“A Fierce and Brutal People”: On Islam and Muslims in Qing Law
“A Fierce and Brutal People”: On Islam and Muslims in Qing Law
Abstract and Keywords
This chapter analyzes the shifting terms of legal description and classification employed by the Qing state, revealing the fluid nature of ethnic discourse and the conditions under which diverse strategies were chosen. At the heart of the exploration lie Qing perceptions of a particularly enigmatic ethnic or communal Other, the Sino-Muslims, who lived both on physical frontiers and in the heart of China, and who occupied a marginal position in discussions of Chineseness (they are and are not) and in discussions of non-Chinese peoples (they are and are not).
Qing local and metropolitan officials in the military, foreign relations, revenue, and legal bureaucracies addressed the multiple dilemmas of ethno-cultural difference as their offices demanded, but they did not work within a single system of categories or vocabulary, nor did they have a unified and consistent system of precedents to guide them. Rather, they acted upon and responded to difference across the empire and to change over time with a variety of categorization and perceptual schemes. Emperors and officials did have a commitment to continuity, but the policies handed down from the past—that is, the lessons of history—could justify many (though not all) courses of action, so we must examine perception and identification at specific moments in particular contexts.
The received view of Chinese culture, in both China and the West, postulates a monolithic sinicization model, in which a benevolent center gradually acculturates and then assimilates the barbarians who (voluntarily or not) submit to its obvious moral power Historical sources reveal a much more diverse repertoire of perceptual and categorization modes in state dealings with Others. Qing rulers and their servants used law—by which they meant criminal statutes and their associated punishments—as a primary tool for controlling the societies over which they ruled, including many ethnic and cultural Others. This essay analyzes the shifting terms of legal description and classification employed by the Qing state, revealing the fluid nature of ethnic discourse and the conditions under which diverse strategies were chosen. At the heart of the exploration lie Qing perceptions of a particularly enigmatic ethnic or communal Other, the Sino-Muslims,1 who lived both on physical frontiers and in the heart of China, and who occupied a marginal position in discussions of Chineseness (they are and are not) and in discussions of non-Chinese peoples (they are and are not).
(p.84) The Qing officials whose written documents constitute the sources for this essay certainly did conceive of the world in terms of Us and Them. Their problem lay in defining “Us” in a nexus of shifting ethnic relationships and “Them” in ways that were politically expedient, enforceable, and common-sensical within the multiple discourses of Manchu-Chinese-Confucian government and the multiple pressures under which officials worked. Qing approaches to ethnic difference shared the assumption that people who are different from one another, if allowed to live in proximity, will eventually conflict and cause social disorder. Some solutions involved altering the nature of the subordinate group(s) to eliminate conflict, while others recommended segregation and surveillance as the only effective antidotes to intergroup enmity. All strategies declared the maintenance of order and prevention of conflict to be the duty of the state, though methods differed as specific officials and administrative organs deployed one perception or another in the course of duty. Like models of ethnic difference elsewhere, all Qing perceptions partook of the struggle between understanding Others as inherently different and understanding Others as having learned differently.
Like our familiar sinicization model, what Stevan Harrell has named “the Confucian civilizing project” contends that all people may become civilized by absorption of Chinese literary culture and its moral principles (wenhua, literarization), though some non-Chinese are more capable of this acculturative leap than others.2 Based firmly on the notion of culture as learned, this project arranges peoples hierarchically by their distance from Confucian conceptions of virtue. Even barbarians may become officials of the empire if they study hard and succeed in becoming cultured, and lower-class Chinese criminals forfeit their civilized status by their defiance of Confucian norms. Despite its apparently egalitarian potential, this perception nonetheless enables powerful prejudice by equating civilization and decency with Chineseness as defined by the literary and political elite. Until non-Chinese manage to attain moral (i.e., Chinese elite) values, they must remain in the outer darkness of barbarism, dominated by the smell of mutton. But difference is ameliorable; they can become civilized if they will, aided by proper policies from the civilizing center and its officials.
In strong contrast we find what Frank Dikötter calls “the discourse of race,” which divides humankind into mutually exclusive groups defined by blood, by descent, by permanent genetic markers of inferiority and superiority that cannot be fundamentally altered by education, ritual, or other ameliorating practices.3 In this conception, Others are savage, ferocious, and irremediably different by nature. When civilized and uncivilized meet or live in proximity, this discourse mandates legal and military dominance as the appropriate mode of interaction. Very powerful in both official language and popular attitudes, this way of thinking about ethnic groups sets clear limits to the potential of acculturation as a mitigator of difference. Though opposite (p.85) in their fundamental assumptions about the nature of differentness, the civilizing project and the discourse of race usually intertwine in Chinese history.4
The Qing, as a conquest dynasty, generated another set of perceptions, practices, and precepts for the management of ethnic relations in China.5 The seventeenth-century Manchus expanded their empire far beyond the frontiers of the Ming, absorbing Mongolia, parts of Tibet, and finally all of eastern Turkestan by conquest and alliance. Themselves susceptible to Chinese opprobrium as uncivilized, the Qing rulers defined their empire as a multiethnic hierarchy with themselves at the center The Qing emperor, head of the Aisin Gioro clan, stood as supreme overlord, dominating his many subordinate lords (and their peoples). This imperial construct stipulated submission and tribute to the center as the duties of Chinese, Turkic, Mongolian, Tibetan, and other subject lords (including the lesser Manchu clans), granting them in turn a measure of autonomous rule over their own peoples. To prevent alliances against them, and to maintain social order, the Manchu hegemons delineated and tried to maintain the cultural and physical boundaries that separated the subordinate peoples from one another.6 Though leaning toward a “natural” division (by language, pastoral or agricultural life, climate, martial character, etc.), this strategy does not take a specific stand on whether differentness is inherent or learned but rather allows for wide variety under the emperor’s benevolent rule.
Another distinct approach to differentness focused on ideas and practices, on what was learned within an identifiable group. In the perception of some Qing officials, Islam and Christianity, to take two relevant examples, were pernicious doctrines, sufficient to guarantee that their followers would be wicked, divisive, and disorderly. Their presence constituted a difference of ethnic proportions, requiring segregation or even in some cases extirpation to alleviate the danger. Officials identified specific beliefs or behaviors as heterodox (Ch xie), condemning those who believed or practiced them to barbarian or criminal status.7 Learned doctrine, in other words, can itself be perverse or evil, overcoming whatever propensity for good its holders might possess. Only by repudiating heterodoxy (and those who continue to espouse it) might a person or group return to the protection of imperial benevolence.
A final categorization scheme bifurcated all humankind into good people and bad people, maintaining that both may be found in all groups. Government’s objective should be to protect the good from the bad, to punish the bad by law, to enable the good to live peacefully. All are the emperor’s children, but only the good deserve His Majesty’s grace. Biological at base, this strategy intersects with neo-Confucian notions of individual allotments of clear and turbid qi, 8which allow broad but not unlimited scope for individual will to overcome a poor natural endowment. It could also be used to isolate the leaders of seditious movements, rebellious sects, or perverse (p.86) doctrines—who are obviously bad people—from their ignorant and deceived followers, who might be good though stupid. One’s inborn character, including the desire to improve oneself, is so important that good doctrine will not necessarily change the wicked, nor heterodoxy deter the good. Variants of this perception appear often in imperial edicts, demanding of His Majesty’s subjects only that they expel the wicked from their midst and live up to their own potential for good.
