Introduction
Purposes and Form of a Muslim History in China
THE NEED FOR HISTORIES
Muslims have lived in China since the eighth century, but no comprehensive account of their 1,300-year history has ever been written in a European or Middle Eastern language. From the vantage of either Chinese or Islamic studies, Euro-Americans lack even an intellectual context for a focused treatment of this important cultural encounter and the millions of individuals who have participated in it. With a few notable exceptions, modern Chinese scholars have restricted themselves to narrow monographs or overblown theoretical models, and the most thorough Japanese book on the subject ends its narrative in the early nineteenth century.
Given the copious primary sources that deal with the Muslims of China, we would expect a larger body of scholarship from foreign academics. The presence of so many Muslims in China has stimulated occasional panic and consistent concern among China watchers, especially missionaries and players of the Great Game of Russian-British imperial rivalry in Central and East Asia. But most contemporary Euro-American scholars of Chinese history, not to mention their students, would be hard-pressed to say much of anything about the history of China’s Muslim residents, except perhaps that they have been there a long time and are counted among the “minority nationalities” (shaoshu minzu) by the government of the People’s Republic of China. But the stories of how they came to be there, how their lives and those of their ancestors might have differed from those of their non-Muslim neighbors, how they differ from one another and from other Muslims, and how their presence might have affected Chinese history do not appear in conventional histories of China, except when Muslims become violent. Islamicists know even less about Sino-Muslims, for mention of this far-flung margin of the Muslim world is generally confined to a few pages of exotica drawn from the thin secondary sources.
This lack of concern has not been exhibited toward the entrance, acculturation, and continued presence of the other three “world religions” in the Chinese cultural area. Despite the difficulty of the sources, the strategies, successes, and failures of Buddhism, Christianity, and Judaism in China have been extensively analyzed, including their sometimes degraded, sometimes exotic status as foreign religions and their adherents’ identification as barbarians, or at least outsiders, who gradually became ordinary and normal—that is, Chinese. The linguistic and intellectual obstacles standing in the way of the study of Islam in China are similar to those facing scholars of other “foreign” religions there, but an additional difficulty lies in the entrenched notion that Muslims are everywhere the same—that Islam, more than the other world religions, demands a strict uniformity of its believers. We can comfortably describe Chinese Buddhists, cope with the cultural complexity of Chinese Christians, and even consider Chinese Jews as both religiously Jewish and culturally Chinese. But not Muslims—they must either be fanatical followers of the Prophet, and thus not Chinese at all, or entirely acculturated, and therefore not Muslim enough to be of significance. This complex of attitudes does not conform to the history of religions in general, certainly not of Islam in particular. For religion alone cannot ever determine how people behave in specific times and places; many other valences of identity constantly play in individual and collective decisions.
In addition to its relevance to the history of both Islam and China, the study of Muslims in China should be included as an element in the much larger study of frontiers both cultural and physical, of cultural contact and syncretism, and of multicultural societies. All over the world a vast variety of hyphenated or multiple cultural identities has been established over the course of millennia.1 For many reasons—trade and long-term sojourning, migration away from military threats or toward natural resources, the expansion and contraction of kingdoms and empires, among others—people moved, acculturated to new human and physical ecologies, and became normal in new places. Some claim indigenous status after only a few generations (e.g., Afrikaaners, Anglo-Americans), while others keep (and/or are kept) separate and remain self-consciously Other after centuries (e.g., the Jews of Europe, the Hoa of Vietnam). The immigrants, the target culture, and the particular temporal moment all figure in the complex processes of mutual adaptation and coexistence.
In the past two centuries the nation-state as humankind’s primary form of large-scale social, cultural, and political integration has become a crucial actor in the dramas of acculturation—those already in progress and the new ones created by colonialism, imperialism, and the break-up or consolidation of empires into “countries.” Modern nation-states fix their borders, including under their sovereign power whoever happens to live in the territory thus enclosed. In most cases, that includes large numbers of people culturally different from those in control. Nation-states also possess the almost unconstrained power to grant citizenship, and thus full humanity, to their residents2 and to count, classify, divide, and otherwise control crucial elements of personal identity.3
Though not utterly new in quality, these capacities of the modern nation-state do differ significantly from those of premodern states. The Qing (1644–1912) empire before the nineteenth century, for example, possessed an ethnological capacity, counting and ordering its subjects on the basis of language and other cultural variables, but it rarely went so far as to tell them who they were.4 When the Qing emperors established and tried to enforce boundaries between their subject peoples, they did so to prevent combinations against themselves and to maintain the normative model of a Manchu center ruling over many distinct subordinate lords.5 Following China’s enforced inclusion in the world system of nation-states, the nation-state centered on cultural China—first in its embryonic Qing and Republican forms and now as the People’s Republic—has created categories of humanity and superimposed them over existing social reality, to which they conform in wildly varying degrees. The People’s Republic of China now employs the awesome apparatus of modern political and social technology to penetrate into local society far more effectively than its imperial predecessors could. In this (to date) century-long process, the Chinese nation-state has inherited, reimagined, and acted upon a modern, hegemonic paradigm of “Chinese” society, one based on the powerful concept of minzu.
