Jonathan N. Lipman

Department of History

Mount Holyoke College

A Story of Violence, Nov. 1990

In a market town south of Kunming, capital of Yunnan province, some Chinese-speaking (Hui) Muslims from the countryside got into a fight with some non-Muslim Chinese--a Muslim child, playing with firecrackers, had damaged some merchandise in a bookseller's stall--and the Muslims beat up a couple of people. The police arrested them, mistreated them severely, and kept them in detention. After two days, their people wondered what had become of them, and two cars full of Muslims, led by a popular local man, came to town to look for them. Rumors quickly spread that these out-of-town Muslims had come to make trouble, and the police armed themselves in their station courtyard.

When the leader of the Muslims opened the station gate and entered the courtyard to inquire about his friends, the police lieutenant, son of a local power-holder, shot him at close range. Fatally wounded, he struggled to his feet, and his companions rushed to help him. All the police then opened fire, killing three of the Muslims and wounding ten. (My informant, who went to town the following day, saw the wounded himself in hospital, all of them in terrible condition.) The local Muslims came to fetch the corpses, and they took the bodies not to their home village but to a Muslim stronghold, the county town of Yuxi. Muslims converged on Yuxi from every direction in trucks and cars, bringing young men in hundreds and plenty of guns (including Uzis and AK-47s). Vehicles filled with explosives, manned by volunteers for martyrdom, were placed in position to blow up the police stations, if it were deemed necessary. An official of the county minorities commission and the Yuxi vice-mayor were taken hostage. Clearly, some of the Muslims were ready to confront the immense power of the state and die in the cause of vengeance for their murdered coreligionists.

The Muslims took over a portion of Yuxi city as their headquarters. After several days, their leaders received word from friends in Kunming that the army had mobilized from provincial bases and was heading for Yuxi. An announcement was made over centrally placed loudspeakers, and some of the men wrote Arabic prayers on white cloths, wrapped themselves in these shrouds, seized their weapons, and prepared to die. But after a dramatic wait, the army, restrained by provincial leaders, stopped short of the city, and the Muslims relaxed a bit. Muslim community leaders engaged in negotiations with civilian officials from the provincial capital, including the minorities commission, and succeeded in forcing the state to pay compensation to the families of the dead and wounded. Each aggrieved family was to receive Y8,000 for a man's death, and the wounded had all their hospital expenses paid by the government. The children of the dead were to be supported by the state to age eighteen. In the end, however, my informant claims to have discovered that the government found excuses to imprison the Muslim leaders by accusing them of drug smuggling, a common enough offense in the Yunnan borderlands, while the policemen were never punished. Though there are no visible signs of tension in Yuxi now, thousands of Muslims will remember the excitement, anxiety, and desire for revenge they experienced in those days.

A Story without Violence, 1990-1997

A Muslim quarter in the city of Zhengzhou, an important railroad junction in northern Henan province, was terribly dilapidated--its buildings old-fashioned and rickety, its markets and alleys insalubrious (in fact, nauseating), its people poor. Like other Chinese city governments, the reform-minded Zhengzhou administration wanted to undertake sweeping reconstruction in order to sanitize the city, improve its image and public health, and create significantly enhanced housing stock for the rising middle classes (almost all of them non-Muslim). Not coincidentally, many government officials and businessmen, especially contractors, would be able to make a great deal of money from the project. They recommended that the single-story buildings of the Muslim quarter be torn down, except for a few historic mosques, and be replaced by multi-story housing which would be too expensive for the majority of Muslims. This would have resulted, naturally, in the scattering of the Muslims to more affordable suburban neighborhoods where they could find apartments, shops to do their business, and markets to sustain them. This scenario had already played out in a number of other cities, and the Muslims were well aware of the potential consequences of urban renewal.

The complaints from the Muslim quarter were very loud. Muslims, they claimed, must live together near their mosques and their businesses, so relocation would destroy the solidarity of the local Huizu, the Chinese-speaking Muslim "minority nationality" in the official terminology. Sensitive to minority concerns (some claim overly sensitive), the city government negotiated with community leaders, who demanded that the government build xincun, "new urban villages" (equivalent to American public housing projects) in the same location as the old quarter and at prices the Muslims could pay. Ground was broken in the late 1980s, and now the quarter surrounding the Great Mosque, with its nearby women's mosque, is made up almost entirely of government-built apartment houses. The ten-year-old buildings are already shabby, but the new ones were finished only three years ago and remain beautiful, though their grounds have not yet been entirely finished. The residents are almost all Hui Muslims, of the lower and middle classes, and their new homes provide much better conditions than before. At no point in the process did anyone threaten violence.

Who are the Huizu?

In a recent book, Masumi Matsumoto argues that two "minority nationalities" (shaoshu minzu) dominated the CCP leadership's perceptions of China's non-Chinese inhabitants well before 1949: the Mongols (Mengzu) and the Chinese-speaking Muslims (Huizu). She claims that, because many Mongols and Chinese-speaking Muslims lived in and around the Yan'an (northern Shaanxi) region which was the CCP's headquarters from 1935 to 1949, those two, among the fifty-six "minority nationalities" currently recognized by the People's Republic of China (PRC), occupy unique places in the PRC's "nationality policies" (minzu zhengce) and "nationality discussions" (minzu lun). As the most familiar minzu, the ones most easily available for study and application of experimental policies before 1949, these two certainly had an impact on the CCP all out of proportion to their size.

They also presented the CCP with very different problems of definition and governance. The Mongols strongly resembled Stalin's definition of a "nationality" (Rus. narod). That is, to the CCP's leadership and their representatives, the Mongols appeared to live almost exclusively in their own historic and well-defined territory, to use primarily their own language, to practice nomadic pastoralism in contrast to agriculture, and to possess a relatively homogeneous culture marked by common folkways such as food, music, horsemanship, and wrestling.

The Hui may have looked similarly well-defined from the vantage of Yan'an and its rather limited hinterland, but once the CCP took national power, its minzu experts discovered that the widely distributed, culturally diverse, Chinese-speaking Muslims resembled the Mongols hardly at all. Indeed, the presence of considerable numbers of Muslims throughout the Chinese culture area has created difficulties of both perception and policy for every China-based state since the Ming empire. Living in every province and almost every county of the PRC, the Hui have managed simultaneously to acculturate to local society wherever they live and to remain effectively different from their non-Muslim neighbors. Most of them use local Chinese language exclusively, and they have developed their "customs and habits" in constant interaction with local non-Muslims, whom they usually resemble strongly in material life. Intermarriage has made them physically similar to their neighbors (with some exceptions in the northwest), but their Islamic practice and/or collective memory of a separate tradition and history allow them to maintain distinct identities. In short, they are both Chinese and Muslim, a problem that must be solved within many local contexts, for there is no single isolated territory occupied primarily by Hui people which could serve as a model for Hui all over China.

From the 15th to the 20th century, Hui (or the earlier Huihui) was simply the Chinese word for "Muslim." The ethnographers of the Qing empire divided its subjects by language, so that male Chinese-speaking Muslims, being Sinophone, had to wear the queue, the pigtail that signified Chinese submission to the Manchus. The Turkic-speaking peoples of the Tarim Basin, on the other hand, though also Muslims and subjects of the Qing, were exempt from the queue, except for their leaders, who had to travel to Beijing to pay homage to the Manchu emperor. The Qing empire's northwesternmost territory, created in the mid-18th century, was often called Huibu, the territory of the Muslims. Various types of Muslims received descriptive ethnonyms--the chantou Hui (Muslims who wrap the head) were the turban-wearing Turkic-speakers of the Xinjiang oases (today subsumed within the Uygurs), the Sala Hui were the Turkic-speakers of northeastern Tibet (today's Salars), and the Chinese-speaking Muslims now included in the Huizu were sometimes called the Han Hui, the Han Chinese Muslims.

