witnessed the highest incidence of political violence in Xinjiang since the Urumqi riots of 2009, with over 120 casualties. Recorded incidents included the suppression of local protests by security forces using live ammunition, indiscriminate attacks by homegrown Uyghur militants, and murky incidents in which people labeled as “terrorists” are killed in episodes that suggest at a minimum widely disproportionate use of force by security forces.

One should of course bear in mind the distinction between what constitutes tensions between specific ethnic groups and the Chinese state on the one hand, and inter-ethnic tensions on the other hand. Inter-ethnic polarization and antagonism of Han Chinese seem to be more acute in Xinjiang and, to a lesser extent, in Inner-Mongolia, than in Tibetan areas. But in both cases mutual stereotyping is only the by-product of what remains the main driver of ethnic relations today: assimilative State policies.

Conversely, there is little prospect of ethnic tensions decreasing in the near future, as the State appears more committed than ever to the accelerated “domestication” of Tibet and Xinjiang (no other ethnic group in China comes even close to these two regions in terms of territorial significance to the Chinese state.)

The key problem is that these domestication efforts are predicated on two fundamentally mistaken assumptions. The first one is that substantially raising living standards among ethnic communities will extinguish potential ethno-nationalist aspirations. Examples around the world show that it does not, all the more if target groups have no say in the design and implementation of these policies and continue to face everyday discrimination and no meaningful political representation.

The second mistaken assumption is that “separatism” is a relevant concept to reading the political situation in Tibet and Xinjiang. The truth is, neither region stands a remote chance of actually separating from China today. So it is difficult to comprehend why “the fight against separatism” should continue to dominate the political order of these regions. The State would be better off casting off what is in essence a phantom-menace—one that fuels ethnic polarization, prevents the accurate diagnosis of policy problems, and gives primacy to a security apparatus responsible for widespread human rights violations, including the endemic use of torture.

There are vigorous debates in China policy-circles about the need to reform ethnic policies—and not only in Tibet and Xinjiang. (Professor James Leibold has recently written an excellent overview of these debates in an East-West Center monograph). Unfortunately, the debates tend to perpetuate the practice of not including the people at stake, whether Uyhgur, Tibetan or others.

To find an illustration of Beijing’s continuing hyper-politicization of the ethnic issue under the new Xi Jinping leadership, we need to look no further than the fate of the Uyghur economist Ilham Tohti, arrested last month and branded a state security offender. His crime was simply to have been suggesting that the root of discontent in Xinjiang where to be found elsewhere than in “separatism.”

Although one can indeed note a rise in the number of incidents involving Uyghurs and Tibetans since 2008 over the decade before that, the answer requires a broader context and depends on what you mean by “ethnic” and “tension.”

As Enze Han points out, not all administratively-designated Chinese minzu (the term once translated “nationality” in the Soviet sense and now translated as “ethnic group”) are restive. Uyghurs and Tibetans are, certainly, but by far the most numerous and restive minzu are those Chinese who belong to the Han majority. Annual official P.R.C.- and Chinese academic estimates show that the number of “mass incidents” has mounted into the tens of thousands, sometimes surpassing one hundred thousand. These protests, usually inspired by land-grabs, egregious corruption, labor disputes, illegal taxation, or environmental damage, take the form, on the one hand, of SMS and Internet chatter or marches and sit-ins, to riots, suicides, bombings, physical attacks and the expulsion of officials from entire towns, on the other. It would be interesting to compare per-capita Han unrest with that of Uyghurs and Tibetans. Given that we know of only a few dozen incidents among Tibetans and Uyghurs over the past few years (some large and violent), I wouldn't be surprised to find that however you quantify it, non-violent or violent, incident for incident, perpetrator for perpetrator, casualty for casualty, there has been as much or more “tension” among Han than among Uyghurs or Tibetans in recent years.

Indeed, it is far more dangerous for Tibetans or Uyghurs to raise their voices than for Han.