THE MINZU PARADIGM
The People’s Republic of China divides its citizens into fifty-six minzu, a word of late nineteenth-century Japanese origin with no obvious English equivalent. It probably originated as a translation of the German das Volk and is now variously rendered as “ethnic group,” “nationality,” “(a) people,” and “nation.” Qian Mu, in Dennerline’s English, thought of it as the “whole people’s descent group,” and believed that its existence and power grew out of the Chinese lineage and the universality of the rituals and norms of propriety.6 The minzu was conceptualized in late nineteenth- and early twentieth-century China as a powerful, invisible cement binding together the descendants of the Yellow Emperor, a definition that clearly illustrates its genealogical connection to late Edo and Meiji period (eighteenth–nineteenth century) Japanese nativism and more distantly to German Aryanism and other European racialist theories, as well as to indigenous Chinese discourses of race.7 In the twentieth century, minzu, and various other compound words using the zu component to claim familial (i.e., genetic) descent, have been critical elements of the re-creation of Chineseness, deeply affected by evolutionism (especially Neo-Lamarckism) and the racial-eugenic theories expounded by Euro-American scientists.
The People’s Republic of China, borrowing from the Stalinist “nationality” policies of the Soviet Union, has transformed the term into a bureaucratic classificatory tool. Since the 1950s it has been part of the governing project of the People’s Republic to “identify” China’s minzu, to classify and count the people within its borders as members of these minzu, and to educate and provide services and policies for them appropriate to their minzu. The state also undertakes verification and reification of the primordial quality of minzu membership and identity through institutions and policies ranging from ethnological and linguis tic research institutes to collections of minzu folktales to special schools for minzu children.
Justifying this immense enterprise, now involving over a billion souls, required the creation of a hegemonic narrative, a unified story that could demonstrate the bedrock truth of minzu continuity and consanguinity in the past, for the present. That narrative rests primarily on the teleological imperatives of Lewis Morgan’s five-step journey from primitive matriarchy to socialist civilization, but, like all such national stories, it also embodies an ideology of domination, the superiority of Us over Them. In this account, each minzu, at its own pace and according to its own environmental and historical conditions, has followed the most advanced minzu, the majority Han people, toward higher steps on the ladder of history. The Han minzu, which is supposed to include the vast majority of China’s citizens—most (but not all) of the people we would call “culturally Chinese”—has also been subjected to this most Procrustean of narratives, but not in the same way as the “less advanced” peoples. For Han—that is, Chinese—history, unlike other minzu histories, constitutes the story of Civilization or Culture itself and thus represents the Chinese version of History, the linear and rigidly structured narrative of progress that philosophers of the European Enlightenment imagined would happen to everyone, sooner or later.8
The “minority nationalities,” formally the shaoshu minzu but often called simply minzu to distinguish them from the Han, have been placed in their “proper” historical positions by the construction of their individual narratives, published as jianshi, “simple histories.” These books all tell more or less the same story, embellished with local detail and ethnological descriptions. From the Mosuo “living fossils” of Yunnan (who cling to archaic practices such as sex outside marriage) to the pastoral Mongols (many of whose families have been farmers for centuries) to the remnants of the Manchus (some of whom have had to be convinced to be Manchus), the various minzu move from primitive to slave to feudal to capitalist to socialist modes of production, progressing most effectively when they acculturate to Chinese ways and learn from the progressive classes of Han society.9 The universal promulgation of these narratives, and the huge master narrative of the Han on which they are modeled, has generated contemporary Chinese History, a story that can be amended or decorated but never questioned. These histories constitute a crucial part of the “civilizing project” of the People’s Republic, which builds upon, expands, and transforms the efforts of earlier political centers—Mongol Yuan, Chinese Ming, Manchu Qing, Chinese Republican—to express their dominance over peripheral peoples. Now a centralized nation-state, twentieth-century China has reified the shape and size of the Qing empire at its greatest extent, minus much of Mongolia and some of Vietnam, as what China has always been. Historical figures as varied as Chinggis Khan, the Dalai Lamas, and the Kangxi emperor and the vast territories of Tibet, Turkestan, and Mongolia may thus be unambiguously identified as Chinese.
The problem of hegemonic narratives and the category systems on which they are based—that is, the objectification of the Other—lies close to the heart of this book. Euro-American scholars generally have accepted description of the “non-Chinese” people of China—as is done unproblematically in the People’s Republic—as members of fifty-five clearly distinguishable minzu. But upon even cursory examination, the supposedly exclusive minzu categories break down, become muddled, invite deconstruction.10 Consider some questions germane to this book: How is “Muslim” defined in the People’s Republic? What is the relevance of this religious category to the minzu paradigm, which eschews religion as a determinant of minzu status? Are people Muslims by heredity, even if they do not practice Islam or believe in its tenets? How much intermarriage outside one’s own minzu is required before ethnic identity shifts? These questions reiterate the difficulties Euro-American states have had with the categories “Jew,” “Black,” “Indian,” and many other putatively genetic boxes, underlining the rigidity and inflexibility of state-established categories as hegemonic devices, not simple descriptors of ethnological reality.