There was no name for what is now called the Huizu, and there were few institutional connections among the Muslim communities scattered all over the Qing empire, except for trading networks, which allowed Muslim merchants to find a mosque for prayer and ritually pure (Ch. qingzhen, Ar. halal) food as they traveled, and teacher-disciple networks of religious professionals. The Manchu center had a powerful interest in discouraging widespread, unofficial organizations within its frontiers, and voluntary associations of all kinds had long been suspect within the Chinese cultural nexus. Islam, with its outlandish sacred texts and unconventional mosque-based community structure, appeared heterodox to many Qing officials (whatever their ethnicity) and thus bore careful watching. Many magistrates and provincial-level functionaries wrote memorials warning the throne about the strong resemblance between Islam and the Buddhist and Daoist sects, which so often caused social disorder. This perception grew stronger when Sufi orders began to spread in northwest and southwest China in the late 17th century. The Sufi order (Ar. tariqa), with its centralized, hierarchical religious authority vested in the shaykh (Ch. shehai or laorenjia), appeared particularly subversive to Qing officials because of its intercommunity connections through the peripatetic shaykhs themselves or their representatives. In addition, the shaykh appointed imams to each mosque rather than each community selecting its own religious professionals. These were the first formal structures of extra-local power within the world of Islam in China, and the Qing state proscribed and persecuted those which were perceived to be inimical to social order.

The Sufi orders did not, however, constitute "national" organizations to which all Chinese-speaking Muslims, all over the Qing empire, might turn for redress of grievances or influence at court. Quite the contrary, they remained regional, divisive, and pitted primarily against other Muslims rather than attempting to establish Islamic unity. The more "national" type of institution and identity had to await the first glimmerings of "modern" China at the end of the 19th century. Urban Muslim intellectuals from Beijing, Nanjing, Suzhou, and other eastern cities received that "modern" impulse, like their non-Muslim intellectual colleagues, from Japan and (more distantly) Euro-America. They discovered that a new world awaited them, and that a New China had to be created to survive in that world. In northwest China, in contrast, the modernist impulse arrived initially from the Middle East in an Islamic and (sometimes) Islamist package, emanating from the Ottoman empire, Egypt, and Saudi Arabia. For either group of Muslim intellectuals, a national Muslim identity could only be found in the context of a newly emerging Chinese identity. Looked at in this way, the Huizu (Hui nationality), as a Chinese ethnic group (a self-conscious social entity), could only come into existence once a Chinese nation-state had been created to contain it.

During the last years of the Qing and the chaos of the early Republican period, these intellectuals struggled to form national Hui organizations to overcome what they saw as the ignorance and provincialism of their coreligionists. Like their non-Muslim countrymen, they relied for debate and consensus on numerous periodicals, such as Xinghuibian, one number of which was published in 1908 by a group of Muslim overseas students in Tokyo; Islam, which flourished in Henan in the mid-1930s; and Beiping's long-lived Yuehua, founded in 1929, which continued to appear until 1948. In the atmosphere of crisis, Social Darwinist thinking, and powerful Japanese intellectual influence, some of them came to see the Hui as a minzu, a distinct and genetically-defined people (race), while others continued to claim that Islamic religion (usually called Huijiao, the Hui teaching) constituted the essential binding force. Non-Muslim leftist intellectuals such as [Fan] Changjiang agreed with the former position, some imams took the latter stand, while the great historian Gu Jiegang concluded that all peoples of the former Qing empire, including Han Chinese, Muslims, Mongols, Manchus, and Tibetans, constitute a single Zhonghua minzu, the Chinese national family. By the 1930s, the CCP, under Stalin's influence and drawing on detailed research done by its own scholars, had begun to refer to the Chinese-speaking Muslims as the Huizu (or Huihui minzu).

May-Fourth-style intellectuals were not alone in working to create a national identity for the Chinese-speaking Muslims in a new China. Imams influenced by Islamic modernism from the Middle East also founded institutions to create a new consciousness for the Hui. Beginning with Sino-Arabic primary, secondary, and normal schools (Ch. Zhong-A xuexiao), they took education in both religious subjects and the secular, modern curriculum, which they called Chinese after its medium of instruction, to be the responsibility of Muslim communities. Even in parts of the remote northwest, scholars basing themselves in the Qing period Islamic-Confucian canon designed schools to teach a mixed Islamic-Chinese curriculum which rapidly came to include foreign languages, mathematics, science, and other "modern" subjects. Obviously it was crucial that the Chinese-speaking Muslims be defined advantageously within a new China. The impact of powerful government policies might well depend upon whether they came to be considered "insiders" (that is, fully Chinese) or "outsiders" (that is, members of an exotic minority group) in relation to the Chinese national project(s). Muslims never agreed among themselves, for more "progressive" groups claimed status as an independent minzu, while "conservative" imams held that Islam alone defined their identity, and scholars in the Guomindang (Nationalist, GMD) camp argued for a single comprehensive Zhonghua minzu.

On the mainland, at least, the CCP won, and these tensions informed the definition of the Huizu during the "nationality identification" (minzu shibie) project of the 1950s, when the Hui became one of China's oddest minority nationalities--both Chinese and not Chinese, containing people who speak a variety of languages and live all over the country--and a unique object for minority policies at every level of government. Ideologically constrained by its own commitment to Stalin's criteria, the PRC could not recognize Islam alone as the distinguishing feature of the Huizu, but neither could the state ignore the fact that many of the differences between Hui and their neighbors derived from religion.

Pressing political questions emerged for national, provincial, and local leaders and Party structures: What could be done about the legitimate power of the mosque and its religious professionals within Hui communities? How should the state handle the connections between Hui communities, which made them, appear conspiratorially linked in ways that non-Muslim Chinese could not be? How could the long-standing local antagonisms between Hui and their non-Muslim neighbors, or between competing groups of Hui (especially Sufi orders and 20th c. Islamist groups) be tamed or eliminated? How could Hui neighborhoods, often tightly knit enclaves cheek-by-jowl with non-Muslim neighbors, be effectively integrated into China's burgeoning urban life? How might Hui villages, autonomous, sometimes exclusive and (in some places) habitually resistant to state power, be transformed into sites for socialist development? How could Hui be brought into China's national project(s) as compliant "minority nationality" participants? That is, what types of organization could the state allow, or impose, which would enable effective control over the Hui, generally perceived to be antisocial and prone to violence. Few of these questions have been answered finally or consistently in the PRC, and they continue to trouble the state's local authorities throughout China.

Interviewing Hui all over China in the summer of 2000, I asked dozens of people--professors and cabdrivers, farmers and factory workers, businessmen and imams--what all Hui everywhere in China share in common? Hui working as "nationality functionaries" (minzu ganbu) gave me the conventional answer dictated by the government's definitions, namely that all Hui share descent from Arab and Persian sojourners and a host of "national characteristics, customs, and habits." Some informants described an emotional attachment to the genetic claim to homogeneous minzu status, the notion that "All Hui under Heaven are a single family." After considerable discussion, however, most people answered, "Not much," for they accurately perceived that Hui-ness, or "doing Huihui" (Ch. zuo Huihui), cannot be dissociated from its local context. Even Islam, with its vision of a universal congregation of believers (Ar. umma) no longer touches all Hui. Many have become secularists, atheists, members of the Communist Party, and thus have separated themselves to some extent from the religious life of their communities, though by the CCP's own definition they remain members of the Huizu, people we might call ex-Muslims.

A young urban Hui salary man in Yunnan went through a whole litany of differences among his fellow Hui--most of them dictated by geography and local culture--before announcing that there are only three things that all Hui share: the white skullcaps which religious Hui men wear either all the time or (at least) in the mosque precinct; youxiang, the deep-fried wheels of slightly sweetened dough which are part of every Hui festival, feast, and life cycle celebration; and consciousness of common Middle Eastern blood. As for the rest, he said, it depends on where you are and whom you ask. So diffuse and elusive an entity presents the PRC with multifaceted, diverse challenges ranging from military control to urban renewal to ideological uniformity. The remainder of this essay will address both the state's actions and some Hui communities' initiatives, resistances, and compliances.

The PRC and the Huizu

The pre-1949 relationship between the northwestern Muslims and the CCP followed no single unified path. It included the allegiance of some progressive Muslim intellectuals to the Party and ferocious Muslim military resistance to the Long March, as well as the cooperation of some local Muslim elites with the CCP's anti-Japanese policy after 1935. In Gansu, Qinghai, and Ningxia, most Muslims belonged to communities dominated by families of Muslim warlords (all surnamed Ma) who joined the GMD. Some Muslims, however, joined the CCP out of poverty, sympathy for the Party's anti-Japanese position, or opposition to the GMD elites in their communities. Committed to the broadest possible social coalition against Japan and the GMD, the CCP did not, at this point, take a strong stand against religion or religious leadership. After the Long March, Edgar Snow talked to northwestern Muslims fighting in the Eighth Route Army, who told him, "The Chinese and the Moslems [sic] are brothers; we Moslems also have Chinese blood in us; we all belong to Ta Chung Kuo [Da Zhongguo, China], and therefore why should we fight each other? Our common enemies are the landlords, the capitalists, the moneylenders, our oppressive rulers, and the Japanese. Our common aim is revolution."