All of these perceptual modes were available under the Qing; both officials and common people drew on them as they lived and worked within a multicultural polity. Since they were never encoded as creeds, these perceptual schemes could be utilized situationally or proclaimed as universal truth, depending on the speaker’s position, goals, and personal proclivities, thereby preserving their fluidity and accessibility.
Categorizing the Sino-Muslims
The Sino-Muslims have presented problems of categorization and control for all of the states that have ruled China since the Song dynasty. Acculturated to local Chinese ways, sinophone but obviously different from ordinary Chinese, they have not even been easy to name. Many terms have been used in official discourse: wushi fanke and tusheng fanke (fifth-generation and native-born foreign sojourners) in the Song; semu ren (subjects of various categories) in the Yuan; huihe (Uyghurs) and its later variants huihui and hui (Muslims) in the Ming; hanhui (Han Muslims), huizi, and donggan 9 in the Qing; hui (Sino-Muslims) in the Republic; and hui minzu (the Hui “nationality”) under the People’s Republic. Other, less savory descriptors have also been common in the official and unofficial record, words like huizei (Muslim thieves), huifei (Muslim bandits), luanhui (disorderly Muslims), and hui written with the “dog” radical. All of these categories have focused on the differentness of the Sino-Muslims from ordinary Chinese, defining these particular people as what we might call an ethnic Other.10
These ethnonymic strategies have influenced historians by defining the terms of discourse, none more than that of the current Chinese government. The “civilizing project” of the Communist government has found the Sino-Muslims troublesome to define, compared to its other non-Chinese citizens, because they are not a distant frontier people, not a territorially, linguistically, or even culturally distinct Other11 Choosing the Procrustean constraints of the minzu paradigm, the People’s Republic has created the hui as one of the oddest of the fifty-five “minority nationalities.”12 They hardly fulfill any of the four orthodox Stalinist criteria for minzu-hood—they have no common territory, no common language (except Chinese), no common economy, and very little common culture apart from their religion and its associated practices. Historically related to one another primarily by their (p.87) adherence to (or descent from adherents to) Islam, the members of this hui minzu have nonetheless been told, and now generally claim themselves, that they are a minzu united by bonds of consanguinity and psychological nature.
Religion, in this hegemonic paradigm, is only one of a number of common cultural characteristics that define a minzu. Thus, the Chinese-speaking Jews do not constitute a minzu because they have become entirely Han (despite their ancestors’ belief in a foreign religion), while believers in Tibetan Buddhism have been placed in various minzu (zang, meng, yi, etc.) defined by their languages, territory, and other markers of culture besides religion alone. But according to the apparatus of minzu “identification” and the Nationalities Commission, the Sino-Muslims, plus some Muslims who speak Tibetan, Bai, Tai, and other Southeast Asian languages, are a single minzu. It has proved impossible for late twentieth-century Chinese scholars to escape from this essentializing paradigm, enforced as it has been by a wide array of state agencies and local arrangements.13 Foreign scholars, too, have found it convenient to write as if the minzu paradigm constituted an accurate description of pre-1949 social reality.
Certainly the Sino-Muslims posed special challenges to the Qing, but not because of their transregional unity or minzu consciousness. In Qing conceptions of ethnic relations, the dangers of communal conflict and chaos increased if people defined as outsiders left the frontier to live in the Chinese culture area itself. There, since the Yuan, the Sino-Muslims have constituted the most widely distributed group of domesticated Others. They live all over China, from Yunnan to Heilongjiang, from Gansu to Fujian, and they have maintained their differentness, embodied in their religion, with considerable vigor, while acculturating to the language(s) and local cultures of their non-Muslim neighbors.14 In the Qing scheme of five distinct linguistic groups, the Sino-Muslims could not constitute a “cultural bloc” parallel to the Tibetans, Chinese, Mongols, and Turkestanis (also called hui, meaning Muslims), for they spoke Chinese as their native tongue. Their leaders could not be ensconced as Qing subordinate lords, for the Sino-Muslims had no empirewide or even regional leaders.
In fact, the Sino-Muslims looked remarkably Chinese to the Qing, so the queue was imposed upon them (but not upon Turkic-speaking Muslims), and they were subjected to antimiscegenation regulations in their relations with non-Chinese-speaking Muslims. Despite their obvious and sometimes contentious differentness from non-Muslim Chinese, the Sino-Muslims lay theoretically within the Chinese category in the Qing scheme of ethnocultural boundary maintenance. They had special names (hui, etc.) and clearly were not han, but they could not be considered entirely non-Chinese. From the conquest onward, Qing judges, their superiors in the bureaucracy and the Manchu emperors themselves wrestled with the ambiguous status of the Sino-Muslims and the difficulty of defining and controlling people who belong (p.88) simultaneously to two exclusive groups—Chinese and Muslims.15 Sino-Mus-lim legal affairs came under the jurisdiction of the regular civil officials, not the military or the Lifan Yuan (except for cases adjudicated in Xinjiang, Tibet, or Mongolia), an administrative decision that emphasized their Chineseness. But their non-Chinese quality could be called upon as an explanation when they broke the law, attacked non-Muslims, or organized to oppose state authority. The Qing solidified their perception of this complex identity in the term hanhui, Han Muslims, which both officials and common folk often used to separate the Sino-Muslims who lived in Xinjiang from the Turkic-speaking Muslims. (This term cannot be tolerated, of course, within the post-1949 minzu paradigm, for Han and Hui are now and always have been two distinct minzu.)
The Qing Emperors and their Officials
To clarify the tensions among various perceptions of Sino-Muslim difference, let us consider officials’ proposals and the Qing emperors’ pronouncements on the Sino-Muslims. During the Qing conquest itself, a Muslim-led army of resistance caused the invaders considerable trouble in Gansu province, where large numbers of Sino-Muslims lived (and still live). Seeking a solution to the violence that had disrupted Gansu, an official of the Board of War recommended segregation of the Sino-Muslims so that their inherent barbarousness might be tamed: “Their customs are different and this results eventually in mutual suspicion…. Forbid them to breed horses or to keep weapons. Command their religious leaders to take charge, regulating their movements back and forth. Let them all cultivate the soil, and so gradually allay their ferocious natures.”16 This metropolitan official believed that peaceful husbandry could work a deracinating magic on the savage Sino-Muslims, as long as they were kept far from the nearest non-Muslim settlements and deprived of their martial matériel. In the event, however, the demands of conquest and empirewide control prevented his suggestions from being implemented, and Muslims remained in the cities and towns of Gansu, often in close proximity to non-Muslims.