The People’s Republic of China, claiming to divorce its minzu paradigm from religion, divides its “Muslim” citizens into ten minzu, each supposedly distinguished by common territory, language, economy, and psychological nature.11 Of these ten, nine are held to occupy their own ancestral land and speak their own languages, though several are now predominantly Sinophone. The tenth, the Hui, whose members make up almost half of China’s Muslims (and ex-Muslims), constitutes the default category. Under the Ming, Qing, and Republican regimes, the word Hui meant “Muslim,” and Islam was called the “teaching of the Hui” (Ch. Hui jiao). Muslims were distinguished from one another by additional ethnonyms: the Turkic-speakers of the Xunhua region were called Sala Hui, the turban-wearing residents of the eastern Turkestan oasis cities were called Chantou Hui, and Chinese-speaking Muslims were called Han Hui, among other names. Since the 1950s, however, only a Muslim or descendant of Muslims who lives in China but does not belong to one of the nine linguistically or territorially defined Muslim minzu is a Hui.12 Most of the Muslim actors in this book would now be considered to belong to the Hui minzu—that is, they were Chinese-speaking Muslims—but they would not have used that name themselves.
The most common ethnogenetic account of the Hui minzu, found in the Huizu jianshi among many other sources, claims that during the Ming period (1368–1644) the Muslims of China became a minzu, despite their lacking at least three of the four defining characteristics of such an entity.13 This Hui minzu is characterized by common descent from the foreign Muslims of the Tang (618–907) to the Yuan (1279–1368) period, a wide geographical distribution in China, and exclusive use of the local vernacular, usually but not always a form of Chinese, outside of ritual life. No such ethnonym exists in the Ming sources—there all Muslims are called Huihui, an erroneous generalization of the earlier Huihe, meaning Uygur.14 Hui, a shortened version of that same word, was used to mean “Muslim” in Ming and Qing texts, a meaning changed decisively by twentieth-century governments intent on establishing minzu (contrasted to religion) as the crucial valence of ethnic identity in modern China.15
The People’s Republic of China has been remarkably successful in imposing the language of the minzu paradigm on its entire population, including scholars and intellectuals of the “minority nationalities” themselves, so as a foreign historian I find myself in the position of disagreeing at the fundamental level of vocabulary—the meaning of words—with teachers, colleagues, and most of the Chinese secondary literature on this subject.16 The reader should certainly be aware of these differences and judge the arguments in this book not on their conformity to a familiar vocabulary but on their historical merits. My subjects are predominantly Chinese-speaking Muslims, but I shall not call them Hui unless I am referring to the period of the People’s Republic. Because the word Hui is now entirely subsumed in the Hui minzu, for historical narrative I prefer the categorical term Sino-Muslim, which combines Chinese linguistic and material culture and Islamic religion without relying on an anachronistic category scheme that would lump them together genetically with Tibetan-, Tai-, and Bai-speakers, among others. To go further, I find the entire minzu paradigm, with its putative antiquity of ethnic consciousness and common descent, to be highly suspect in regard to the Sino-Muslims. One of this book’s purposes lies in examining what actually happened in some parts of China in order to test the now politically enshrined minzu version of Chinese and Sino-Muslim history.
For the same reasons, I shall attempt to use ethnonyms contemporary to the sources of my narrative for non-Chinese-speaking Muslims, so Salar will appear in the Qing, and Dongxiang only in the twentieth century. Perhaps most annoying to contemporary readers, I shall try to avoid the word Han as well, preferring “Chinese,” or, in some cases, “non-Muslim Chinese,” as a more neutral marker. Though it certainly appears in pre-twentieth-century sources, “Han” did not mean what it does now, and historical accuracy demands that we understand what it meant then. As Almaz Khan has persuasively argued for the Mongols, Pamela Crossley more cautiously for the Manchus, and Frank Dikötter for the Han Chinese, I shall demonstrate with regard to the Sino-Muslims that ethnic consciousness of the genealogical kind inherent in the minzu paradigm is largely a modern phenomenon, based in a hegemonic ideology that belongs to the nation-state, not to premodern empires.17
CATEGORY SYSTEMS
Clearly one of the difficulties of a Muslim history in China lies precisely in determining what the categories—the limits that states, cultures, and other hegemonic systems place on what meanings are possible by control over the lexicon and syntax of primary and secondary historical expression—were at a particular time. Since I am writing for an English-reading audience, I confine myself largely to the category systems available in this language, some of which do not exist in Chinese or the other languages in which my sources are written. Like minzu, words such as “religion” (Ch. zongjiao), “Muslim” (Ch. Musilin, or earlier Huihui and Huijiaotu), “ethnicity,” “China,” and “nation-state” (Ch. guojia) have no neutral or precise referents but rather represent pieces of historically constructed category systems within discourses of power.18 From the outset, then, I shall be clear about their meanings.
Religion
The differentness of various Muslims from normative Chinese definitions of themselves, from the idealized descriptions of “Han” culture and society, does not always lie in the four Stalinist criteria that qualify Muslims for minzu status. Indeed, the minzu paradigm, as applied in the People’s Republic of China, embodies only some of the category problems of a Muslim history in China. In this book, I have categorized and separated the various actors from one another primarily by creed, religious association, and the panoply of practices associated with Islam. “Muslim” represents a powerful valence of identity, engaging us with religious criteria invariably mediated by words and practices originating far from China.