During the anti-Japanese war, the CCP moved away from the Comintern's earlier policy, which allowed self-determination for minority groups as nations, and toward a United Front policy in which all patriotic elements within the great family of the "Chinese nationality" (Ch. Zhonghua minzu) should be encouraged to resist Japan together. Mirroring Stalin's reversal of the right of secession for "minority nationalities," the CCP proclaimed the indivisible unity of China (by which they meant the entire territory of the former Qing empire) and the responsibility of all citizens to love and protect the motherland. The CCP leadership had already decided that China was to be a multi-minzu state, for on their Long March they had met not only Muslims and Mongols but also the culturally diverse (and hard to control) non-Chinese peoples of Guizhou, Yunnan, Sichuan, and eastern Tibet. Following the lead of Sun Yat-sen, they declared the Hanzu, a vast mosaic of peoples living from Siberia to the tropics, the overwhelming majority of China's population, to be a single, undifferentiated nationality, "children of the Yellow Emperor," sharing blood and history. All the rest became "minority peoples," for they belong to China-the-country and should be regarded as younger brothers in the great minzu family. According to the CCP analysis, each people will inevitably follow the most advanced--the Hanzu--through Lewis Morgan's teleological stages of advancement toward the light of socialism and communism.

This rosy picture, of course, could never be implemented in practice. Even the process of deciding which peoples of China were to be designated as distinct minzu engaged the Party and its ethnographers in contentious and highly politicized debates in the mid-1950s. Over 400 groups "applied"--that is, they were considered sufficiently minzu-like that scholars were sent to investigate and report on their conformity to the official definition(s) borrowed from the Soviet Union. Party committees at the central and provincial levels made the final decisions. The case of the Huizu, among many others, reveals that Stalin's carefully enshrined terminology, still repeated regularly in academic and popular literature on the minorities, was honored as much in the breach as in practice. The members of the Huizu, as currently defined in the PRC, have no common language, no common territory, and no common economic life, though they are widely held to be genetically inclined toward skill at doing business in the marketplace. As for common psychological make-up, or culture, Islam itself constitutes their sole common heritage, and their "customs and habits" tend to differ from region to region except for those which derive from their religion.

Since the "nationality identification" of the 1950s, and especially since 1978, communication and transportation have improved all over China, and the state's designation of the Huizu as a coherent social entity now has considerably more validity than it ever did in the past. Because the state has been willing to fund official minzu institutions of all kinds, members of the Huizu are now aware of their minzu identity (which they certainly were not in 1949) and that their minzu has (for example) "minzu costumes," "minzu folksongs," and "minzu literature" both ancient and modern. There are Huizu research institutes, Huizu exhibits at minzu theme parks all over China (as well as Hawaii and Florida), and Huizu variety performances on television on New Year's eve. After years of "minzu work" coming from the state, many Hui are entirely convinced of the common blood they share with all other Hui.

Some Hui take pride in the institutionalized affirmation of their "national" existence, while others find it false, condescending, or downright silly. According to a young Hui worker, "we all loathe those dreadful 'minzu village' theme parks; they're just places for Han to go and feel superior to the primitive natives." Pointing to his jeans, t-shirt, and baseball hat, he said, "I'm certainly a Hui, but does this look like Huizu clothing to you?" At the same time, many Hui earn their living by research, collection, exhibition, and reification of their "national" traditions, including scholars in work-units all over China dedicated to the study of the Huizu heritage, dating back to the Mongol period or even the Tang but heavily focused on the present.

There is no question that the 1300-year presence of Islam and Muslims in the Chinese culture area has produced a remarkable synthesis. The 17th-19th c. Confucian-Muslim texts, for example, constitute a rare example of profound Islamic philosophy in a "non-Muslim" language. Sino-Arabic calligraphy is a striking adaptation of a hallowed Middle Eastern art form to the formal aesthetics of Chinese culture. The PRC's definition of the Huizu, however, invariably distorts what have been highly localized evolutionary processes. In virtually any realm of life except religion, we may find as much difference among Hui all across China as between Hui and their non-Hui neighbors.

Structures and Personnel of Authority: The Mosque and the Minzu Ganbu

As a system of religious authority, Islam has been relatively decentralized since the eighth century. Sunni Islam in particular, to which practically all Muslims in China adhere, has never been led by a centralized church or hierarchically organized clergy. Each congregation or mosque (Ar. jami'a) can employ religious professionals to lead prayer, to preach, to teach, and to interpret the religious law (Ar. shari'a) and the sacred texts. In conflictual or recondite cases, these learned leaders (Ar. 'ulama) may refer questions or doubts to famous scholars at majors centers of learning, who may reply with an authoritative opinion (Ar. fatwa), but these opinions cannot be implemented by any universal system of religious enforcement. As the Islamic community came to include large numbers, and finally a majority, of non-Arabs (and non-speakers of Arabic), this internal flexibility allowed congregations in the southeast Asian archipelago, in the Atlas Mountains, in the Caucasus, and in China (among many other places) to follow the orthodox tenets of their faith without imitating the social or political forms of Islam's Arabian homeland.

In the vast Chinese culture area, with its diverse ecologies and social landscapes, Muslim communities have thus fit into many different local scenes without following a rigid blueprint. They certainly share many characteristics which distinguish them from their non-Muslim neighbors--the centrality of the mosque and its professionals, for example, and the sacred texts in Arabic which mark them as outlandish, however thoroughly they have become "local." Since the 14th century, however, Muslims in China have come to use local Chinese as their sole vernacular, adding Arabic, Persian, and Turkish words to it as an internal lexicon of recognition and authenticity. They have placed themselves in economic niches which take advantage of their intercommunity connections, and they rely on one another for hospitality, halal food, mosques for prayer, and trustworthy trade partners in the face of the non-Muslims. But they have few formal structures of intercommunity authority, except for the Sufi orders mentioned above, so they have generally been able to adapt to local (non-Muslim) contexts without threatening either social order or the state and its local representatives.

In the People's Republic of China, the centralized structures of the Huizu, in parallel with those of the other "minority nationalities," have been created as hierarchies by the state. The "nationality cadres" (Ch. minzu ganbu) of the Hui are organized into at least two sets of organizations--one dealing with religion, the other (in theory) resolutely secular and "ethnic," to handle minzu problems. The China Islamic Association (Ch. Zhongguo Yisilanjiao Xiehui, shortened to Yixie), which includes members of all ten officially-defined Muslim minzu, takes charge of Islamic religious affairs and mediates all open, "legitimate" contacts with Islamic organizations and Muslims outside the PRC. The central Yixie in Beijing, for example, allocates places in the government-sponsored pilgrimage (Ar. hajj) delegations to Mecca, highly desirable to Chinese Muslims because the state foots the rather large bill for transport, lodging, and food during the month-long journey. In addition, there are religious affairs offices (Ch. zongjiao shiwu ju) at the provincial and local government levels to deal more directly with local religious issues. The Central Nationality Affairs Commission (Ch. Zhongyang Minzu Shiwu Weiyuanhui, shortened to Minwei), on the other hand, deals with political and social issues involving all of the shaoshu minzu. The Yixie, Minwei, and religious affairs offices have provincial and county level structures, sometimes headed by Han but staffed largely by minzu ganbu, and members of these organizations can see their functions as quite separate, though they may have overlapping personnel and might often work together.

In theory, at least, these structures allow members of the Huizu to take charge of their own communities. Some informants, however, scoffed at the idea that the minzu ganbu and their hierarchical minzu and religious organizations have any power or influence. One Xi'an Muslim shopkeeper told me that "We solve our problems here ourselves, without any interference from them. The minzu ganbu are useless." His neighbor, who also has a shop, claimed that no minority person would ever be given a position with influence. A Kunming Muslim worker agreed with this conclusion, and said that the state authorities use the minzu ganbu as a front, as window-dressing, while giving all the effective jurisdiction to Han functionaries. "They would never trust a Hui with any real power."