Once the wars of conquest were completed, the three great Qing rulers could eschew the racial component of that official’s characterization and proclaim their impartial benevolence toward Sino-Muslims and non-Muslim Chinese, repeating the slogan “Equal benevolence toward Chinese and Muslim” (Ch han hui yishi tongren). In a 1694 edict preserved in Sino-Muslim sources, the Kangxi emperor wrote: “Considering that those Han (Chinese) officials who receive royal appropriations regularly attend court only once a day, while the Muslims who do not have any royal appropriations still worship God and praise the sages five times daily, the Han are certainly inferior to the Muslims.”17
(p.89) This did not stop eighteenth-century local and provincial officials from raising the specter of Sino-Muslim bestiality and violence, requesting strict action lest the Muslims’ mere presence among the Chinese cause social chaos. Risking severe imperial disapproval, Chen Shiguan, a Shandong judge who later became a grand secretary, argued in 1724 from his own experience that the Sino-Muslims are not the same as other subjects, and that Islam itself should be prohibited: “It is a perverse doctrine that deceives the people and should be banned by law. Those who enter it do not respect Heaven and Earth and do not worship the gods, instead setting up their own cultic deity [Ch zongzhu]…. They aid the evil and harm the people. Please force them from their [perverse] teaching and destroy their mosques.”18 The Yongzheng emperor responded that though Islam is a foolish and not particularly popular religion, it comes down from antiquity to its low-class adherents and does not present any particular danger to society.19 In 1729 he reiterated his impartiality in an edict to the Grand Secretariat:
All over the direct [-rule] provinces, the Hui [Muslim] people, having resided there from of old, are enumerated as part of the population and are all children of our country. It follows that they cannot be regarded as separate. Over the years secret memorials have frequently been submitted arguing that the Hui [Muslim] people maintain their separate religion, speak a foreign tongue, wear strange clothes, and are fierce, perverse, and lawless, and demanding that they be strictly punished and placed under restraint.20 I deem, however, that the Hui [Muslim] people have their religion because their ancestors bequeathed them their family habits and local customs…. As long as they peacefully keep their customs they are not to be compared with traitors, lawbreakers, or those seeking to delude and lead people astray…. Our court looks on them with the same benevolence as on all.21
Despite this clear imperial opinion, the following year Lu Guohua, an Anhui judge, memorialized in language similar to Chen Shiguan’s, claiming that Sino-Muslims’ differentness deluded the people. He requested that the Throne make an example of the Muslims and punish them severely. The emperor replied peevishly that Lu’s memorial was harsh and absurd, removed him from office, and ordered him back to the capital for punishment. The emperor thus pitted his own understanding, based on an ideological combination of supreme overlordship and the Confucian civilizing project, against the ethnocultural prejudices of some of his own officials, who had the job of actually governing mixed communities of Muslims and non-Muslims. We should note for reference below that neither Chen Shiguan nor Lu Guohua was posted to a “Muslim” area such as Gansu or Yunnan; both of them served in the north China plain, where substantial communities of Sino-Muslims lived among vastly superior numbers of non-Muslims.
In the Qianlong period, the imperial stand against systematic anti-Muslim (p.90) prejudice began to erode, along with social order all over the empire. Creating significant precedents, the Qianlong emperor sometimes followed his local officials in singling out the Sino-Muslims as especially (and congenitally) violent and troublesome. Dou Bin, a senior official in Guangxi, had experience in the Northwest and memorialized in 1750 against the appointment of a Sino-Muslim as military commander in Gansu: “This sort of people [Sino-Muslims] put violence before everything and have no loyalty to the state. The rich among them make trouble and the poor go in for thieving. They are basically different from ordinary folk. Now Ha P’an-lung [Ha Panlong]…too is a huizi. Granted that he is not lax in discipline, but what if he shows religious sympathies?”22 Rather than claiming han hui yishi tongren or referring to the many Sino-Muslims who had loyally served the dynasty as civil and military officials, as his father and grandfather might have done, the Qianlong emperor responded, “This memorial is highly commendable. Noted.”
One of the dynasty’s most distinguished provincial officials, Chen Hong-mou, served for two terms during the 1740s as governor of Shaanxi, which had one of the empire’s largest Sino-Muslim populations. His experience there impelled him to compose and put in force a provincewide “Covenant to Instruct and Admonish the Muslims” in 1751. That document includes a variety of discriminatory legal practices aimed at curbing what Chen saw as the naturally “fierce and brutal” (Ch qianghan) character of the Shaanxi Huihui: “How can they become accustomed to defiance and unrestrained criminality, entirely without scruples? The Chinese fear the Muslims as if they were tigers, and the Muslims look on the Chinese with hatred. Because of their wild and intractable character, [the Muslims] are seen by local officials as nothing but unruly and rebellious people from the edges of civilization.”23 In his covenant, Chen isolated a variety of criminal acts commonly committed by Muslims: collective brawling, rape, stealing ripe grain from the fields, cat burglary and fencing stolen goods, loosing flocks of sheep into others’ fields, livestock rustling (for immediate and covert butchering), creating disturbances at public theatrical performances, disrupting market trade by attacking those who sell pork products, violently contending with one another over religious issues, and mutual protection in cases of feud or criminal behavior.
Aware that these crimes were most often committed by young men operating in gangs, Chen recognized that “among [the Muslims] are many law-abiding, experienced, steady, circumspect, and magnanimous people, who know what should be feared and are inclined to delight in obedience.”24 Nonetheless, he found that the only rational way to prevent “Muslim violence” lay in making the local Muslim leaders themselves legally responsible for the behavior of their communities. He distributed copies of the covenant to religious (p.91) professionals and lay elders, insisting that they should be punished as principals in crimes committed by Muslims if they failed to report them to the local officials. Though Chen’s covenant was never adopted by the Board of Punishments, and we do not know if the Qianlong emperor approved, it presages the gradual transformation of the Qing code over the next century: “Now I have received the imperial decree strictly forbidding relying on mobs to behave violently. The orders are extremely strict, and the legal categories not light. I, the governor, personally fear that more Muslims than Chinese rely on mobs to do evil. The crimes committed by Muslims must be more serious than [those committed by] Chinese…. Although there are Chinese who commit these crimes, overall they are not so grave as Muslims.’”25In this view, Muslim criminals should be treated more harshly than Chinese criminals because they are Muslims, a perception that would haunt Qing legal opinion for the remainder of the dynasty.
In 1762 Ebi, governor-general of Shaanxi-Gansu, again characterized the Muslims as inherently violent and proposed similar techniques to deal with their lawlessness. Noting that the mosques lie at the core of every Muslim community, and that every mosque has a religious leader (Chjiaozhang), Ebi memorialized:
The Hui in the country districts administered from Sian [Xi’an] traditionally stick together and rely on solidarity and strength of numbers to insult the Han. Truculent and ruffianly, they go openly to rob and steal, and hamlets are really at their mercy. If by chance the culprits are exposed, the matter is just treated in the ordinary way and no more, and some officials are so feeble that when they get involved in strife with the Hui [Muslims] they do not even report it unless there is loss of life…. Your Majesty’s servant is having notices printed showing the crimes which the Hui are wont to commit, together with the various punishments prescribed in the regulations. These notices will be issued to the religious heads at each village with orders to display them in the mosques. The religious heads…will also be made responsible in case of lawless activity to come forward with advance information…. The severest measures are necessary to deal with the degree of savagery now existing.26
Ebi instituted a self-policing system similar to Chen’s, making the imams responsible for the behavior of their congregants, and the emperor concurred, with a cautionary warning that “measured pace and perseverance” would be needed.
As both Chen and Ebi correctly perceived, the mosque and its religious professionals constituted a crucial difference between Muslim and non-Muslim communities everywhere in the Qing empire. As community center, Koranic school, courthouse for judgment under sharia law, martial arts training hall, and meeting place for daily and weekly communal worship, the mosque (and (p.92) its leadership, both lay and professional) made up an institutional core unmatched in non-Muslim society, except perhaps in sectarian Buddhism and Daoism, or in Christian communities. By singling out the imams, and by lumping all Muslims together as truculent and ruffianly, Ebi advanced the process that would soon lead to statutes specifically discriminating against them.