Though scholars of religion can supply strict and objective criteria for membership in the umma, the universal Muslim community, like minzu definitions they begin to bend and become more malleable under the pressure of historical circumstance. In public behavior, not only in China but also in Muslim heartlands as well, many adult Muslim males, not to mention women, do not pray five times a day, some stint on their charitable obligations, and many do not fast during Ramadan, but they are nonetheless unambiguously Muslims (some would call them “bad Muslims”). We have even less information to test Muslim identity with regard to matters of the heart such as faith in God. From studying Muslims in China, I would argue that “being a Muslim” strongly resembles “being a Jew” or “being a Christian” in the vast variety of religious, psychological, social, political, and intellectual states it might describe.19 The sources for this book define “Muslim” in many ways, ranging from silent participation to religious conviction to genealogical descent, and readers should be aware that no single definition beyond self-ascription or community membership informs the text. Euro-Americans must be particularly cautious in ascribing specific collective consciousness or behavior to Muslims, for we are bound with special tightness by our own discourses of superiority and hegemony with regard to the Islamic world.
Sects, Orders, Teachings, and Solidarities
The Muslims of northwest China have divided themselves into a bewildering number of groups by affiliation with leaders, ideologies, and religious communities.20 The terminology describing these structures will be crucial to this historical narrative, for religious solidarity functioned as a central valence of identity in northwest China. Muslims also identified themselves with particular places, genders, age-cohorts, professions, and more, but the sources for northwestern Chinese history focus to a great extent on expression of loyalty to specific Muslim leaders, taken to embody the principles and character of their solidarities. I shall use the generic term “solidarity” to refer to all such groups, for they are, to a greater extent than many other components of identity, voluntary and self-consciously solidary.
The English word “sect” has in Islamic studies come to refer exclusively to groups that can consider other Muslims to be nonbelievers (Ar. kafir). In modern times the Sunni/Shi’i division is referred to as sectarian, while most others are not, and since that distinction is not relevant to this history of Islam in China, “sect” does not appear in this book with reference to Muslims. Islamicists use the word “order” to refer to Sufi brotherhoods, whose rise in China will be described in considerable detail. Chinese, of course, possesses its own terminologies and categories for such groups, and controversies have arisen over whether a particular group is a jiaopai (teaching), menhuan (Sufi group with hereditary leadership), or some other sort of pai (faction). Qing texts often refer to Laojiao (Old Teaching) and Xinjiao (New Teaching), while contemporary studies combine specifically Muslim terms with minzu categories, for example, Salazu de Zhehelinye (Jahrīya adherents among the Salar minzu). Because so many of the actors in this narrative are Muslim solidarities, usually personified in their leaders, I shall define them with care when they appear.
Ethnicity
The complex definitions surrounding the English term “ethnicity” continue to provide scholars with fertile ground for contumely. Rather than engage in the theoretical debate, I shall follow the definition proposed by a recent book: Members of an ethnic group share consciousness of solidarity by virtue of sharing (putative) common descent and common customs or habits, and they similarly share consciousness of opposition to other such groups of “different” ancestry and customs. However ethnic groups might arise (and there is considerable controversy on the issue), they do seem to develop (are invented or transformed) “in situations where a group is confronted in some way by an outside power with whom it is in competition for resources of some kind, whether they be material . . . or symbolic.”21 This definition allows for ethnicity to be processual, rather than fixed by a list of characteristics in anthropological time, and it denies the primordial, eternal qualities often ascribed to ethnic groups by their members and their enemies. This definition also does not conform to that of a minzu, for the latter is rendered ahistorical by construction of objectified markers—in official doctrine, these are common territory, language, economy, and psychological nature—rather than processes of consciousness or opposition, so I shall not use the two words interchangeably.
Empire and Nation
The state must be an important actor in any history of Muslims in China, and here, too, problems of definition arise, especially as we observe the transformation of the Qing empire into the Chinese nation-state in the nineteenth and twentieth centuries. In a recent book James Hevia notes some crucial differences between Manchu imperial hegemony and earlier (specifically Ming) indigenously ruled Chinese states, such as the “multinational, multilinguistic, and multiethnic” nature of the Qing polity and the consequent necessity for the Qing rulers to create a powerful center and an effective balance among their subordinate lords.22 We cannot place “the Muslims” in a single position in this state-centered model (nor did the Qing), for they lived in so wide a variety of cultural and political circumstances and in so many different relationships to the state. Some (e.g., the Turkic-speakers of Altishahr) were perceived as vassals, others (e.g., the Chinese-speaking Muslims of Gansu) as domestic subjects like the Chinese; some individuals received hostile attention from officials, while others achieved high rank. Transforming itself in historical time, and affected by internal and external forces beyond its control, the empire, like the personal or collective identities of its subjects, should be viewed as processual rather than fixed, and this requires special care in the construction of an apparently straightforward narrative.
Even Qing authority over the Muslims of northwest China, the central subjects of this book, cannot be described as monolithic or consistent. Some of the non-Chinese-speaking peoples, the Salar for example, had been governed by tusi, local families that received hereditary patents of office and a degree of autonomy in local affairs from the Qing state.23 After completing the conquest of eastern Turkestan in the 1750s, the Qing appointed local notables as hereditary lords (Tur. beg) over the urban Turkic-speakers of Xinjiang, but the Chinese-speaking Muslims of Shaanxi and Gansu remained entirely under the jurisdiction of the centrally appointed regular civil officials, though the military played a major role in local politics. The Qing perceived differences among the peripheral groups—in their ability or inability to use the Chinese language and in their historically demonstrated “governability”—and established local authority accordingly, altering its structure as local and regional conditions changed.