Disagreeing with this opinion, a Ningxia minzu ganbu holding joint positions in the provincial Minwei and Religious Affairs Office has participated actively in local conflict resolution throughout his career. He sees his work as a mediator within the Huizu as a serious contribution to social order, preventing the kinds of conflicts which had in fact polarized the northwestern Chinese Muslims from the 1760s until the 1930s and led to sanguinary violence. But some Ningxia Muslims openly expressed their contempt for such people, who join them in oppressing us. The minzu ganbu of course denied this, claiming that people who hold such opinions are ignorant of the vital contribution made by the religious and minzu affairs functionaries.

The construction of top-down bureaucracies to handle Huizu affairs might seem to homogenize at least the administration of local Hui communities all over China, and this has indeed been one effect of the PRC's consolidation of effective central power in the last half century. But this has not had much leveling effect on the types of conflicts, problems, and distinctions which confront Hui in their local contexts all over China. As the two stories at the beginning of this essay indicate, some areas are more violent than others, presenting the authorities with intractable problems of rivalry and collective memory of conflict stretching back into the distant past. In contrast, some Hui communities are engaged with the state (cooperatively and/or conflictually) in solving what appear to us to be much more modern, "normal" dilemmas.

The persistence of Islamic authority--the 'ulama (usually called ahong in Chinese, from Per. akhund, "teacher"), Sufi shaykhs, and elders of the local mosque--continues to provide an internal alternative to the formal structures of Yixie and Minwei, one which sometimes appears subversive to non-Muslim officials. Ahong must be very careful not to tread on the toes of state functionaries or to invade their realm of authority. One elderly religious teacher in Xining, a man of eminent descent and great learning in the Muslim tradition, talked to me for over an hour without admitting that anything ever goes wrong in his community, or that religious conflict had ever occurred in the large provincial city in which he constitutes the highest Islamic authority. Though evidence to the contrary exists, he felt he had no choice but to describe the state structures, and his own mediatory role, as entirely successful in keeping social order.

The Primacy of the Local: Hui as Middlemen

Both Chinese and Muslim, the Hui may be seen as occupying the cultural margin of Chinese civilization, part inside and part outside. More literally, many Hui communities are physically located on the edges of the Chinese culture area, where it abuts southeast Asia, Tibet, Central Asia, and Mongolia. In these locations, the Hui often serve as middlemen, brokers between Chinese and non-Chinese cultures. They are not alone, for members of other ethnic and cultural groups can also perform these functions, but the Hui have proved themselves uniquely suited to the position. In Lintan Jiucheng, a market town in southern Gansu, one Hui solidarity has over a century's experience in economic brokerage between China and Tibet. Tibetan is taught beside Chinese and Arabic in the community's schools, and they have designed their economic activities to maximize exchange between the mountain and grassland products of the Tibetans and the lowland artisanal and industrial production of cultural China. Knowing that the Tibetans have particular tastes in textile colors and designs, for example, this Muslim group has purchased two silk factories in Hangzhou, far away in eastern China, where they manufacture cloth to suit Tibetan preferences. They transport the goods by rail and road to Lintan Jiucheng, where they are re-packed and loaded onto small pick-up trucks, which the individual brokers use to reach their tent-dwelling customers in the high country.

Hui communities like Lintan Jiucheng's may be found all around the great semi-circle which surrounds the Chinese culture area on its inland side. Hui have also placed themselves throughout China proper, in villages, towns, and cities where they are surrounded by a vast majority of non-Muslims, overwhelmingly Han Chinese. In those locations they constitute a kind of internal frontier, where they sometimes continue their middleman function. For example, Hui families all over China have taken jade transport, jade carving, and jade selling as a specialty. Even in Beijing and Shanghai, some of the most famous and prosperous jade dealers are Hui. The Muslims have undertaken this work in part because the most important domestic source for jade within the PRC (and the Qing empire in centuries past) lies in Xinjiang, China's huge northwesternmost province, whose population was until the mid-20th century overwhelmingly Turkic-speaking and Muslim.

More quotidian trades have also attracted Hui merchants. In part because of their religion's dietary prescriptions, many Hui engage in butchering (eschewing pork, of course), in inn-keeping, and most obviously in recent decades, in preparing and selling food. The halal (Ch. qingzhen) restaurant is ubiquitous in China, assumed to be cleaner than its non-Muslim counterparts and to serve tasty, slightly exotic specialties such as Xinjiang fried noodles, Central Asian bread, and lamb stew. In the past few years, high-end Hui restaurants have appeared in major Chinese cities, serving halal versions of the same dishes as non-Muslim establishments--roast duck, whole fish, crab, and more. The more common Hui food stall is located in a market street, and its customers sit on stools at small, rickety tables to eat halal ("Hui") adaptations of local everyday foods. Like their non-Muslim neighbors, Muslims in northwest and north China prefer noodles, steamed bread, and other wheat products as their main grain foods (Ch. zhushi), while Yunnanese, Sichuanese, and other southern Muslims eat their meat and vegetables with rice.

Xi’an. The Muslim quarter of Xi'an is clearly distinguished from nearby non-Muslim parts of the city not only by its old-style housing (see below), but also by its dense concentration of Muslim restaurants, bookstores, mosques, shops, and schools. In the past twenty years, the residents of the quarter have prospered in business and rebuilt many of their substantial mosques. The largest and oldest of them, the Great Mosque, dates from the Ming period and has been designated as a national historic site by the PRC government. Since the Great Mosque has been included on the flourishing Xi'an tourist circuit, hundreds of thousands of visitors make their way on foot through the quarter's alleys each year. Responding quickly to this potential market, the Hui of the quarter have become souvenir dealers on an enormous scale--one can buy "Muslim" objects, such as white skullcaps, calligraphic scrolls in Arabic, or porcelain decorated with Arabic inscriptions, as well as conventional "traditional Chinese" knick-knacks from the hundreds of stalls which crowd the narrow walkways.

Conspicuous economic success marks the Xi'an Muslim quarter, but so does Islam. In the early 1990s, religious leaders became increasingly uncomfortable about the sale of alcoholic beverages by Hui foodsellers in the quarter. Most Xi'an Hui follow the Qur'anic injunction against consuming alcohol, but merchants eager for non-Muslim customers found it necessary to acquiesce to their clientele's demands for alcoholic drinks with their food. During his Friday sermons, a courageous ahong began to denounce the sale and consumption of alcohol, and a committee was formed to ban alcohol from the quarter. This organization, which has few parallels in other parts of China, agitated openly, held rallies, and confronted merchants who refused to cooperate with the ban. After several years of success, the committee was suddenly declared illegal by the local authorities, since it had not applied for permission to incorporate as an organization. Though it received support from all of the Xi'an area's eighteen mosques and even from non-Muslim merchants operating in the vicinity, the committee would only be allowed to operate if it accepted a CCP member in its ranks and allowed him to report on all its activities.

This particular conflict, as yet unresolved in 1998, placed the quarter's Muslims in a potential confrontation with the state, and with non-Muslim merchants and customers who resented the alcohol ban. Activist Hui had skillfully used the state's own minority policies, which permit socially acceptable "minority customs and habits" to flourish, but they were unwilling to resist the government's label of "illegal organization" (Ch. feifa zuzhi) by direct confrontation or going underground. No national Hui organizations joined the struggle, though provincial Yixie representatives voiced their support for the alcohol ban during an investigatory meeting. Its substance--alcohol avoidance--was universally Islamic, but this was a local battle, fought by local Hui, over a local issue.

Sanpo. East of Xi'an, on the south bank of the Yellow River in Henan province, lies Sanpo, an entirely Hui community of 4,500 near Zhengzhou. Sanpo differs from nearby Han Chinese communities in the centrality of its ten mosques in collective life and its devotion to the trades of the tanner and furrier, especially in sheephides and wild animal pelts. These occupations, and the marketing of furs and leather which results, have proved most profitable, but they also give the village a powerful and penetrating smell. Some Sanpo villagers have migrated to the hill country of eastern Gansu, where they raise sheep on barren hillsides, then bring the hides to Sanpo for processing. The products are sold to clothiers and to boutiques in the major cities of eastern China, and more recently to Europe, Japan, and the United States as well. As a sideline, some enterprising Sanpo merchants have also invested in the manufacture of woolen rugs for export.