There is now a well-recognized script for episodes of Han political protest: First, cause a “mass incident” to bring local grievances to the attention of the Party center in Beijing. As long as protestors stick to local issues and do not attempt to build solidarity among similarly aggrieved workers, farmers, pensioners or pollution-victims nationwide, often they can get away with surprisingly little or even no sanction whatsoever. Meanwhile, if the incident gains sufficient attention, the central Party rides in to prosecute miscreant businessmen and cashier corrupt officials. The underlying systemic problems remain unaddressed, but local tension is relieved somewhat and the Party looks good. By now, this escape valve is the de facto way of dealing with corruption, land-grabs, pension-theft, pollution and the like in the absence of a free press, free courts or other democratic checks on officials and the rich elites with whom they collude.

But we use different terminology and lenses for incidents involving Tibetans and Uyghurs than for incidents involving Han Chinese alone. In the official narrative of the PRC—and, generally, in the language used by foreign observers as well— Uyghur and Tibetan political incidents are considered almost by definition to be ethno-nationalist, “splitist,” or “separatist,” if not “terrorist.” This divergent understanding means that Uyghurs and Tibetans cannot use the script, cannot access the now common escape valve when subjected to local injustices. Their communications are censored more closely; police are immediately sent to violently repress the few peaceful demonstrations that do occur. The suppression of a demonstration by Uyghur students in 2009, sparked the Urumqi riots. Loyal, moderate, but critical voices, such as that of Uyghur economist Ilham Tohti, are more quickly and thoroughly silenced.

Such across-the-board labeling of incidents involving Uyghurs and Tibetans as “ethnic,” “separatist,” “communal” and so forth can become a self-fulfilling prophecy. By signaling to Han and other groups that there is a tolerated way to voice concerns, demonstrate, and otherwise act out their grievances, while brutally repressing similar dissent when it arises from Uyghurs and Tibetans, the Chinese party-state is reinforcing those very ethnic lines of division they claim they want to dissolve.

Last week I came across two separate examples of “jihadi propaganda” circulating across Uyghur media platforms. The first was a video posted on Facebook. The second was an audio file posted on WeChat, the mobile app that has recently become the communication tool of choice for young Uyghurs in Xinjiang. I use scare quotes for “jihadi propaganda” because up till now this term, commonly used in translations of Chinese media reports on violence in Xinjiang, has provoked in me a highly cynical reaction.

Like many of my colleagues in the West I have for several years argued that Chinese media reports of incidents of “religious extremism,” “terrorism” and links to the shadowy “East Turkestan Islamic Movement” mask a reality of local violence sparked by local grievances. As Jim Millward points out, similar things happen all over China but they are not labelled as terrorism. Too often over the past year, reports from the Uyghur villages caught up in these incidents have detailed police, engaged in a new crackdown against extremism, entering the homes of religious families and attempting to remove women’s veils, upon which the men bring out their knives to protect their family honor.

One of the eminently moderate and sensible things that Ilham Tohti dared to say before his arrest, in an interview with The Financial Times, was that the suppression of everyday religious practice in Xinjiang was liable to provoke further discontent with the state and encourage stronger religious feeling.

Robbie Barnett rightly cautions us against careless use of terms and urges us to think about the politics underlying the state dropping the term “nation” in favour of “ethnicity”. For many Uyghurs I knew in the 1990s it was indeed all about the Uyghur nation, but today the people I speak with are much more interested in being “proper” Muslims. How and why has that come about?

There’s no denying the upsurge in new kinds of religious belief and practice that has swept Uyghur communities at home and in the diaspora, most markedly I would argue since the 2009 riots. They’ve brought about changes most visibly in women’s dress codes but they’re also marked by the disappearance of alcohol from many Uyghur restaurants, the call to prayer beeping out from mobile phones, a new enthusiasm for the Ramadan fast.