In local or regional history, we cannot simply examine central policy or imperial pronouncements and assume their implementation by “the state.” Rather, we ask, within local structures of dominance, “Who is the state?”24 The state’s formal and informal apparatus in northwest China over the past three centuries has included a fair number of Muslims, some of them conventional graduates of the military and civil examination system, others holders of less obvious (but no less real) state-sanctioned authority. The empire governed many of its peripheral subjects from a considerable political—as well as physical—distance, and this, too, distinguishes it from the more intrusive modern nation-state, with integration and participation on its mind.
SOURCES AND PERCEPTIONS
Our interpretive problems do not end with the meanings of words. To repeat a common truism, historical writing depends to a great extent on what people choose to remember, usually by writing it down but by other means as well, and this is determined by what they think is important. For a history of Muslims in northwest China, most of our primary sources come from officials, literati, gentry, and other people associated with the state. The northwest has had the lowest literacy rate of any part of the Chinese culture area, so few Muslims wrote Chinese well enough to compose their own histories. Those who did tended to work, like their literate non-Muslim contemporaries, in the vocabulary and from the point of view of the imperial center. They did not necessarily agree with one another, for Chinese historiography and intellectual history are rife with conflict over interpretations of the past, but all of them, more or less unconsciously, utilized culturewide notions of human nature, of historical causation, of the working of human society to inform not only their analyses but also their decisions on what to include when they wrote.
In Chinese writing about Muslims over the past three hundred years, one theme overrides all others—that of violence. During the Ming and Qing periods, it became axiomatic that Huihui, or (if the writer was careful) some Huihui (meaning Muslims) were by nature fierce, predatory, and hard to control.25 Like similar dominant stereotypes elsewhere in the world, and others rife in China itself, this image has a history; we can follow its formation in real time, as specific incidents combined with more diffuse perceptions of peripheral or “different” people to arouse fear and loathing.26
Particularly after the Qing conquest of eastern Turkestan—that is, by the mid-eighteenth century—the peculiar conditions of northwest China, pressuring the Sino-Muslims from both sides of their frontier homeland and on both aspects of their hyphenated identities, brought catastrophic change, which can be narrated in increasingly detailed and textured form. I am especially concerned with the weight of the stereotype of Muslims as violent people, which figured in state and local sociopolitical and military practice from the mid-Qing into the twentieth century. To narrate a history of violence does not mean to accept the image as true. Rather, as a historian, I weave the stories the sources tell into a history of perception that lay beyond the consciousness of their authors. Hindsight, after all, has to be good for something.
EXPLANATIONS OF SOCIAL DISORDER
From 1949 into the 1980s Chinese historiography has been dominated by the supposedly materialist analytical construct of class struggle, the sole motive force of change in the Maoist vision of the past. In contemporary Chinese History, the hegemonic narrative centering on the Han minzu, class struggle remains a primary explanation for social disorder (including violence): the masses who are the subjects of history and objects of state control rise up against feudal oppression at moments of sharpening social contradiction, with varying degrees of success. Within that rigidly enforced conception of human motivation, however, Chinese historians have creatively integrated older causal conceptions from imperial China’s Confucian discourse.
As the historian’s focus narrows to specific events and individuals, the hoary ethic of “praise and blame historiography,” canonized by the Han dynasty historian Sima Qian (145–87 B. C. E.?), appears alongside dialectical materialism, attaching judgments of “good” and “evil” to the crucial actors—emperors, rebels, officials, warlords, intellectuals, and more. Since Chinese nationalism (often permeated with Han minzu chauvinism) has come to dominate explanatory modes, the decision as to what is good and what is evil depends largely on whether the event or actor contributed to the good of the nation, its most progressive classes (the workers and peasants), and, in some cases, the minzu, as it is currently perceived. Historians therefore set out to judge whether Ma Hualong was a righteous rebel or a wicked bandit, whether Ma Hongbin was a virtuous warlord or a feudal warlord, whether Chiang Kai-shek was a reactionary villain or a national unifier. The objects of historical judgment include not only individuals but also collectives of various kinds, including Muslim solidarities, entire peoples or nation-states, residents of a particular place, foreigners (in general or divided into subgroups), and minzu. Such judgments have powerful impact on the construction of hegemonic narratives, while their praise-and-blame purpose ineluctably influences the ways Chinese historians read their sources.
The definition of “good,” and for whom the person or event in question might be good, lies at the core of all such judgments. Prior to the rise of the modern nation-state, the conventional nostrums of Confucianism set the standard—goodness lies in acting out benevolence, righteousness, justice, filial piety, female chastity, and so forth. For officials and other public figures, loyalty to the throne—to the person of the emperor or his family—stood as an important definer of the good, as long as the dynasty’s virtue could be seen to remain intact. The Mencian justification of rebellion against an unjust authority that had lost Heaven’s mandate might also condition the evaluation, with appropriate hindsight, of a historical actor. Like many historians of China, I shall have occasion to mention the proverb “A winner a king, a loser a bandit.” In the nineteenth and twentieth centuries the good of the nation-state, its leader, and the minzu has become the standard for historical judgment, joined since the rise of the Communist Party by that of the “progressive” or “virtuous” social classes and of the Party itself.