Marked as different from their non-Muslim neighbors by their religion, their trade in sheep, and their village’s odor, the Sanpo Hui are also quite distinct within the world of Islam, for their village has five mosques reserved for, and led by, women. Women's mosques (Ch. nüsi), as distinguished from a curtained-off women's section in an ordinary mosque, may be found in very few parts of the Muslim world. In China, they exist primarily in the North China plain (with outliers in Xi’an and Lanzhou),and in Yunnan, where virtually every significant Muslim community has at least one. Muslims from elsewhere in the world find the women's mosques peculiar at best, unnecessary or even heterodox at worst, but local Muslims in China defend them as perfectly compatible with Islamic orthodoxy. They believe that women can find a more satisfying spiritual and community life within their own institutions, taught by their own female imams (Ch. nü ahong).

From these brief vignettes we may move toward the conclusion that relationships between the Hui (as individuals, as communities, and as a "national minority"), their non-Muslim (or non-Hui) neighbors, and local government are determined more by local issues, conditions, and personalities than by any national agenda on either side.

The Hui vs. the State: Negotiating urban renewal

The Muslim quarter of Xi'an, just behind the Bell Tower at the center of the old city, is distinguished not only by its religious institutions and its anti-alcohol committee but also by its antiquated housing. Xi'an has undergone a thorough urban renewal in the past twenty years, and only the Muslim quarter remains as a remnant of the old days, its many low, run-down buildings constituting a blot on the urban landscape, according to city planners, who much prefer modern apartment blocks with flush toilets. Like the Muslim sections of Zhengzhou and many other cities, this Hui quarter (Ch. Huiminfang) has been under increasing pressure to submit to government plans for gentrification and resettlement of some of its population in the suburbs. The Muslims, too, would like to have running water, modern apartments, and wider, cleaner streets, but not at the cost of values and social conditions which they perceive to be essential to their community life.

Like the Muslim urbanites of Zhengzhou, the Xi'an Hui have lobbied their local government, but their negotiations have taken place in an entirely local personal and political context, without involvement by central authorities and without any systematic reference to experience elsewhere, though several informants referred informally to the positive models of Jinan and Shenyang. They have demanded that any urban renewal plan leave room for them to maintain their solidarity, and some of them have organized to define their community's interests. Though they certainly desire to be seen as progressive, these Hui also have deep anxieties about the survival of their community life, their extended families, their mosques, and their businesses, fears which neither local nor provincial government planners have been able to allay.

Ma Liangxun, the leader of the Xi'an Hui community in relations with city and provincial authorities, holds the official position of vice-chairman of the Religion and Minzu Affairs Committee of Shaanxi province. He is also the brother of a leading ahong of the quarter, Ma Liangji, who recently retired as the religious leader of the Great Mosque and now devotes himself to Muslim construction and commemoration projects. But Ma Liangxun is not a businessman, so when plans were first made to rebuild the Hui quarter, the city government sought a Hui business leader to direct the contracting and concentration of capital for investment in renewal, but none could be found. A successful Hui entrepreneur managed to achieve a government position from which to play the middleman, but his efforts failed.

Reacting to the potential transformation (or elimination) of the quarter, a group of local men, including small entrepreneurs, academics, schoolteachers, and religious professionals--not the most visible leaders but solid citizens with credible voices--formed the Xi'an Islamic Cultural Study Society (Ch. Xi'an Shi Yisilan Wenhua Yanjiuhui) in 1995. Their work has included holding conferences, publishing edited volumes of essays, welcoming guests foreign and domestic, conducting formal and informal relations with Muslim communities in other parts of China, and, most relevant here, doing research on the Hui quarter in order to provide data to residents and to the local and provincial governments. This formal group, permitted by the state under the category of "popular [Ch. minjian] organizations," has worked publicly and openly as an advocate for the Hui community. One of its intellectual leaders, Zhu Songli, thought that he could help to ease the tension and conflict over urban renewal in the Muslim quarter by gathering information and listening to the residents. Everything was done through official channels, in no way resembling the violent confrontation in Yunnan described at the beginning of this essay:

We worked out an outline for investigation and research, then submitted it to the superior departments responsible for the work and the relevant provincial leadership. In the past half year, we have invited concerned people from every walk of life in the Hui quarter to several discussion meetings to hear their opinions. We have made a number of specific visits [to individuals] and produced and distributed more than a hundred questionnaires. Naturally, our work is still quite superficial and the materials we have gathered very limited and scattered. But undertaking this research has nonetheless impelled us to think about many problems, some from the comments of our members and others discussed in the reactions to the questionnaire, and has given us considerable inspiration. The opinions of Xi'an Muslims on urban renewal, reported both by Gillette and by the Cultural Study Society, differ by the respondent's wealth and education, by religious affiliation and personal piety, by individual goals, and even by family size. Numerous objections have been raised to the government's plans, but no unified counterproposals have come from the Muslims. Rather, they continue to discuss urban renewal issues with one another and, through official and unofficial representatives, with the state. As of the summer of 2000, no renewal projects had yet broken ground in the quarter. Only a few old public apartment buildings had been vacated and torn down, but work had not begun on any replacements for them.

Yinchuan. Facing a similar urban problem, a mosque community in Yinchuan, capital of the Ningxia Hui Autonomous Region, took a very different path. Their 19th century mosque, built in the old Chinese style, had been desecrated during the Cultural Revolution, and they very much desired a new mosque in the heady, vibrant reform economy of the 1980s. On the same site as their old mosque, they constructed an "Arabic-style" complex, with a dome and two towers, and also built half a dozen large apartment buildings surrounding it. The housing, some of it owned and operated by the mosque as an endowed property (Ar. waqf), generates income for the religious institution. It also guarantees that this particular Hui community will not be broken up by gentrification or further urban renewal. Though the complex does not contain retail businesses, there are market spaces nearby where the Muslims can shop and operate their own trades. The flexibility of this community is due in part to the relative anonymity of their location in a small provincial city--Yinchuan, unlike Xi'an, does not lie on the major national or international tourist circuits, and this mosque had been located on the outskirts for centuries.

At least one conflict did mar the mosque's happy renewal. Like all modern mosques, this one broadcast the early morning call to prayer through an amplifier and loudspeakers, located in the tall minarets, just before dawn. Pious Muslims, of course, find this call, beginning with a resounding "God is great!" (Ar. Allahu akbar), to be a welcome invitation to communal worship, to the satisfying ritual of the new day. Non-Muslim families living in the nearby apartment buildings, however, unaccustomed to Islamic religious routine, resented being suddenly awakened by their neighbors' amplified and unintelligible noise. Arguments occurred, unpleasant but not violent, and the mosque's leaders agreed to a meeting. The city religious affairs office played a part in the negotiations, and a compromise was reached by which the first call to prayer of the day was made unamplified from the mosque's tower, while the remaining four were announced as usual through the sound system. This was certainly a far cry from the battles and massacres of the region's past. Sufi orders had flourished and fought one another in the area, and a major Muslim rebellion against the Qing established its headquarters across the Yellow River from Yinchuan in the 1860s. Tens of thousands of people died in the empire's campaign of suppression. By the 1990s, the Muslims of this mosque, at least, had established their presence as legitimate, their religious institutions as an ordinary part of the landscape, and their voice as one of moderation and negotiation.

Confrontational Hui: Various Locales, Various Solutions

If people in pre-modern China knew anything at all about the Chinese-speaking Muslims, they "knew" that they are violent people, prone to ganging up on hapless non-Muslims, to protection rackets and extortion, to feuding among themselves and to violent crime. In the Qing period, law cases and memorials from the northwest, from the southwest, and from the North China plain--all the parts of China proper in which Chinese-speaking Muslims were concentrated--described "those people" as fierce, brutal, and antisocial. As the story which begins this essay demonstrates, that stereotypical image has not faded from memory in some parts of the PRC. The Yunnanese policemen who opened fire on the Muslims in the police station courtyard thought they had much to fear from a trouble-making gang of Hui. The army has been called out on a number of occasions in the past decade, in a number of provinces, to deal with violence involving Hui. Many non-Muslims, especially those living near dense settlements of Hui, still fear for their lives and deplore the state's "soft" policies on minorities. One non-Muslim scholar in Beijing claimed that the Hui receive myriad special benefits because local government is so afraid of them, so concerned that "Hui violence" not reoccur.