Chen Shiguan, Lu Guohua, Dou Bin, Chen Hongmou, Ebi, and many other Qing officials believed that the closed nature of Sino-Muslim communities and the differences between Sino-Muslims and their non-Muslim neighbors led directly to criminal behavior, which was in any case inherent in the Muslims’ character Reacting to growing social disorder and his officials’ perceptions, the Qianlong emperor accepted Ebi’s plan of action, singling out Muslims and their mosques as sites of potential danger requiring special treatment. In late 1762, only a few months after Ebi’s memorial, the Board of Punishments recommended, and the emperor accepted, the first statutory distinction between Sino-Muslims and Chinese within the criminal code (see below).27
Social violence increased all over the Qing empire after the middle of the Qianlong period, but the court did not hew to a consistent line regarding Muslims who broke the law. In 1819 the Jiaqing emperor, echoing his grandfather more than his father, ordered Muslims to be treated just like other subjects under the law. Responding to a memorial from Censor Zhang Yuanmo, who requested that Muslims not be allowed to serve as officers under local magistrates, the emperor held that all provinces have criminals and asked why we should select the Muslims to be punished with stricter statutes.28 Only two years later, however, his successor made the opposite argument in response to a similar request. Dai Junyuan asked the Daoguang emperor for special regulations to deal with violent crimes committed by Muslims, calling them exceptionally arrogant and insolent, gathering crowds for brawling at a moment’s notice. The emperor agreed.29 But in 1853, even in the shadow of Muslim rebellion in Yunnan, the Xianfeng emperor reiterated his ancestors’ old adage of han hui yishi tongren. When Li Jiaduan and Yuan Jiasan reported “Han-Hui violent incidents” in Anhui that year, he responded, “They are all the Court’s children…and must be treated with equal benevolence…only equal treatment can quiet these conflicts.”30
Thus we cannot isolate a clear trend, though the Qianlong emperor certainly did open the door for discriminatory statutes. The emperors and their officials continued to call upon different perceptions and corresponding legal formulae depending on their posts, their experience, and their policy options at a particular time. Local officials in Muslim areas of north China (e.g., Shandong, Anhui) seemed more inclined to request prejudicial statutes or policies than their imperial masters, but that hypothesis must be tested with more thorough reading of the sources.
(p.93) Muslims in the Qing Criminal Code
Criminal codes and books of legal precedent are not merely guides for judges and clerks. They embody finely tuned gradations and nuances of domination and can inform us of subtle transformations within the state, including reactions to specific kinds of behavior perceived as antisocial or antistate. Examining changing precedents and punishments over time, we can learn not only that a particular act was considered illegal, but how illegal. The Qing state did not simply categorize its subjects into criminals and law-abiding subjects. Among the guilty, individuals were distinguished by the distances to which they might be banished and their status in exile, the number of blows they might receive with the heavy or light bastinado, the period of time they might wear the cangue, the specific words that might be tattooed on their faces, the length of time they might be imprisoned, and the severity of pain they might undergo while being executed, revealing Qing official ranking of the seriousness of their crimes.
The Kangxi and Yongzheng emperors had not been willing to distinguish Sino-Muslim from non-Muslim in criminal law, but the Qianlong emperor was, at least after 1762. This represents a change not only in Board of Punishments policy but also for the Qing state as a whole. No longer would the Confucian civilizing project dominate imperial discourse on the Sino-Muslims; ethnic distinctions based on inherent characterization of those particular Others as barbaric would also be approved and codified. Having an anti-Muslim ax to grind, or under pressure from anti-Muslim gentry and militias, local and regional officials continued in the footsteps of Chen Shiguan, Chen Hongmou, and Ebi, proposing communal or ethnic distinctions, which became law as the Qing state reacted to internal disorder with increased coercive violence. When social tensions rose and disorder increased in the Qianlong period, so too did discriminatory judgments against Sino-Muslims. By the mid-nineteenth century, the Board of Punishments had created a body of statutes and precedents that allowed magistrates to punish Sino-Muslims more severely than non-Muslim Chinese for the same crimes.
These legal decisions also gradually reified the widespread social stereotype that Sino-Muslims are inherently more violent and dangerous than non-Muslim Chinese people.31 This perception had bloody consequences for large regions of China between 1781 and the middle of the twentieth century. Some Qing officials also claimed that Islamic doctrine and practice made the Sino-Muslims’ very presence inimical to the imperial order They requested that the emperor ban the outsiders’ religion and remove this danger to peace and tranquility. Despite these perceptions, the Qing emperors and law courts never found Islam itself, as a religious system, to be illegal or dangerous to the state. Islamic teachings belonging to particular solidarities or factions might be judged heterodox because they produced social disorder. (p.94) But even during and after the widespread “Muslim rebellions” in the Northwest and Yunnan, Islam continued to be recognized by the emperors as a legitimate religious doctrine that could encourage virtue among the emperor’s subjects, despite the claims of many officials to the contrary.
Having approved Ebi’s 1762 plan (which mirrored Chen Hongmou’s 1751 covenant) to treat Muslim and non-Muslim criminals and communities differently, making Muslim religious leaders responsible for their congregants’ behavior, the Qianlong emperor and his administrators were soon asked again to discriminate between Muslims and non-Muslims under the law. In late 1762, Min Eyuan, a Shandong judge and later governor of Jiangsu, memorialized regarding a case in which local Muslims had formed a gang of thieves, and he requested severe penalties under the legal rubric of “organized larceny.”32 The Board of Punishments concurred in a closely argued response: “As for Muslims who go out to plunder, if the stolen goods are recovered entirely, there can be no increasing the penalty; and if they do not gather together bearing weapons, they may be handled according to the [regular] statute. But if they gather more than three men bearing ropes, whips, and weapons, then leaders and followers should not be distinguished; neither the amount of loot nor the number of offenses should be taken into account. All without exception should be deported to the farthest miasmic frontiers of Yunnan, Guizhou, or Liang-Guang for penal servitude [Ch chong-jun].”33Under the ordinary statute, the amount of loot and the distinction between leaders and followers would have been used to calculate appropriate punishments. When the criminals were Muslims, however, Min Eyuan and the board found that those particular distinctions should not mitigate the penalty. We may surmise, since the documents do not discuss the officials’ motivation, that they believed all Muslim criminals to be equally dangerous to social order The Board of Punishments, in an 1825 supplement to this special Sino-Muslim statute, openly reasoned that Muslims are “ferocious and recalcitrant by nature, frequently banding together” for criminal action.34
In the same vein, a codicil dealing with Muslims was also added to the code regarding the harboring of outlaws and the storing or fencing of stolen goods, specifying penalties equal to those of the original offenders for those who aid them. If Muslims committed these crimes, their punishments could be one or more degrees harsher than those meted out to non-Muslim Chinese who committed the same offense.35Minority communities, with their perceived solidarity in the face of authority, have always proved difficult for the state to penetrate, so when Muslim criminals eluded arrest or escaped into their closed communities, they (and their loot) were much harder to capture than non-Muslims. The officials reasoned that more stringent punishments might discourage Sino-Muslims from helping one another to escape justice or dispose of ill-gotten gains.36
(p.95) Tensions in northwestern Chinese society grew rapidly in the 1770s, some of them inspired by fear and cultural differences in local social interaction, others by poor government, local economics, and the gradual militarization of communities for self-defense.37 The Wei valley of Shaanxi, where social conflict had inspired Chen Hongmou’s covenant, was known for its militias, always divided between Muslim and non-Muslim, whose stockades and strong points dotted the hills and mountains both east and west of the provincial capital. Tusanga, a Shaanxi judge, memorialized in 1773 to request special punishment for Muslims who engaged in armed robbery. He argued that the Muslims depend on their strength to plunder others; since no special statutes exist to control them, they can easily evade severe punishment.38 Like Min Eyuan, he concluded that special Muslim statutes would be more effective, and the emperor concurred. Such a statute was added to the armed robbery section of the Code within a few years, with further precedents over the following decades, making it clear that Muslim robbers would be treated more harshly than non-Muslims for the same crimes.