I’ve recently been involved with a research project on Uyghur Islamic media, and the vast majority of the items I’ve encountered are quite innocuous. I’ve surveyed endless soft pop productions of religious nasheed songs from Indonesia accompanied by white doves and flowers. There’s also a definite presence of what we might term Islamist media, that urges women to be modest and men to renounce alcohol and gambling, using often quite gruesome images of death and decay to inculcate fear of God’s judgement.

But the two items I viewed last week took things a step further. The video showed a young Uyghur man writing to his mother to explain why he was joining the rebel fighters in the mountains, which were portrayed as places of freedom, joy and comradeship. For the first time that I’ve seen, the video explicitly drew on images of the 2009 violence and Chinese police attacking Uyghurs to explain what impelled him to go there: to seek revenge for this bloodshed. What interested me about this video was that it was not made to provoke anger and hatred but pathos. He was writing to his mother for goodness sake! I dwell on this in some detail because I think it’s important to think about how these media productions—these items of “jihadi propaganda”—make people feel.

A group of eleven Uyghur young men did find their way up a mountain in January this year. What they were doing there is unknown. They were shot dead by a special Kyrgyz border guard unit because they “appeared to belong to an organization of Uyghur separatists”. Reports that the eleven men were unarmed have cast doubt on this interpretation of events, and I suspect that we are unlikely to learn the truth of the matter.

To argue about the situation in Xinjiang in terms of “are state actions provoking religious extremism or are state actions justified in the face of rising extremism” is to argue about the chicken and the egg. In either case it seems that Uyghurs, especially Uyghur young men, are being forced by both sides—state and “jihadi” propaganda—into subjective positions which impel them into a cycle of violence, sometimes as perpetrators but more often as victims.

If we return to the original question: “Are ethnic tensions on the rise in China?” I tend to agree with Enze Han’s assessment. In short, we have no way of answering this question with any level of certainty. The sensitivity of ethnic issues in China and the closed nature of society in the P.R.C. mean we currently lack both the methods and the empirical data to systematically determine any trend lines. If the C.P.C. and its security apparatuses maintain statistical data on the quantity and frequency of “ethnic” incidents—disagreements, melees, attacks, riots and other forms of violence—this information is not publicly available, nor is it likely to be under Communist rule.

But if this data were available, I suspect—as James Millward astutely notes—we would find far more “ethnic” tension/unrest among the Han majority (there are daily tiffs, even violence, among different Han migrant communities in Beijing and Shanghai, for example) than China’s comparatively small Uyghur and Tibetan communities. If we stop thinking in minzu (民族) categories, as Robbie Barnett urges us, our understanding of “ethnicity” and “ethnic tensions” in China alters fundamentally.

What is perhaps more worrying, however, is the perception of a looming minzu crisis inside China. With each incident of Uyghur or Tibetan violence, the chorus for policy change increases, at both a popular and elite level. Racist taunts are de rigueur on the Chinese language Internet (even on more “liberal” spaces like Weibo), where one encounters constant gripes about how the Chinese state is “spoiling” those “ungrateful” and “backward” Uyghur and Tibetan minorities. Since the collapse of the Soviet Union in 1991, Chinese policymakers and intellectuals have warned that the P.R.C. could face a similar maelstrom of ethnic implosion unless fundamental changes are made to its ethnic policies: scaling back minority rights and preferences and perhaps even scrapping regional autonomy altogether.

Today, most Chinese believe current ethnic policies have failed. While some call for a federalist solution or more genuine forms of regional autonomy, many more agree with Professor Ma Rong of Peking University that the current approach places too much emphasis on ethnic/minzu differences (unnecessarily politicizing them) and not enough on a shared Chinese national identity. Recently, one of China’s leading and most influential policy advisers, Hu Angang, declared that “the foundation of the China dream (zhong guo meng/中国梦) is the integration of the Chinese nation’s guozu (国族), an old term first employed by Liang Qichao and Sun Yat-sen which literally means “state-lineage/race,” the sort of state-orchestrated “melting pot” that many believe requires the “mixing of bloodlines” (xie yuen ronghe / 血缘融合) through increased inter-ethnic marriage and physical contact.