Under these judgmental, moralistic historiographical regimes—Confucian, nationalist, and communist—the behavior of Muslims in China has been evaluated. Not only did Muslims have to conform to the current standard of goodness, but they also suffered opprobrium from an intellectual elite that held its own texts and values to be exclusively true, thus condemning the Muslims’ texts (and the Muslims themselves) to permanent or temporary barbarism. Confucian scholars usually argued that the barbarian was civilizable, and we will note many cases of Muslims who achieved high rank or reputation by their (Chinese) classical learning or (Chinese) moral action. Foreigners or barbarians were simply those who had not yet been adequately exposed to Civilization, to Culture.
Dikötter, however, notes that another, more defensive set of categorical differences between Chinese and non-Chinese was available in premodern China, in contrast to this Confucian cultural universalism. The alternative view defined non-Chinese as utterly and irremediably Other, different, savage, bestial. Though literati commonly invoked those categories only under pressure from external elites (e.g., conquest dynasties), when the standards themselves were threatened by unCivilized power, they nonetheless provided a potential counterbalance to the ideology of acculturation and absorption of foreigners.27 They also enabled both officials and local gentry, in times of social strife, to call on wide-ranging stereotypes of Others (including Muslims) as evil and disorderly by nature, controllable only through force majeure applied by the state rather than by peaceful exposure to the civilizing influences of literary Culture (or its lower-class correlate, agriculture). In the People’s Republic, with its statutory equality among minzu, “barbarian” has been replaced by “primitive” or “backward” as a characterization of those minzu who must be persuaded or educated into conformity with the dominant cultural norms of the Han minzu, while irreversible savagery is reserved for imperialists, feudal regimes, some foreigners (especially Africans), and, occasionally, the Japanese.
SPECIFICITY AND MULTIPLE CAUSATION
Facing a comprehensive, hegemonic narrative of Chinese history based on inflexible imperial or Sinocentric standards of judgment, we should aim not only to deconstruct but also to provide alternatives that comprehend the ambiguity and multiple causations inherent in any human history. In the construction of parallel or alternative narratives, we can avoid the errors of universalism and overgeneralization that plague the dominant paradigms; we can read the sources in pursuit of local stories revealing the same humane richness of motivation we would seek in order to explain or understand events in our own times and places.
So I present this book not as the history of the Muslims of northwest China but as a history of Muslims in northwest China—one of many, as firmly grounded in specificity as the sources allow. My own proclivities and opportunities led me to the region called Gansu Province by the Qing but which now includes Gansu and Qinghai Provinces and the Ningxia Hui Autonomous Region. There the Sino-Muslims’ bicultural quality is sharply pronounced, for they occupy what Richard White, a historian of European relations with Native Americans, has called the “middle ground,” a place of intimate contact and fear and adaptation, a place in which peoples adjust to their differences while positioned between cultures.28 Though it may be dominated by one side or the other, the middle ground is always ambiguous ground, always capable of multiple interpretations. The middle ground of Gansu, like that of the Great Lakes, enabled a long process of sometimes expedient, sometimes deadly, mutual misunderstanding. Though I believe that Sino-Muslims inhabited similar middle grounds in Yunnan, Sichuan, and Turkestan, and elsewhere in China on a more local scale, those stories will have to await other investigations.
As I explain below, Gansu was the meeting ground of four cultural zones. In the past millennium the dominant cultural and political cores of Chinese, Tibetan, Mongolian-Manchu, and Central Asian civilizations all have lain far from northwest China, making it a frontier of four cultures. Under imperial rule from a capital in eastern China, the region could usually function through negotiation of the terms of relationship and a careful eye to independent powers that might occupy the “other” side of the frontier in Tibet, eastern Turkestan, or Mongolia.29
PROGRESS BY NARROWING
We cannot begin such a history in medias res, however. Even the hegemonic narrative of Sino-Muslim history is not generally known by Euro-American scholars, so the very presence of Muslims anywhere in China must be narrated and explained. This story includes the evolution of social attitudes and socially constructed descriptions of Muslims in China as a group (stereotypes), providing a general context for more local history. Therefore I progress by narrowing in both space and time, beginning with a brief and rather general history of the arrival and diffusion of Islam in what we now call China and moving toward regional, local, and finally individual stories. Driven by the historian’s inexorable logic of chronological time, my narrative cannot sustain the vast scope necessary to be the history of Muslims in China, nor can it amalgamate into accurate generalizations all the personal, local, and specific stories that represent the Sino-Muslims as human beings, as historical actors.
Following chapter 1’s geographic and ethnographic introduction to northwest China, the narrative begins in chapter 2 with an overview that covers almost a thousand years and encompasses the entire Chinese culture area, setting the stage by placing Muslims as unfamiliar elements in the well-known context of Tang through Ming China. As sources become richer and more complex after the Qing conquest of the 1640s, chapters 3 and 4 examine specific Sino-Muslim solidarities and their leaders, covering only a century each, and only in northwest China. When the Chinese nation-state replaced the Qing empire in the late nineteenth and early twentieth centuries, complex processes of change, caused by both endogenous forces and others from Euro-America and from the Muslim world, enveloped the peripheries of China and forced people to make choices under unfamiliar pressures. Chapter 5 therefore narrows further to individuals, presenting four portraits of Sino-Muslim leaders who took very different but equally winding paths toward accommodation with modern China.