Even when punishments for Muslims and non-Muslims were of equal severity, such as tattooing on the face, the Board of Punishments still distinguished between the two.39 A convicted non-Muslim might be marked qiedao (thief), but a Muslim would be tattooed with the characters huizei (Muslim thief). In 1768, acting on the recommendation of Wu Dashan, the governor-general of Shaanxi-Gansu, the Board of Punishments decided to brand some convicted Muslim thieves with both huizei and gaiqian (sentenced to banishment) to highlight both their crime and their ethnicity and separate them from non-Muslims.40
From ancient times, Chinese law had provided for mitigation of a harsh sentence if the convicted offender was an adult only son with aged or ailing parents, grandparents, or great-grandparents. Systematized in the Qing code, the practice of filial mitigation (Ch liuyang) allowed the convict to care for his parents before and after their death, undertaking the sacrifices that would allow their spirits to rest in peace. The conditions for its implementation were strict—request by an imperial official, no other male descendant over the age of sixteen, old folks over seventy or ill—but for those who met them, filial mitigation limited punishment to one hundred blows, eliminated any other penalty, and returned the convict to his family. The criminal code, in a codicil dating from the Jiaqing period, specifically prevented filial mitigation in cases of Muslims who engaged in armed robbery: “Regarding Muslims who plunder in a group of more than three, carrying ropes, whips, and weapons, and are sentenced to penal servitude in the Southwest…. Filial mitigation may not be granted.”41
A case from Shaanxi in the Daoguang period confirms that local officials implemented this statute in practice. A Muslim gang led by Mi Tianxi engaged in public brawling (Ch gong’ou), killing Jia De, and filial mitigation (p.96) was requested for one of the Muslim gang members, Mi Chongshour The judge decided that because Mi, a Muslim, had been convicted of armed brawling in a group larger than three and sentenced to penal servitude, he could not allow filial mitigation. This judgment was reversed on appeal. In its statutory review, however, the Board of Punishments affirmed that there might be cases in which filial mitigation for Muslims would not be granted.42
From many dynastic predecessors, the Qing inherited the custom of imperial amnesties, the acme of royal benevolence, which released convicted criminals of many classes from their sentences on occasions of imperial celebration. The monarch could exclude from compassionate amnesty any offenders too dangerous to be pardoned, and so ferocious were violent Muslims held to be that they could be prevented as a group from receiving His Majesty’s favor In the Xing’an huilanwe find amnesties of the Daoguang era that do not cover “Muslims sentenced to military servitude for collective armed crimes” and “Muslims sentenced to serve as foot soldiers for collective larceny”43 The exemptions are illustrated with a case from Shaanxi. Ma Yin, a Muslim, had been exiled in 1829 for collective larceny but managed to escape. Shortly thereafter an amnesty was proclaimed that would have released him from both his sentence and the penalty for escaping. Avoiding recapture, Ma Yin again gathered followers and committed armed robbery. When convicted for that second crime, because he was a Muslim he was given a harsher sentence—penal servitude in Xinjiang and tattooing on the face—despite the imperial amnesty.44
Beginning in the mid-Qianlong period and increasingly in the nineteenth century, we find the Board of Punishments equating violent Muslim offenders with the very worst elements in society. In the view of the Board of Punishments, they are none of them the same as ordinary people who commit violent acts but rather habitual, hardened, brutal criminals who must be punished especially harshly. A statute from the Daoguang period makes this connection explicit, noting that the fearsome brigands of Hebei and Anhui, and the Muslims of all provinces, form bands to feud and kill. Summarizing the matter, the Board of Punishments recorded the judgment that the Muslims of every province and the savage gangs of Nanyang, Runing, Chenzhou, and Guangzhou districts in Hebei should be considered to be the same.45
Though most Sino-Muslims never came under the purview of these criminal statutes, the perceptions embodied in the code certainly affected the Muslim expectations of and relationship with the state, at least its justice system. Muslims were not the same as other Chinese subjects of the Qing, nor would they be treated equally under the law. After the mid-Qianlong period, officials could (and did) freely equate them as a group with the most violent members of Chinese society, describing the entire Sino-Muslim population as qianghan, fierce and brutal, language that did not appear in the edicts and rescripts of the Kangxi and Yongzheng emperors.
(p.97) Distinctions within Islam
Distinctions between Muslims and non-Muslims who committed violent acts came to be formalized in the Qing code after 1762. Few of the cases mentioned in the brief summary above involved crimes committed in Gansu, the stronghold of the Sino-Muslims in northwest China. Officials knew that Gansu constituted a very different environment for Qing relations with its Muslim subjects. There the Sino-Muslims lived adjacent to non-Chinese-speaking Muslims of many cultures (Turkic, Tibetan, Mongolian) and in close contact with the entirely Muslim regions to the west. Gansu was a region where Sino-Muslims could be affected not only by changes within the Qing empire but also by influences emanating directly from Turkestan and the Middle East. Gansu officials had the choice of treating Muslims under the conventional code or under special regulations and a fairly lax legal regimen reserved for fan, non-Chinese. In the increasingly turbulent late eighteenth century, they used both, claiming that doctrine as well as ethnicity or evil character caused criminal behavior Though Chen Hongmou had noted the presence of feuding religious solidarities in Shaanxi before 1751, he had insisted that they all be treated equally under the law.46 In Gansu starting in this period, legal distinctions between orthodoxy and heterodoxy, between harmless religious groups and potentially subversive ones, were utilized by Qing officials to isolate some Muslims from other Muslims, to stigmatize some Muslim solidarities collectively as lawless, pernicious, heterodox, and inherently dangerous to social order.
In most such cases, Gansu Muslims brought the original accusations against one another, not for “collective armed robbery,” “fencing stolen goods,” or “armed brawling” but for ideological offenses, pitting one version of Islamic orthopraxy against another before a non-Muslim judge. Revealing the Sino-Muslims’ willingness to accept the Qing state’s legitimacy as the most effective authority available, these cases also constitute a useful analytical contrast to the statutory changes outlined above. Conflicts between Gansu Muslims often involved physical violence and destruction of property, but they were adjudicated under statutes also used to evaluate sectarian Buddhists or Daoists, and local officials sometimes chose to judge them as non-Chinese. In the Northwest, the Muslims were stereotyped as dangerous not only because they robbed and gathered armed bands but also because they feuded with one another as religious solidarities, their violence spilling over onto other frontier peoples who came to fear them as inherently fierce.47
In the local gazetteer of Xunhua (now in Qinghai but then in Gansu), a small town on the south bank of the upper Yellow River, we find recorded a conflict between two groups of Muslims, whom the Chinese texts call Fore-Breakers (qiankai)and After-Breakers (houkai),referring to whether they advocated breaking the daily Ramadan fast before or after prayers:
(p.98) In 1748, the After-Breaker Ma Yinghuan went to Beijing to accuse Ma Laichi of teaching heterodoxy to delude the people. Former Gansu governor Huang Tinggui investigated the matter and found that Ma Yinghuan should be punished for “false accusation” [with the penalty Ma Laichi would have received had he been found guilty].48 Following the precedents on conflict prevention, [Huang sentenced Ma Yinghuan to] penal servitude. He further instructed that when Fore-Breakers and After-Breakers conduct funerals, their religious leaders should not invite both litigants [together] to recite scripture. He ordered in the case file that incidents of disruption must cease. Ma Laichi and his son Ma Guobao thereafter traveled to and fro spreading their teaching.49
Apart from the superficial matter of the timing of meals, the Islamic grounds of conflict do not appear in this legal text, probably because the magistrate and his superiors did not try to understand them beyond the overt claims of the litigants.