Be that as it may, a close examination of a number of recent incidents reveals that "Hui violence" is not evenly distributed, is not simple rapacity or collective sociopathy, is not always directed at non-Hui. Some Hui do behave violently (so do some Han, of course, but their minzu identity is not blamed), but most do not. The Muslims of the North China plain (especially Henan), once famous for banditry and predation, have since the 1930s become much calmer. The Muslims of southern Shaanxi, feared for their solidarity and bloody-mindedness in the Wei River valley hinterland of Xi'an, were all killed or driven out during the "Muslim rebellions" of the mid-19th century. So the following cases come primarily from Yinchuan, Qinghai, Gansu, and Yunnan, frontier zones in which Muslims remain concentrated and their reputation as inherently antisocial people has persisted.

Ningxia. The Ningxia Hui Autonomous Region occupies barren plains and mountains through which runs a ribbon of green, the narrow plain watered by the Yellow River as it meanders between the Ordos desert and Liupan mountains on the east and the Helan mountains on the west. Near Yinchuan, one of China's oldest manmade irrigation systems enables farmers along the canals and rivers to grow paddy rice only a few hundred yards from the desert sands. Long an area of mixed religions and ethnicities, Ningxia currently has a population about 80% Han Chinese and 20% Hui, with a small number of Mongols in the west of the province. The irrigated areas (mostly to the north) have prospered in agriculture, industry (including coal mining), and trade, while the southern sections, mountainous and arid, have been the site of incredible poverty. Though there are Hui all over the province, for the past few decades "Hui violence" has been isolated in the destitute, water-starved south.

In 1992 a conflict erupted within the Banqiao branch of the Jahriyya Sufi order over leadership of the branch. Followers of the two candidates in southern Ningxia, the impoverished and mountainous part, began to feud over the leadership. Knives were drawn, guns brought into play, and pitched battles fought in and around the market towns and mosques. The violence escalated so fast that over fifty people had been killed before the army arrived. One informant learned through the rumor mill that at least one side had used homemade artillery during the affray. All of the dead were Hui, and none of the participants had made any overt moves against the state. The army separated the two sides without firing a shot, and the work of pacification, judgment, and reconciliation began with the arrival of a provincial work-team made up of 1,500 (yes, one thousand five hundred) officials, the majority of them Muslim minzu ganbu.

The job took eighteen months. By the end, one of the contenders for the leadership of the Jahriyya and some of his followers were doing time in a provincial prison, but they were not executed or locked away for life, as has often been the case with Chinese citizens involved in violent incidents. The work team members were satisfied that they had dealt with the local antisocial elements, but they came away appalled by the poverty and deprivation they had seen. Within two years, plans had been made to alleviate those conditions, and by the year 2000 road-building projects, newly dug wells, and agricultural extension stations had been located throughout the three counties of southern Ningxia. Central and provincial governments have both contributed funds to the effort; Muslim institutions and private individuals do so as well. One wealthy Yinchuan informant personally supports four children in southern Ningxia, sending them a monthly contribution for nutrition, schoolbooks, and other expenses through his mosque's charitable fund.

Yinchuan city, in the northern part of the province, presents a strong contrast to this conflict. With a large Muslim population and a long history of internecine feuding among Sufi orders, traditionalists, and fundamentalist solidarities, the city could be a hotbed of violence. But informants ranging from imams to professors to shopkeepers told me that the problems had been solved (like the early-morning loudspeakers in the western suburb), that there had been no violence there for a long time. They attributed this rapid improvement to one of two sources: either the minzu ganbu (especially the Yixie) had done their propaganda work well, mediated among the potentially feuding communities, and fostered an atmosphere of mutual respect; OR that same atmosphere had been created by effective cooperation and negotiation among the ahong of the various mosques, the leaders of the competing Muslim solidarities. All informants agreed that members of one mosque can attend services at the others, that ahong routinely give sermons to one another's congregations regardless of solidarity membership, and that members of one congregation are often invited to others for charitable feasts, funerals, and festivals. As I stood in the courtyard of the Xiguan mosque, I saw several of the mosque officials, including an elderly ahong, riding off on their bicycles and was told that they were going to a funeral at another nearby mosque. That simple sight constitutes strong evidence for local variation among Hui communities when compared both to Yinchuan's own past and to the violence which tore apart southern Ningxia in 1992.

Xining. As mentioned above, the most eminent ahong in Xining, the provincial capital of Qinghai province--which contains the northeastern corner of the Tibetan cultural region, where it abuts the Chinese, Mongolian, and Turkic cultural regions--spent over an hour telling me that everything there is fine. Tibetans and Muslims never quarrel, Muslim groups all get along, everyone shares mosque space for festivals, Han Chinese never insult Islam, and "unity" (Ch. tuanjie) characterizes all minzu relationships. Understanding his position as a public leader, in a delicate and closely watched frontier zone, I did not press him or present evidence to the contrary (which is plentiful) during our conversation.

The previous evening, a young shop clerk sat with me for over an hour and talked about "what happened back in 1993." That year (my informant was a teenager at the time), a newspaper in Sichuan published articles and advertising deemed insulting to Islam and Muslims, juxtaposing Islamic material and pictures of pigs. Protests spread all over the country, and Xining Muslims planned a demonstration. Unlike a similar situation in 1989, the Xining authorities in 1993 refused permission for a parade, but the Muslims marched anyway. Already defying the local representatives of the state and dissatisfied with their reaction, the Muslims decided to take their grievances to Lanzhou, hundreds of kilometers to the east, and then to Beijing, a place of immense power to which few if any of them had ever been. As they left town heading east, the army moved to block the bridges over which the road wound through the narrow mountain passes. Finding their way obstructed, some of the marchers decided to swim the river instead, and five of them drowned.

Clearly this was not "Muslim violence," though it certainly constituted disobedience to local authority. It did nourish an already-existing sense among some Xining Muslims that the army (and by extension the state) does not care about them, that they were willing to watch Muslims drown, that they would not allow Muslim voices to be heard. For non-Muslims, this incident offered further evidence that Muslims will misbehave if given the chance and confirmed their prejudice that Muslims will sacrifice themselves instantly and thoughtlessly for their religion or their ethnic group. Though these stereotypes are not limited to Xining--indeed, they are quite general in China--this particular incident caused only local upset, and it is known among ordinary Muslims elsewhere in the country only as a distant report.

Gansu. Gladney has described recent brawls between members of competing Muslim solidarities in Linxia, a prefectural city in southern Gansu province. It is located on the east side of a mountain range, west of which the population is largely Tibetan, while to its east, beyond the Tao river, lie predominantly Han counties. In short, Linxia represents the cultural frontier, a riverside market town where Chinese and Tibetans met, their economic interactions mediated, to some extent, by Chinese-speaking Muslims. Linxia's large population of "minority nationalities" (mostly Muslim) qualifies it to be an autonomous prefecture, and its minzu and religious affairs offices have been constantly busy settling or controlling disputes of all kinds.

To this potentially volatile mix of ethnic difference and frontier location the modern world has, since the late 19th century, added a new ingredient--illicit drugs. Eastern Gansu produced a substantial portion of China's domestic opium crop in the late Qing and Republican periods, but the plant was largely eliminated during the anti-drug campaigns of the 1950s. The state's proscriptions remained effective through the Cultural Revolution period, but since 1978 the booming commercial markets, lax enforcement, and ubiquitous government corruption of the PRC have reintroduced the drug trade to China, including the Muslims of the northwest. According to a recent news story, many Linxia Muslims are engaged in smuggling opiates, including heroin, from Yunnan, near the border of the Golden Triangle, with destinations as far away as Xinjiang, where some dealers also obtain supplies of opiates from Afghanistan via the former Central Asian republics of the Soviet Union.

In either case, already-established transportation networks staffed by Muslims would be available for moving the drugs from the frontier towards markets in China proper, which local people call "the interior" (Ch. neidi). This repeats a common pattern from earlier in the century, when Muslim farmers grew poppies and Muslim merchants moved raw or processed opium to market. But now the Muslims are also users. The same article reports an alarming rise in local consumption, with hundreds of young Linxia Muslims already addicted. In Linxia as elsewhere in the world, the combination of rapidly acquired wealth and drug use has enhanced the frontier's propensity to tension and violence. In 1996, Linxia informants professed themselves afraid, nostalgically recalling the Cultural Revolution and its strict social controls as halcyon days of peace and social stability.