In the Qing and Republican periods Muslims played an important role in constructing Chinese identity—as Gu Jiegang claimed they would for a modern China—in part by resisting homogenization, by subverting the dominant definitions of Chineseness, by remaining different but present, the “familiar strangers” of my title. The results of this history of tension were often brutal and violent, well reported in the self-consciously moralizing histories of the keepers of public order, who were also the defenders of the monolithic hierarchy of virtue with themselves at the top and distant, uncivilizable foreigners at the bottom. But some of the Sino-Muslims’ history took place within domestic settings that historians cannot reach, as Muslims and non-Muslims intermarried, or in marketplace encounters that ended in deals, not brawls, or in mutual and satisfactory adaptations resulting from centuries of ordered contact.
Over the past thousand and more years Muslims have come to belong in China, to think of it as their only home, to use its cultures and languages as their own. And Chinese culture, in many of its regional manifestations (especially, but not exclusively, in Gansu and Yunnan), has developed a “sense of itself” in part through constant interaction with and gradual inclusion of the Muslim strangers who lived there. Their myriad of stories cannot be told only in the aggregate, in terms of ethnic groups or regions. Rather, we meet individual human beings, follow their decisions and indecisions, try to understand their position as participants in what appear to be two cultures, as demanding and as exclusive and intolerant of domestic difference as cultures are prone to be. Complex choices about adaptation had always been available to Sino-Muslims, but the richness of our twentieth-century sources allows us to examine individuals as they faced the wrenching changes wrought by modernity.
THE VIEW FROM THE EDGE: FAMILIAR STRANGERS
The large processes examined here—acculturation, resistance, incorporation, integration—should not be allowed to obscure differences among places, times, and persons. In a book covering 1,300 years I have sacrificed a measure of specificity, especially in the early chapters, but I have tried to suggest in anecdotes the richness of sources that awaits us. Only a novelist’s or poet’s hand could describe adequately the extraordinary physical and human terrain of northwest China, but I hope to do it pale justice with maps, photographs, and narratives. This book is intended to create interest not only in further study of Sino-Muslims, but also in the broad issues of difference and conflict their stories embody. Most, if not all, of the contemporary world’s nation-states are coming to terms with domestic minorities—people who belong to more than one culture simultaneously, people who live in the “middle ground” and who have created new and syncretic cultures there. (Another recent book called Familiar Strangers deals with the Gypsies in the United States.)
It would be feckless to claim that Gansu somehow represents China as a “typical” region. Gansu lies at the frontier, at the margin, of Chinese culture (though not of the contemporary Chinese nation-state), and thus, despite its apparently outlandish mix of cultures, it can provide us with what has now become a clichéd advantage—the liminal view. The Sino-Muslims of northwest China indeed live in a dual liminality, at the outer edges of the two worlds with which they identify as Chinese and as Muslims. I have written Familiar Strangers as a history of becoming and then being Chinese while remaining Muslim, of the evolution of a sense of home. If our understandings of Chineseness cannot encompass the history of Gansu, then we should revise them. If our generalizations about Islam, about cultural contact, about acculturation and sociocultural distinction cannot help us explain the society of northwest China and its processes, we should change them.
1. For the creation of such combined identities among the Sino-Muslims, see Lipman, “Hyphenated Chinese.”
2. Rubenstein, Cunning of History.
3. Anderson, Imagined Communities, esp. chap. 9, “Census, Map, Museum.”
4. James Millward has found exceptions to this generalization at the northwestern edge of the Qing empire, law cases in which officials indicted Chinese-speaking Muslims (whom Millward calls Tungans) from Gansu who moved to Xinjiang as merchants. After acculturating in language and marrying local Turkic-speaking women, they cut off their queues. Were they Chinese, and therefore to be punished for defying the empire’s tonsorial strictures on Chinese people, or non-Chinese, and therefore free to wear their hair as they chose? Both men protested that they had been forced to cut off their queues by invaders from Kokand, but Nayancheng, the pacification commissioner, believed that they would not have hesitated to do so themselves, so he punished them with exile. In addition, he requested a judgment from the Board of Punishments regarding their marriages to Turkestani women and was permitted to beat them with the heavy bamboo, separate them from their wives, and reinforce their sentence of exile under an antimiscegenation statute added to the Altishahr regulations only after these cases were decided (Beyond the Pass, chap. 6).
5. Hevia, Cherishing Men, chap. 2.
6. Dennerline, Qian Mu, 8–10.
7. Dikötter, Discourse of Race, esp. chaps. 3 and 4.
8. On the Enlightenment narrative of History, see Duara, Rescuing History, chap. 1. To contrast Han History with those of the various minzu, see Harrell, “Civilizing Projects,” esp. pp. 25–27. I have borrowed the notion of a “civilizing project” from Harrell’s article.
9. For a detailed essay on the jianshi of the “Miao minzu” of southwest China, see Litzinger, “Making Histories.”
10. Many of the papers in Harrell (ed.), Cultural Encounters, undertake this work, especially those by Harrell, Diamond, Litzinger, Khan, Borchigud, and Hsieh.