From other evidence, we know that Ma Laichi was a Naqshbandi Sufi. He had been initiated by a disciple of an important Turkestani religious and political leader, Hidayat Allah (also called Khoja Afaq); studied in the Muslim heartlands; brought a new set of Muslim texts back to Gansu from the West; and advocated an unfamiliar form of social and religious organization, the Sufi tariqa.50One of his most divisive innovations lay in abbreviating the lengthy scriptural recitation customary at life cycle celebrations. Reciting a shorter text, which Ma Laichi had brought back from the Middle East, saved time and money for ordinary Muslims, who paid the religious professionals to read for them. Ahong 51who did not follow Ma Laichi’s teaching were thus deprived of some part of their income, which went to the new group.52
As part of his accusation, Ma Yinghuan claimed that Ma Laichi had founded Mingshahui, “Bright Sand Societies,” at whose meetings initiates had sand blown into their ears.53 His intention clearly lay in associating Ma Laichi’s Sufis with Daoist or Buddhist groups, always suspect in Qing eyes for their bizarre ritual practices and propensity for sedition. Ended by Governor Huang’s summary judgment, Ma Yinghuan’s suit failed to prevent Ma Laichi and his son from initiating many northwestern Muslims into their tariqa.
The Sufi content of Ma Laichi’s teaching certainly created some of the motivation for conflict within Muslim communities. But we must also look at the Qing legal grounds on which Ma Yinghuan brought suit, for the Qing officials’ understanding of this conflict determined its legal result, and their perception surely differed from that of the litigants. Ma Yinghuan accused Ma Laichi of xiejiao huozhong, teaching heterodoxy to delude the people, a very serious crime.54Based on the Qing state’s experience with potentially subversive Buddhist and Daoist groups, the statute concentrated on activities associated specifically with them, some of which coincidentally pointed directly at Sufis as well. For example, it prohibited meetings that took place (p.99) at night, which Muslims must do in order to eat together during Ramadan and Sufis often did as part of their mystical practice. The supplementary statutes forbade such practices as writing charms, preparing sacred writings (especially esoteric, encoded texts, such as those in Arabic and Persian), and collecting contributions, all normal religious activities for Sufis, though the statute’s authors knew nothing of Islam or Sufism.
Governor Huang found against Ma Yinghuan and ordered him punished, but a Qing court might well have approved the argument that Sufis practiced heterodoxy. In a legal culture that valued conservatism and harmony, Ma Laichi’s group might have appeared innovative and divisive. Its Sufi practices were certainly unlike those of non-Sufi communities, since the tariqa created new, specific, personal bonds of religious loyalty and communication between communities where only coreligionary ties had existed before. If Ma Yinghuan had been able to prove his allegations of sand-blowing or other bizarre practices, he might have been more successful, given the Qing’s anxiety about and the tariqa’s social resemblance to Buddhist and Daoist sectarianism. Fortunately for Ma Laichi and his Sufi solidarity, the label of New Teaching (Ch xinjiao), a pejorative and long-lasting name, did not stick to them as a result of this lawsuit.
After 1748, Ma Laichi and Ma Guobao continued proselytizing on behalf of their tariqa, moving from one Muslim community to another all over Gansu in a Sufi pattern very different from that of conventional Muslim clergy, who were attached to a single mosque and stayed there to serve the religious needs of their own congregations. Divisions among Muslims devoted to different leaders (and different ritual modes) spread as far as Shaanxi.55 More significantly, in 1761 another returned pilgrim, Ma Mingxin, founded a second Sufi group in Gansu, and conflicts between the two Sufi orders dominate the Qing records from that point onward. Unlike the criminal cases brought in “the interior” (as Gansu folks call the rest of China proper), the Gansu suits brought Islamic doctrine and religious behavior before the benches of Qing judges, with disastrous results.
Among Sino-Muslims, Ma Laichi’s Sufi order came to be called the Khafiya, Arabic for “silent ones,” because they repeated their mystical litany silently. The followers of Ma Mingxin, in contrast, chanted aloud, waved their heads and arms, even danced as they repeated their formulae of praise to God, so they were known as Jahriyya, the “vocal ones.”56 In official Qing parlance, however, these Arabic names had no standing. Because Ma Mingxin returned to Gansu long after Ma Laichi, the Jahriyya was perceived as having destroyed a once-stable social order, and it earned the negative appellation New Teaching, while the Khafiya, along with non-Sufi Muslim congregations, came to be called Old Teaching (Ch laojiao).57Since they were rivals for religious, social, and economic power, conflict between the two orders and their leaders came to the attention of the authorities almost immediately after Ma Mingxin’s (p.100) return from the west. In 1762, Khafiya adherents in Xunhua filed suit against the Jahriyya for deluding the people with a new teaching. The magistrate, who had neither precedent nor competence to judge Islamic orthodoxy, or any way to associate either position with Qing orthodoxy, ordered the litigants to return to their native places and feud no more.
Since neither was able to achieve dominance, the two orders competed in building new mosques and initiating new members. In 1769, a Khafiya leader drew up an indictment against the Jahriyya for “violating religious rules,” and the judge ordered three new Jahriyya mosques temporarily closed, with two of them to be reopened when peaceful relations were restored. He also ordered Hemaluhu, a Jahriyya leader, to wear the cangue (for reasons unspecified in the text). Hemaluhu appealed to a higher court, where the Khafiya filed a countersuit. After a thorough investigation at several levels of government, unable to find particular merit with either litigant, the judge punished both sides for false accusation. This solution satisfied neither side, so they continued their competition for Gansu Muslims’ allegiance and contributions.
The matter soon came to violence. In 1773, twenty Old Teaching (Khafiya) families in a Gansu village converted to the New Teaching (Jahriyya). Old Teaching adherents started toward town to file a complaint with the magistrate, for this transfer of allegiance represented a substantial loss for their order, but brawled with their rivals along the way and killed one. The magistrate concluded the case by assessing a fine, but two months later New Teaching men killed four Khafiya members. The murder was adjudicated under the less severe statutes for non-Chinese, so one New Teaching leader was sentenced to wear the cangue and the case was closed.58 Thereafter, when feuding between the groups escalated toward full-scale armed confrontation, the officials could not control it.