Yet, any forced or rapid increase in inter-ethnic contact will only exasperate ethnic tensions and violence, in the short term at least. It’s worth considering for a moment, whether any radical cure to China’s “diseased ethnic policies” might actually end up further harming or even killing the “ethnic patients”? The admittedly provocative suggestion by Professor Barry Sautman, that China’s ethnic policies are actually more effective than those of both India and America, merits serious consideration. China has its fair share of ethnic problems, but are they any more serious than those confronted by other multiethnic societies across the globe?

For someone like me who is old enough to have the honor of being at the receiving end of ethnic violence during the Cultural Revolution, the current tension is a piece of cake, as it were. Having said this, there is no denial that we must address ethnic conflicts seriously. In doing so, however, we should not be tempted into a mode of showing sympathy only to the hapless minorities without taking account of the profoundly emotional state of the Chinese. Unlike in the past, today many Chinese blame the State for giving the minorities too much privilege, to the extent of spoiling them. That this sense of Chinese victimhood in the hands of minorities is shared by an increasing number of Chinese scholar-politicians is alarming. The very survival of the C.C.P. and the P.R.C., they argue, now hinges on nothing short of scrapping the Regional Nationality Autonomy system and the affirmative action policies. In other words, ethnic tensions are now regarded by the Chinese populace and its intelligentsia not as something marginal, nor as part of social issues or processes, but fundamental to the existential identity of China and its nation form.

This sense of Chinese victimhood and the proposed redressive measures raise a number of questions about the institutional arrangements that underpin ethnic relations in China. First, to the extent that the Regional Nationality Autonomy is one of the three basic systems of the P.R.C., can one surgically remove this alleged ‘tumor’ without undermining the entire political structure of China? Second, critics of China’s nationality policy, both within and without, have blithely ignored one important thing, i.e. the ‘special relationship’ between the C.C.P. and China’s ethnic minorities whereby the C.C.P. tried to liberate minorities from the Chinese Nationalist oppression and minorities supported the C.C.P. bid for power in China. Is China prepared to publicly and constitutionally rescind this covenant between the Party and the minorities, thereby abolishing the C.C.P.’s United Front strategy, its most prized strategy for victory? To say the least, insofar as the P.R.C. is a party-state, the road to a “normal” nation-state as yearned for by many Chinese, especially liberal intellectuals, is not going to be a smooth one, and the obstacles are not just the ethnic minorities.

While current ethnic tensions may be explained in reference to a number of ideologically inflated terms such as Chinese colonialism, nationalism or imperialism, or minority splittism or terrorism, we should account for the intense emotion or indignation expressed by both sides. Historically, the C.C.P. has introduced a powerful class-ethnic morality into ethnic relations by distinguishing its members as ‘good Han’ as opposed to the Chinese Nationalists whom they denounced as ‘bad Han’. Such a moral differentiation was instrumental for the C.C.P. to build ‘minzu tuanjie’ (national unity and amity among nationalities), which implies that after the Revolution, the Han Chinese, shorn of bad nationalists, are all ‘good’, and minorities, who have been subjected to so-called ‘democratic reforms’, are equally good people. Thus, ethnicity in contemporary China is supposed to be relationships between good ethnic groups. The problem with this conception of good ethnicity is that it has deprived both sides of self-reflexivity, leading to a propensity to blame others. This explains the profound sense of mutual deprivation among the Han and minorities today; neither side admits to being wrong.