A hundred kilometers southeast of Linxia, across a stretch of grassland above 8,000 feet in altitude, lies Lintan Jiucheng, mentioned above as a center for Muslims engaged in the trade between cultural China and cultural Tibet. The Xidaotang, a Muslim collective with over 10,000 members, has had its headquarters there for almost a century. Valuing commerce, and therefore social order, the leaders of the Xidaotang have taken a conciliatory stand toward government regulation and the ethnocultural values of their suppliers and customers, preferring to deal peacefully between Tibetans and Chinese.

The Xidaotang has also mediated within the world of Islam in China. Its founder, Ma Qixi (1857-1914), studied the Muslim tradition in Central Asia and was thoroughly conversant with the Arabic canon, but he also received a classical Chinese education. Once he became an ahong and teacher, that combination led him to focus his curriculum on the Islamic texts in Chinese, the Han kitab, which had been produced during the Qing period. Without watering down their Islamic orthodoxy or orthopraxy, the Hui of the Xidaotang have built China-wide production and distribution networks, including numerous retail shops using their commercial name, Tianxinglong. As mentioned above, they teach Tibetan in their community's schools, and informants within the solidarity confirm a relatively respectful approach toward their Tibetan clientele, an attitude not shared by most Chinese. In part because of their powerful local influence in ethnic relations, Lintan Jiucheng has remained relatively calm for the past half century, and there have been few incidents of communal violence. Though drugs may be smuggled there--the town lies on a major trade route between Yunnan and the northwest--none of my informants mentioned addiction as a social problem.

Yunnan. The ethnic and historical complexities of Gansu, Qinghai, and Ningxia are more than matched by those of Yunnan. China's southwesternmost province, bordering not only several southeast Asian countries but also a frontier of the Tibetan cultural region, has been the site of sanguinary confrontations, massacres, displacements, forced migrations, and diverse interminglings of peoples and cultures since it was incorporated into a China-based empire by the Mongols over seven hundred years ago. Muslims came to Yunnan in large numbers just at that time, soldiers and officials in an army commanded by Sayyid Ajall Shams ad-Din (Ch. Sai Dianchi), one of Qubilai's most successful generals. They stayed, even after the end of the Mongol empire's rule over China, and they have been an ordinary part of the social landscape ever since. Their relationship with the state reached a low point in the mid-19th century, when Du Wenxiu (1827-1872), a local Muslim, proclaimed Dali (western Yunnan) the capital of "the state which pacifies the south" (Ch. Pingnan Guo), fought off the Qing's armies for years then perished after a lengthy siege. Scholars debate whether Pingnan Guo was a "Muslim state" or not, but many of Du Wenxiu's core followers were Muslims, and the Qing armies (one of them led by a Muslim) did kill thousands of Muslims while pacifying the rebellion.

Some parts of Yunnan, especially in the west and south, were more engaged in Pingnan Guo than others, despite a province-wide distribution of Muslims. The Muslims of the poverty-stricken northeastern region (Zhaotong) pride themselves more on their lineage genealogies than they do on their adherence to a 19th c. Islamic state. Several informants, including university-based scholars, emphasized that the patchwork, somewhat disconnected quality of the Muslim worlds of Yunnan continues to exist today. A linguist told me that there are at least four distinct dialect zones within the Yunnanese form of Chinese, and that Muslims are residents of all four. My own field experience bore out the contention that very few generalizations can accurately describe the Hui in the region, apart from the obvious unities of religion and mosque-based community structure.

Apart from the violent incident described at the beginning of this essay, one potent moment stands out in recent history, as narrated by local Muslims. Everyone in Yunnan (and every Muslim in China) knows that the state destroyed the village of Shadian, a sizable south Yunnan Muslim community of several thousand souls, toward the end of the Cultural Revolution. I could not go to Shadian, but several informants in Kunming and Yuxi were willing to tell me what happened, and their various accounts contain the truth that "everyone knows." In the early 1970s, conformity to Han Chinese standards and values constituted the most important tenet of minzu policy in China. Anything perceived as "different" was maligned as counterrevolutionary and squelched. All over China, mosques became sites of bullying, sacrilege, even murder and suicide as Muslims were forced to raise pigs in the courtyards, and even aged ahong were forced to violate their own religious beliefs by eating pork.

Shadian was no exception. The state, through its local officials, imposed a production quota of pork on the village, and few were willing to resist, since the policy was well known and closely linked to patriotism and love of Chairman Mao. In addition, disobedience would have brought severe penalties. But then local Han officials came to the village and began to browbeat the Muslims. Old ahong were made to crawl on the ground and make pig sounds, and local policy demanded that Muslims not only produce but also consume pork in order to be "good Chinese revolutionaries." A secular Muslim leader, living in Kunming, wrote editorials in the provincial newspaper criticizing and decrying Islamic pork avoidance as superstition and feudal practice.

Local Han cadres, eager to "help" the Muslims become good Chinese by overcoming their hidebound and antisocial resistance to pork, decided to place pig bones in the wells in order to accustom the Hui to the taste. Discovering the pollution, the Muslims refused to drink the water and determined to resist what they perceived as murderous, sacrilegious behavior from their Han neighbors and government officials. Shadian, like many Yunnanese Muslim communities, had a long tradition of metallurgy, blacksmithing, and weapons manufacture. Over the next two years, as the county mustered an all-Han militia to oversee the "counter-revolutionary" village, the Muslim men made new guns, which included sophisticated automatic weapons, and organized a Muslim paramilitary of their own. Battles between the two armed groups left a number of men dead. The authorities, not understanding the depth of the Muslims' determination, sent ordinary policemen to deal with the problem, and the Muslims killed them.

The People's Liberation Army (PLA), summoned from provincial frontiers and distant bases, surrounded the entrenched farmers and small entrepreneurs of Shadian, who had only small arms and no artillery. After negotiations failed to persuade the Muslims to allow the PLA into the village, the army opened fire and razed the village entirely. None of my informants knew how many people had died in the assault, but one said that the only survivors were those who had dug shelters beneath their homes. According to Gladney, the army destroyed the village as a warning to others, but after the fall of the Gang of Four in 1976, the state rebuilt Shadian quickly--one informant thought this was done in hopes of eradicating the memory of the army's brutal suppression. The county and provincial governments provided special grants once reconstruction began, so Shadian developed more quickly than most Yunnan villages because of the incident, and today it prospers in trade and agricultural production.

Closer to Kunming, up in the central Yunnan hill country, the Muslim village of Najiaying also had a tense confrontation with the state, but with a very different denouement. Like Shadian famous for gun making, Najiaying's Muslims certainly could have decided to take on the forces of the state. When Shadian "rebelled," the ahong of Najiaying organized the men to withstand a siege and prepared the community for armed confrontation. But my informant, among other village leaders, talked the men out of suicidal resistance and, when the army arrived in force, surrendered the village without firing a shot. This decision strongly resembled that of Najiaying’s ancestors back in the days of Du Wenxiu and Pingnan Guo. Ma Rulong, a Muslim general serving the Qing, came to their broad valley to exterminate the Jahriyya Sufis at nearby Donggouzhai, but Najiaying surrendered without loss, except for a few men (including my informant's great-grandfather) who went to Donggouzhai to share their coreligionists' resistance and fate. After both surrenders, the village was able to participate in the recovery of the local economy and prospered. It currently has 7,000 indigenous residents, and another 7,000 "outside" workers, mostly from elsewhere in Yunnan, serve as the lower level labor force.

Shandong. The most recent local violence involving Hui occurred not in a remote frontier region but right in the Chinese heartland, Shandong province. In December, 2000, international news networks reported deadly clashes there between police, allied with armed non-Muslim paramilitary units, and crowds of Muslim protesters. No fewer than five Muslims were killed in Yangxin county, Shandong, and Muslims from all over eastern China were said to be in Yangxin to protect their coreligionists and to seek revenge.

The cause of these violent outbreaks lay in long-term misunderstandings between local Muslims and non-Muslims over what constitutes respectful coexistence. In late September, a non-Muslim street vendor, trying to take advantage of the Hui Muslims’ reputation for cleanliness and tasty food, had (in his ignorance) put up a sign advertising "Muslim Pork." Relying on local Hui sources, Agence France Press reported that Muslim leaders both staged public protests and attempted to petition local government for redress of this insult. (Lateline News claimed that outraged Muslims killed the vendor and another non-Muslim, but none of the other stories confirmed this.) Rather than negotiating the removal of the sign, county-level officials accused the Muslims of serious violations of the law (because of their protest gatherings) and arrested several leaders. In a sad replay of Qing and Republican period incidents, November saw Muslims and non-Muslims gradually polarized, and police clashed several times with unarmed Muslim demonstrators, whom they accused of illegal assembly.