In 1780, New Teaching leaders gathered large contingents of armed followers to attack Old Teaching villages, which organized for defense but suffered considerable losses. The local officials, to whom the Old Teaching villagers fled for redress, did not even bother to hold trials or assess punishments, claiming that such conflicts among frontier peoples were reflections of their savage nature and could not be tamed by civilized law. Instead they sent troops, but the inexperienced detachments did little to prevent violence. Old Teaching leaders then appealed to the provincial authorities in Lanzhou, who sent middle-ranking military and civilian officials to investigate. Moving toward Xunhua at the head of a platoon, they proclaimed their intention to eradicate (Ch xi, wash away) the New Teaching if it did not respect the law (Ch ru bu zun fa). 59New Teaching leaders, accurately recognizing that their solidarity had been singled out for punishment, met the officials on the road, put the troops off their guard with a deception, and killed them.
(p.101) From this point in the spring of 1781, the documents subsume the Old versus New Teaching rivalry in a different legal category, that of rebellion. The Qing state could not tolerate violence aimed directly at its officials, the local representatives of imperial majesty, and moved immediately to demonize and then exterminate the New Teaching Muslims who questioned its legitimacy. The narrative above, distilled from official accounts, led them to perceive the New Teaching leaders, doctrine, and organization as the source of evil. Unable to stop the New Teaching Muslims from taking prefectural towns and then besieging the provincial capital itself, Beijing sent Agui, a grand secretary with extensive military experience on the southwestern frontier, to lead a multiethnic army against the rebels (as they were now defined in law and state perception). In a three-month campaign, the Qing forces—including Old Teaching Muslims, Tibetans, Mongols, and provincial troops—crushed the New Teaching’s armed resistance.
In the aftermath and pacification, which included a second armed action by Gansu Muslims against the Qing in 1784, Agui and his staff assessed what legal measures the state ought to take to prevent further violence. Though they blamed the New Teaching and its leaders for the rebellions, their proposals cast a much wider net around many Islamic practices, combining Chen Hongmou and Ebi’s proposals, the xiangyue system of local propaganda and control officers (which translates awkwardly into Islamic forms), strict limits on the movement of Muslim clergy, and a generally discriminatory attitude toward Muslims. That is, over the long run the Qing did not simply blame a single Muslim solidarity, as they had during the intense violence, but rather combined that judgment with a wider sense that Muslims are dangerous and must be controlled, for which they could call on a thirty-year tradition of legal precedents. In a lengthy memorial reproduced in the Xun-hua gazetteer, the officials outlined their recommendations, calling upon statutes and regulations ranging from “teaching perverse doctrine” to “concealment of stolen goods.” They blamed the violence on Muslim clergymen and laymen moving from town to town (which Muslims often did for commercial as well as religious purposes), inadequate supervision by the imams (whom they called xiangyue), coercion or adoption of non-Muslims into the Muslim communities, and the impenetrability of Muslim solidarities by officials. For these social ills, they recommended conventional remedies such as guarantees by local leaders, stronger statutory enforcement, a prohibition on adoptions, and careful investigation by local officials.60For the next century and more, they tried these and many other methods, none of which succeeded in ending or even defusing communal tensions in Gansu.
Even in the midst of legal discrimination and military expeditions against the Sino-Muslims in Gansu, the Qianlong emperor still did not entirely abandon the idea that Islam and Muslims could be ordinary and acceptable in the Qing empire. In 1782, a Cantonese Muslim seminarian named Hai Fu- (p.102) run was arrested in Guangxi. The literary inquisition was in full swing, and the prefect impounded all of Hai’s Muslim texts for examination and indictment. The books included Arabic and Persian “scriptures,” which the officials of course could not read, and a number of Islamic apologetic texts in Chinese. Zhu Chun, the Guangxi chief inspector, memorialized that the Chinese texts were “presumptuous and reckless…containing more than a few wild and deviant passages,”61 and he recommended harsh punishment.
In a pair of firm edicts that summer, the Qianlong emperor denied Zhu’s charges, analyzing Islam as a harmless religion, albeit a foolish one (as his father had opined some decades before), and insisting that Muslims deserved the same protection as other subjects of the empire:
The sacred texts which they regularly recite consist of books handed down from of old containing no really scurrilous or plainly seditious language. Furthermore, the phrases in these books to which Chu Ch’un [Zhu Chun] has drawn attention are on the whole crude expressions which cannot be described as violent and rebellious. These are simple, ignorant Hui [Muslim] people, faithful to their religion….
The sacred books which they revere are household knowledge among the Muslims. There is no difference here with the Buddhists, Taoists, and Lamas. Surely they could not be exterminated and their books burned!62
Clearly the emperor had no intention of discriminating against Islam itself, or against law-abiding Muslims in his domains, so he chastised Zhu Chun for overzealousness, and Hai Furun was released unharmed.
Eighty years after this case was submitted, Zuo Zongtang, fighting for the Qing dynasty’s life against the Taiping, Nian, and northwestern Muslim armies, also entered into the perceptual battle over the place of Muslims in China and the reasons for their violent, antisocial behavior. He spent almost five years in Shaanxi and Gansu, then five more in Xinjiang, ending separatist threats to the dynasty and consolidating Qing sovereignty over the entire Northwest. His initial campaign in Gansu destroyed the New Teaching (Jahriyya) headquarters near Ningxia, and Zuo often named that Muslim solidarity and its leader, Ma Hualong, as the root cause of Muslim insurgency and violence:
The reason why the New Teaching must be prohibited is that it claims to be from God and makes ridiculous prophecies. This group’s behavior is very strange and often lures foolish Muslims into willing slavery. The victims often are trapped into conspiracy without knowing how and are even willing to face execution without the slightest regret…. This makes the New Teaching a potential danger to the empire…. [Ma Hualong] healed the sick and granted children to those who prayed for the birth of children…. When the New Teaching is eliminated…then Shaanxi and Gansu can expect to be safe for a hundred years.63
(p.103) But after more years of fighting and with the benefit of hindsight, in a variety of written documents, he named more diverse causes, including the local non-Muslims: “Regarding the disaster of Muslim insurgency for the past eight years and the reason for the commencement of violence in Shaanxi and Gansu, the fault lies entirely with the Chinese [Han].” He was also aware of the importance of tensions within the Muslim communities as causes for violence, as his lieutenants later summarized: “Ma Hualong was a New Teaching Muslim, holding to different religious objectives than those of the Shaanxi Old Teaching Muslims, and they were mutually incompatible [lit., ‘fire and water’].”64
Zuo utilized several modes of distinction in the Qing repertoire to “explain” what had happened in the northwest between 1861 and 1872. His contradictory analyses of the causes of rebellion—that it was the fault of the non-Muslims, that it was the fault of the New Teaching, that it resulted from Muslim rivalries—can be reconciled if we recognize that he had different audiences and different purposes as he described the rebellions to his staff, to his superiors in Beijing, to members of his family in letters, and to posterity in reports written for the permanent record.
We have observed a variety of discursive modes for comprehending difference in Qing official and legal discourse on the Sino-Muslims. The Confucian civilizing project appears in suggestions for deracination, in proclamations of imperial benevolence, in encomiums for Muslims who succeed in the examination system. The discourse of race, on the other hand, dominates anti-Muslim memorials that demand special discriminatory powers for the state to control those congenitally violent people. All of the Qing emperors (some consistently, some not) proclaimed han hui yishi tongren, their vision of a multiethnic empire dominated from the Aisin Gioro center Even after the Qianlong emperor made discriminatory statutes part of the code, he and his successors continued to proclaim that ideal when circumstances permitted.65 Post-1762 analyses of social conflict in Gansu focus on the New Teaching, the Jahriyya Sufi version of Islam, as a locus of wickedness, a pernicious doctrine, which must be rooted out in order for Sino-Muslims to live peacefully in the empire. Finally, officials and emperors sometimes expressed the desire to rid the empire of bad (“weed”) Muslims, who might be known by their doctrine or their behavior, and to keep the good ones in peace and harmony. As Chen Hongmou wrote, there are many Muslims who are “law-abiding, experienced, steady, circumspect, and magnanimous”—that is, they are good people, who deserve His Majesty’s benevolent protection.