There is also something very special about ethnicity in the Chinese cultural milieu. The dominant Chinese cultural ideology embedded in Confucianism is profoundly relational and ethical, and it is now promoted as China’s soft-power globally. Domestically, this ethical relationality dictates that minorities cannot maintain their “autonomy” or “difference” from the Chinese without offending Chinese ethical sensibility. What results is a Chinese moral rage at so-called minority “splittism.” The term splittism is not limited to territorial secession; it is also understood to be a deliberate centrifugal distancing from a righteously good China and the Chinese people. One can then reasonably wonder how high-handed punitive measures against alleged minority violation of China’s ethical ethnicity can help to win minority “heart” to China. What is China’s soft-power towards its minorities, today?

To take for granted the categories provided by the Chinese state is a grave analytic error. The fifty-six nationalities into which the P.R.C. divides its citizenry have an arbitrary relationship to ethnographic reality. The most extreme example of this arbitrariness is that Mandarins and Mins are equally considered 'Han'—tantamount to labeling Swedes and Englishman 'Germans'. Turning to the 'low population nationalities' (shaoshu minzu / 少數民族), the Rgyalrong are classified as Tibetan, although they are not and speak a wholly different language. As a result of their classification the Rgyalrong are taught primary school in Tibetan, resulting in a absurd perversion of mother tongue language policy.

Most P.R.C. nationalities are so low in population, so poor, and so marginalized that they are unable to articulate politicized resistance to the state in a way that garners international media attention. To see this process as evidence that ‘assimilation/sinification is perhaps the dominant trend throughout the nation’ is to celebrate cultural marginalization and oppression as a component of an inevitable ‘progress’ toward national homogeneity. The United States embraced such a discourse, as the fate of its indigenous peoples amply testifies. The organization of other polities (Belgium, Switzerland, the U.K. etc.) puts the lie to this chauvinist Hegelian perspective.

The discourse of ‘ethnic tension’ encourages us to mistake hatred of the state as conflict among citizens. Consider self-immolation. The Tiananmen Square incident (天安門自焚事件 23 January 2001) and the Tang Fu-zhen incident (唐福珍自焚事件 November 13, 2009), among many cases, show that this extreme expression of dissent is familiar to Chinese as well as Tibetan protesters. In setting themselves on fire out of disillusionment and desperation Tibetans are embracing an established Chinese mode of expression. This tragic phenomenon, far from showing ethnic tension, demonstrates the great extent to which Tibetans are actively participating in the dominant trend of sinification.

With a pen stroke the Chinese state created its nationalities half a century ago. Now, a tradition of incompetent governance continues to stoke the grievances that find articulation in what the state misidentifies as ethnic ‘splitism’.

A couple of years ago (in the volume Demystifying China) I noted that the construction of the category of ‘Han’ as but one subdivision of ‘Chinese’ eventually…

“entailed the construction of the category of ‘national minority’ … Modern China has incorporated diverse peoples such as the Tibetans and Mongols—peoples with a conscious historical experience of independence from China, replete with an awareness of themselves as a people possessing their own states, states with bureaucracies operating in their own languages. Incorporation into China has necessitated their uniform reduction—along with other groups sometimes numbering only in the tens of thousands, and having no similar national history—to the common status of ‘minority nationalities’.”

While it is true that the term “ethnic group” is problematic in English, in Chinese it is the unchanged linguistic regimen of the shaoshu minzu that prevails. And it remains a poor fit for a significant group of peoples subsumed under it, especially those people who sincerely see themselves as conquered, and now subject to deleterious Chinese domination in many aspects of life within their own domains. Indeed, in talking about ‘ethnic tensions’ it’s worth remembering that, with some exceptions, these tensions most prominently pit the dominant group and its state on one side against this or that minority nationality on the other.

And for Tibetans and Uyghurs, certainly, the structure is pregnant with the ingredients that precipitate the sort of tensions that the original question implies. The fact that the structure is top down, and exists in an atmosphere in which the peoples subject to it have no free voice with which to fully contest it—to whatever extent they feel moved to do so—further exacerbates tensions. The restrictions on free speech are, of course, not unique to the situation of the shaoshu minzu. But it is such a pervasive thing that it is easy to pass over it in discussions of specific minzu policies. The arrest of Ilham Tohti in January, and the announcement of trumped up charges against him (as well as the arrest of several of his students who are likely to be coerced into providing ‘evidence’ against him) speaks very much to this point.