Serious violence, however, did not occur until someone hung a pig’s head in front of a mosque on December 8 or 9. Several days later hundreds (perhaps thousands) of Muslims from Mengcun, a Hui stronghold in nearby Hebei province, set out on the road to join the protests in Yangxin. Stopped by armed police and militia at a roadblock, they refused to disperse or return home and (according to police) became rowdy. Fighting broke out, joined by unofficial Han bystanders. As frightened by a Hui mob as their colleagues in Yunnan had been in 1990, the police fired into the crowd of protesters, killing five and injuring as many as forty.

By December 18, a negotiation process had begun, reparations for the dead and wounded had been offered, and as many as 2,000 cadres had descended on the region to calm the violence. Lateline News reported a Hong Kong human rights spokesman as saying, "This thing is heating up all over the country. Hui people are very upset about this." But no national, regional, or even provincial outbreaks occurred--indeed, the December 18 story is the last news I have been able to find about the incident. Inquiring by e-mail of friends all over China in January, I have found that Muslims know about the violence, regret it, even resent it and openly blame the state for it. But none has any intention of going to Shandong, organizing a national protest, or undertaking any public action. Once again, we find local authorities dealing not with national ethnic conflicts but rather with local Hui problems, which they will try to solve without supra-local repercussions.


How can we explain the differences between rebellious Shadian or Donggouzhai and overtly compliant Najiaying? Between Shadian's fate in the 1970s and that of Yuxi two decades later? Why was the Zhengzhou Muslim quarter able to negotiate construction of xincun to salvage its neighborhood's location and structure, the Xi'an Huiminfang's efforts have thus far produced no results (though their neighborhood has not yet been razed or "renewed"), and Yinchuan’s Xiguan mosque has renewed itself and its neighborhood? It seems to me that all of the stories in this essay can only be understood through the localness of Hui communities, despite their common Muslim religion and (state-defined) minzu identity. Chinese scholars unify all of these phenomena in a conceptual universe dominated by the notion of minzu. Indeed, they posit two simultaneous interlocking processes--ethnicization (minzuhua) and localization (diquhua)--as responsible for the formation of the Huizu within the Chinese cultural matrix.

Without accepting the conventional timing of these two processes (for Chinese scholars place the former in the Ming period rather than the 20th century), we may understand their importance to the formation and maintenance of Hui identity in contemporary China. Hui intellectuals emphasize the national quality of Huiness, its shaoshu minzu core, while many other Hui stress the local in discussing who they are. Religious leaders and pious individuals, of course, place greatest importance on Islamic religion as a unifying valence of identity, but they also recognize its limits. Despite the claim that "all Muslims under Heaven are one family," most Hui ahong do not connect themselves easily or comfortably with separatist Turkic-speaking Muslims in Xinjiang and their sociopolitical ambitions. After all, the vast majority of Hui, even some of those who have traveled extensively in the Middle East, are clearly Chinese in their language, material culture, and textual lives outside the mosque. However much they might identify with Muslims elsewhere--even unto donning Arab clothing and headgear for photo opportunities--Hui are not members of Malay or Turkish or Persian or Arab or any other "Muslim" culture in which Islam is a "natural" component of identity. On the contrary, they must distinguish themselves constantly from the overwhelming majority of Chinese-speakers, who are not Muslims, while still remaining part of the only culture and polity in which their identity makes sense--that of China.

Seen in that light, this study of the Hui suggests some important cautions to foreign observers interested in China's "minority nationalities." First, "the Hui" do not exist as a unified, self-conscious, organized entity. Some would argue that no ethnic group conforms to these criteria, but our commonsensical notion of "the Tibetans" or "the Uygurs," discussed in endless newspaper articles and web postings, indicates that many of us believe that they do, or should. The Hui have national leaders, but they are all empowered and thus, to some subjective extent, delegitimized by their intimate association with the state--through the Yixie, Minwei, universities, and other government-approved organizations. If the above analysis is correct, then the separatist Eastern Turkestan movement in Xinjiang, the Republic of Mongolia, and the Dalai Lama's leadership of a substantial portion of Tibetans--all headquartered outside of China--set a model for minzu identity which the Hui (and, I would suggest, at least some other minzu) do not, indeed cannot, follow.

Second, some Hui communities are more difficult, sensitive, volatile, and potentially violent than others. This could be due to historical memory of confrontation and desire for revenge, to bellicose or inflexible Muslim leadership, to local geographical or economic conditions which militate against harmony with non-Muslim neighbors and/or the state, to insensitive or downright discriminatory policy or behavior from functionaries at several levels of government. We have seen negotiation between Muslim leaders and state authorities succeed in Zhengzhou, prevent confrontation in Yinchuan, allow Hui survival in Najiaying, while Yuxi, Shadian, and southern Ningxia exploded in violence. If communities as similar and geographically proximate as Shadian and Najiaying could have such different histories, how much more disparate must community histories be in Gansu, Henan, Beijing, or elsewhere?

Third, we cannot ignore the power of minzu policy and its underlying vision of "the minorities" (including the Huizu) as primitive peoples who require the leadership of the advanced Han minzu in order to advance toward the light of modernity. This mixture of condescension and fear toward non-Chinese people has much power in Han society There can be no question that some Hui resent this attitude and its attendant policies. But others do not, or at least mute their enmity with acknowledgement of Hui achievements and successes, in both the past and the present. The oft-heard contemporary claim that "We Hui can always defeat the Han in business; they are afraid of us," echoes edgy old Han proverbial knowledge--"Ten Huihui, nine thieves," or "Ten Beijing slippery characters cannot defeat a single Huihui." Though this persistent ethnocentrism will always produce small-scale confrontation, even rage and violence, there are no Hui leaders or organizations calling upon all Hui, all over China, to reject the authority of the current system in favor of Hui hegemony or emigration. In this the Hui of China strongly resemble the Muslims of India, who persist in their homeland despite constant tension and occasional open ruptures with a majority society which, to some extent, denies the validity of their sense of belonging and brands them as dangerous and foreign. But the Hui have no Pakistan, no Bangladesh to which they can turn as a "more authentic" homeland, and they constitute an incomparably smaller percentage of the general population.

Finally, as far as most Hui are concerned, no separatist movements or Islamic fundamentalism should undermine the unity of China as a nation-state. We may conclude from the above arguments that the Hui can only be Hui in China, however orthodox or orthopractic they may be in their Islamic lives. Even if increasing international communication raises the consciousness of Middle Eastern issues and Islamic identity among the Hui, this will not result in more than superficial Arabization and calls for "authentic" religion. The small communities of Hui living outside of China--in Turkey, for example, or Los Angeles--have not attempted to set up governments in exile but qingzhen/halal Chinese restaurants, conforming to the pattern of other Chinese emigrants in those parts of the world. Thus, despite the Hui being defined as a "minority nationality," U.S. policy makers should nonetheless regard the Hui as unequivocally Chinese, though sometimes marginal or even despised Chinese. Some among them, especially young and militant ahong, might claim that the unity of the Islamic umma overrides national (Chinese) identity, but this contention cannot be shared by most Hui. Like African Americans or French Jews, the majority of Hui participate as patriotic citizens in the political and cultural life of their homeland, even when antagonistic elements in the society or state challenge their authenticity or loyalty. On the basis of this analysis of the Hui, we cannot predict a centrifugal collapse of the People's Republic of China, in the manner of the former USSR. For now, at least, the cultural and linguistic forces of cohesion and Chinese nationalism currently outweigh the potentially divisive, dehomogenizing capabilities of religion, minority identity, and anti-minority discrimination for the Chinese-speaking Muslims.

Glossary of Terms



chantou Hui

feifa zuzhi

fengsu xiguan


Han Hui

Han kitab







Huizu (Huihui minzu)







minzu ganbu

minzu lun

minzu shibie

minzu zhengce


nü ahong


Pingnan Guo


Sala Hui

shaoshu minzu

Xi'an Shi Yisilan Wenhua Yanjiuhui






Zhong-A xuexiao

Zhongguo Yisilanjiao Xiehui (Yixie)

Zhonghua minzu

Zhongyang Minzu Shiwu Weiyuanhui (Minwei)


zongjiao shiwu ju

zuo Huihui


Allahu akbar




Ikhwan al-Muslimun