11. Muslims may be divided by language group: the Uygur, Qazak, Tatar, Uzbek, Salar, and Kirgiz are Turkic-speakers; the Dongxiang and Bonan speak Mongolic languages; the Tajiks speak a Persian-based language; and the Hui, the default Muslims of China, are primarily Sinophone but include small populations that use Tai, Tibetan, and other “minority” languages.
12. In pre-PRC China a Chinese who converted to Islam would become a Hui, that is, a Muslim. Now that is no longer the case, for minzu is not a category based on religion. Such a person would be a “Han who believes in Islam.” Similarly, a Tibetan Muslim’s minzu status would depend on when the conversion took place. If it were back in the 1780s, the male descendants of that Tibetan would now be members of the Hui minzu. (The minzu status of females often depends on that of their fathers, so only some female descendants—those whose fathers were also Muslims—would now be Hui.) If the conversion took place now, however, that Tibetan would remain a member of the Zang minzu, but one who believes in Islam. Dru Gladney has written an ethnography of the Hui, focusing on the diversity and local character of their cultures and Islamic practice (see Muslim Chinese, esp. chap. 7).
13. A concise version of this argument may be found in Lin and He, Huihui lishi, 1–12.
14. Nakada, Kaikai minzoku, chap. 1. The Uygurs called by this name in pre-Ming times were not of the same culture as twentieth-century Uygurs, though they came from the same part of the world. Members of the contemporary Uygur minzu must by definition be Muslims, and the Tang to Yuan period Uygurs were not.
15. Some scholars, including Gladney (Muslim Chinese, chap. 7) and Harrell (“Civilizing Projects”), argue that the whole notion of “ethnic group” or “minority nationality” can be created only in the context of the modern nation-state, a position with which the evidence presented here leads me to concur (see chap. 6, this volume). Duara (Rescuing History, 52–56) disagrees, arguing against Anderson that the gap between modern and premodern self-consciousness is neither so wide nor so radical as he claims. Duara does, however, recognize the nation-state as a new kind of community, despite its potential ties to “archaic totalizations.” There is an enormous literature based on the assumption that the Hui minzu was formed in Ming China and that it was firmly established by the Qing conquest. The simplest, most essentialized description may be found in Huizu jianshi, 13–21. An earlier, somewhat more academic treatment, with substantially the same conclusions, may be found in Xue Wenbo, “Mingdai yu Huimin.” Many of the texts that will enable an accurate assessment of Islam under Ming rule have been collected and analyzed by Tazaka Kōdō in his magisterial Chūgoku ni okeru Kaikyō (esp. vol. 2, sec. 4–5).
16. I have had stimulating, frustrating discussions on this subject with colleagues in Beijing, Gansu, and Ningxia. They generally concluded that I, being a foreigner and a non-Muslim, did not understand correctly, while I found that they could not tolerate my argument, which challenged one of the crucial underpinnings of their view of the world, the minzu paradigm.
17. Almaz Khan, “Chinggis Khan”; Crossley, Orphan Warriors; and Dikötter, Discourse of Race.
18. It is both significant and fascinating that many of these Chinese words—minzu, zongjiao, guojia—were first created in Japan as translations of Euro-American terms.
19. Arguments among Muslims on “what it means to be a Muslim” are as common as the remarkably similar debates over the meaning of Jewishness in Jewish communities all over the world. Muslims have rhetorically, ritually, or even politically placed other Muslims outside the umma over many issues, a fact we often forget in our desire to formulate a unified Muslim Other. I would not argue that anything is possible within Islam—or Judaism, for that matter—but rather that the range is wider than we usually imagine, and that this variation is germane to the study of Muslims anywhere.
20. Gladney, Muslim Chinese, appendix A, includes a list of the names of many of these organizations or solidarities.
21. Harrell, “Civilizing Projects, 28.
22. Hevia, Cherishing Men, chap. 2. As Pamela Crossley recently noted in a personal communication, “Empires pinpoint centers, nation-states construct boundaries.”
23. The tusi had originally been established by the Mongols, and the Ming continued some of their appointments in the limited peripheral areas under their control. The Qing, of course, ruled an empire twice the size of the Ming state.
24. I am grateful to Stevan Harrell for this insight, based on his fieldwork in southwest China.
25. Needless to say, this characterization is applied almost exclusively to male Muslims. I have found almost no premodern Chinese sources that even mention the existence of Muslim women except as victims of massacre, objects of pacificatory action by the state, or statistics in census or taxation records. A conversation in May 1996 with Ms. Ma Yaping, a graduate student at the Gansu Nationalities Institute in Lanzhou, confirmed that Chinese scholars have not found any sources, either. Ms. Ma must restrict her research on Sino-Muslim women in Gansu to the People’s Republic (post-1949) for lack of information on any earlier period. For more detailed analysis of this theme, see Lipman, “Ethnic Conflict.”
26. I have dealt with this problem of perception in detail in Lipman, “Statute and Stereotype.”
27. Dikötter, Discourse of Race, chap. 1, sec. 2.
28. White, The Middle Ground.
29. In this millennium, only the Xixia kingdom of the Tanguts, a Tibetan-speaking people, created an independent polity in this region, but many local power-holders (including twentieth-century warlords) were able to establish semi-independent satrapies under distant regimes.