And this brings up something else. How useful is it to approach the question just by quantifying the number of clashes between the state and, say, Tibetans or Uyghurs, during specific periods? This approach scants the effect of repressive measures, e.g., the barring of Tibetans from outside the Tibetan Autonomous Region from freely traveling to Lhasa, targeted arrests (meant to send broader messages), etc. Is the level of tension only to be understood from observable protest activities? I think this is far too simple. One must beware of not falling into the trap of measuring not the level of tension, but the effectiveness of repression.

However arbitrary the idea of the fifty-six minzu is, it’s the categorization that the state, and most citizens, works within. There’s certainly a rich brew of provincial stereotypes, and inter-communal tensions extend throughout society, down to feuds between one rural village and another carried over into the world of Beijing migrants. But these problems are seen as ones of local governance or law enforcement, not as carrying political weight through being “ethnic” issues. Crime is the classic example; when a Henanese gang fights a Hunanese gang, that’s an issue of law enforcement (i.e, which side gives the largest bribe to the local police), when a Uighur gang fights a Henanese gang, suddenly the second gang is “Han” and it’s an conflict that can have national reverberations. One of the side effects of the last year’s growing violence in Xinjiang has been that the Beijing police, and I assume other police forces elsewhere, have felt unleashed from any restraint in dealing with migrant Uighur, a heavy-handedness strongly applauded by the Han competing for the same space.

To pick up on Rachel Harris’ point, I’ve heard at least one Chinese strategist express private concerns about a growing Islamist presence even among the Hui, a massive and extremely well-integrated group, driven through Wahhabi influence through government-sponsored trade and exchanges between Ningxia and Saudi Arabia. This seems highly marginal, but there’s a whole shadow world of Islamic education for young men in China, usually through travel and study with individual teachers in small groups; several young Muslims I know have spent a year or two moving between cities to find the right teacher, or even traveling to Southeast Asia. Much of this is intellectual and mystical, but these are people looking for an identity who are somewhat vulnerable to being radicalized.

But there’s little sense of Islamic brotherhood between groups. The Hui’s growing presence in Xinjiang, their economic competition with the Uighur, and, most of all, the perception of them as agents of the Chinese state, going back to Nationalist times, means the Uighur generally despise them with a passion. Some Uighur even single out the Hui’s history of Sufiism as evidence that they were never “real Muslims” to begin with, despite the Uighur’s own numerous Sufi practices, picking up on Wahhabi language imported from Middle Eastern conflicts.

Although Hu Jintao chose the phrase “harmonious society” as the slogan for his time in office, the number of what are officially termed “mass incidents” in mainland China increased to over a hundred thousand per year during his tenure. As James Millward noted, that unrest is recognized by the state as to some extent the legitimate expression of grievances, while protests by Tibetans and Uyghurs are viewed quite differently—not as grievances, but as movements against China staged in collusion with foreign or exile “forces.” And this led to the Tibetan protests of 2008 being received in China by a wave of what one scholar at Lanzhou University termed “Tibetophobia, akin to Islamophobia in the West," leading to routine incidents of extreme suspicion and outright hostility towards Tibetans throughout the country. Since then, hotels in Han areas of China refuse to accommodate Tibetans or Uyghurs and are required to report any Tibetans, Uyghurs or sometimes Mongols who enter their premises to the Public Security Bureau. In one well-known case in 2008, a group of Tibetan scholar-cadres from Lhasa was flown to Beijing by China Central Television to appear on a discussion show about the excellence of China’s ethnic policies­—only to find that no hotel in Beijing would permit them to register.

Since then, the immediate tensions arising from the 2008 events have subsided and the intensity of nationalistic fervor generated during the Olympic year has also softened. But the Tibetans and Uyghurs have never been model minorities: this is not new. The tensions between these two groups and the State have always existed and will continue to do so. This is as much to do with historical memories and territorial disputes that have lasted centuries as with current policies; these are not recent inventions. In the case of the Tibetans, until recently the Tibetan populations had very little contact with Han Chinese except in border regions. Incidents of open conflict between the two groups based on ethnic animosity or tension were rare in the past, and never approached the type of visible mass ethnic riots common in multi-ethnic cities around the world today.

Even now, most Tibetans live in areas that are relatively isolated ethnically and very rarely come into contact with Chinese. Similarly, the vast majority of Hans have no contact with Tibetans. So the type of ethnic tension that is prevalent in Western cities is not visible in China, and most Chinese cities do not have ethnic ghettos with no-go areas for different groups. This is starting to change as the phenomenon of mass migration pushes Chinese to outlying areas and slowly generates ethnic enclaves. Tibetans, too, are now migrating to urban areas, particularly Xining and Chengdu, so urban ethnic ghettos may soon arise, with the problems that come with them.

The tension has to be gauged at two levels: on the one hand, in the relationship between the State and its citizens, and, on the other hand, in the relationships among the citizens. The neo-liberal homogenizing policies of the P.R.C. have created restive natives who view those policies as not of benefit to them or their communities and see the current trend as designed to relegate them to second-class citizenship. But the relationship between citizens cannot be generalized: we cannot view the entire Han population as anti-Tibetan or the Tibetans as anti-Han. The revival of Tibetan Buddhism within China (and in much of South East Asia) owes as much to Han patronage as to Tibetan devotion. Even current centers of Tibetan protest, such as Kirti monastery in Ngaba in Sichuan, are largely funded by Chinese Buddhists. The popularity of Tibetan Buddhism in urban areas is evident, and there is a growing exoticization of all things Tibetan within Chinese popular culture too, as illustrated by the popularity of the online fiction series “Tibet Code” (zangdi mima), soon to appear as a DreamWorks-China co-production. Even the winner of a recent Chinese Idol contest was a Tibetan, Phurba Rgyal (蒲巴甲/Pu Bajia).

These are not signs of outright ethnic animosity towards Tibetan people. But the State will need to face deep-seated issues of history and intolerance if it wishes to find any way to begin viable dialogues with its Tibetan or Uyghur populations.

As most of the commentators have argued in various ways, the key term in the question—“ethnic tensions”—carries in it a flawed set of assumptions. Some other more productive starting points might have been: Are extra-legal forms of discrimination against minorities on the rise? How are protests against the actions of state authorities interpreted differently for different minzu? Is majority and state tolerance decreasing for the expression of certain forms of identity, and if so why and with what effects? Has the association between the state and the Han Chinese gotten closer?

A certain interpretation of “on the rise” might call, as some contributors have pointed out, for a quantitative analysis of data that is not available. Even in its absence, though, it is clear that cleavages have formed in new situations and spaces. Friends in Lhasa note that ethnicity has become relevant in everyday socializing in work units where it had not in the recent past been an important line of difference in the daily routines of work and leisure. Moreover, since 2008, Tibetans and Uighurs have been turned away from hotels, restaurants, and other public spaces and private service industries. This type of profiling is based sometimes on official minzu status as inscribed on identity cards, but at others on assumptions made by readings of phenotypical characteristics, comportment, and dress.

Widely divergent geoeconomic dynamics and historical memories create different situations for variously positioned minzu. As a counterpoint to Tibetans, Uighurs and Mongolians, China’s Koreans have been seen as “model minorities.” Yet it’s difficult to imagine state authorities tolerating situations in Tibet and Xinjiang analogous to the activities of missionaries from South Korea in Yanbian.