Amid all these shifting terms of description and classification, we find constant interplay between benevolence and prejudice, between “separate but equal” treatment and blatantly unequal statutes, tension reflected in the (p.104) behavior of emperors, officials, and the Sino-Muslims themselves. This fluid ethnic discourse shifted in accord with local conditions, regional and central politics, and the ideological stances of crucial actors. If the entire repertoire of modes was always available in theory, certainly the imperatives of office, of local ethnic enmity, of imperial majesty pushed both officials and commoners toward particular choices. In relatively peaceful years and places, proclamations of imperial benevolence could hold sway, but when violence occurred (whether “Islamic,” as in Gansu, or simply perpetrated by Muslims, as in Shandong or Anhui), discriminatory models could be called upon to justify state violence. Local officials and gentry, overwhelmingly non-Muslim, reified the “Muslim” valence of identity as a marker of difference, identifying the Sino-Muslims as uncivilized, whereas emperors could (and did) justify selective coercion or punishment while still invoking han hui yishi tongren.
The materials presented here demonstrate that local officials often desired to isolate Muslims as different and inimical to social order, both before and after 1762, but could be prevented from doing so by the Throne. They also show a clear difference in perception and recommendations among officials serving in north China, in Shaanxi, and in Gansu. From experience in Shandong and Anhui, judges argued for discriminatory statutes; Shaanxi governors and magistrates claimed uniquely violent character for the Hui-hui and blamed religious feuding for disturbances; while Gansu civil and military officials condemned the New Teaching as a font of disorder and requested prohibition. But even in Gansu, local conditions dictated careful distinctions. Zuo Zongtang, who fought from central plain to furthest frontier between 1867 and 1878, did not express a singular, clear opinion, preferring a variety of solutions that included forced resettlement, incorporation into the Qing military, and massacre.66 Zuo did indeed xi hui, wipe out the Muslims, around the Jahriyya base near Ningxia and in the Gansu corridor, as acts not of ethnocide but rather of military contingency, for he left large Sino-Muslim communities intact at Hezhou and Xining.
Sometimes the Sino-Muslims were perceived as Chinese, as when they had to wear the queue or be punished by the same laws as other Chinese. Sometimes they were perceived as non-Chinese, when they became subject to discriminatory statutes or military and political repression for belonging to a prohibited Sufi order Always, of course, they were actually both, as reflected in one of their many Chinese names—the hanhui. The changes in the code make it clear that they were perceived as different, but not as different as the Turkic-speaking Muslims of Xinjiang. After all, they spoke Chinese, took the exams, and served the state as ordinary Chinese did. The Qing administered the Sino-Muslim communities as Chinese, never encouraged a Sino-Muslim elite outside the regular examination systems and appointed no beg s or tusi to rule over them. After the mid-Qianlong period, however, they also (p.105) demanded that the Sino-Muslims’ religious leaders take responsibility for their congregants’ behavior or be punished themselves.
The definition of the Sino-Muslims as a legal category stereotyped them as the wildest, most peripheral members of Chinese society, as sinophone and normal but also dangerous. It was that combination which forced so many twists and turns in Qing legal thinking. Though acculturated, they would not assimilate entirely but worked hard to remain different, which made no sense to many officials. In local systems, the Muslims inspired fear and hatred by alien speech (including the Sino-Muslim “insider’s patois” called Huihuihua), alien customs, and an exclusive in-group spirit. The emperors tried to explain that the Muslims just followed the customs and precepts handed down from their ancestors, but their obvious and often contentious differentness prevented local officials and gentry from granting the Sino-Muslim versions of Chineseness any but peripheral legitimacy67 Dru Gladney has demonstrated the diversity of the hui minzu under the People’s Republic, and my historical research has revealed similar variety among the Sino-Muslims under the Qing. Responding to differing local and regional contexts, Qing emperors and officials could not formulate consistent guidelines for relations between Muslims and non-Muslims, nor codify consistent use of state power to control these next-door neighbors who were also Others.
The author would like to thank Professors Kataoka Kazutada, Saguchi Toru, Ma Saibei, Donald Leslie, and the late Joseph Ford for their painstaking research in the primary sources, work that became a guide through the enormous body of Qing law and official communication regarding Muslims.
(1) . I use the term Sino-Muslim as the most neutral indicator of their hyphenated culture. Other ethnonyms are available, and most foreign scholars are content to call these people hui, as they currently call themselves. But because that term has been defined within the minzu system of the People’s Republic (see below), which would anachronistically project contemporary identities back into the Ming-Qing period, I have avoided it. For the same reason, I have also chosen not to use han to indicate the “majority ethnic group of China” as it is currently defined, preferring rather the cultural term “Chinese.”
(2) . The more civilizable earn the appellation shu, “cooked,” while those less susceptible remain sheng, “raw,” because of their stubbornness and bestial stupidity. Harrell, “Introduction,” in Cultural Encounters.
(4) . Note that all of these abstractions of perceptions also exist in conventional European-American thinking about African Americans, Amerindians, and other minorities, sometimes encouraged by government policies such as segregation, official racial or ethnic identification, compulsory public education, and affirmative action.
(7) . Han Yu’s famous essay on the bone of the Buddha demonstrates the same kind of thinking with regard to Indians and Central Asians in the Tang period. The amalgam of beliefs and practices usually called “White Lotus Teachings” became such a doctrine in the late Ming and early Qing periods, a perception that led to numerous and often sanguinary misunderstandings of social reality. Ter Haar, The White Lotus Teachings in Chinese Religious History, esp. chap. 8.
(8) . Zhu Xi and his neo-Confucian followers held that all human beings are endowed with individual allotments of qi, often translated as “vital essence,” which are naturally clear, turbid, or somewhere in between, depending on an individual’s fate. The purpose of self-cultivation (Ch. xiushen) lies in clarifying one’s own qi in order to allow one’s inherent luminous virtue (Ch. ming de) to shine forth into the world.
(9) . Huizi usually, but not always, refers to the sedentary Turkic-speaking Muslims of Xinjiang, the people now called Uyghurs; occasionally, however, it is used to indicate the Sino-Muslims of China proper (as in Dou Bin’s memorial, below). Donggan (also Tungan and Dungan), which Turkic-speaking Muslims call the Sino-Muslims, derives from a Turkic word of uncertain etymology. It sometimes appeared in Qing documents, and in the twentieth century it has become the official ethnonym for Sino-Muslim refugees who settled in Russian (and then Soviet and now independent) Central Asia.
(10) . I shall not enter here into the debates on defining ethnicity. Gladney, Muslim Chinese, and many other publications have persuasively argued for the creation of the hui as an ethnic group in the twentieth century, and especially under the People’s Republic. Though similar in intent, these Qing characterizations did not suffice to constitute the Sino-Muslims as an empirewide solidarity.
(11) . This has presented particular problems for the People’s Republic. Stevan Har-rell discusses the contradictions and complexities of the minzu paradigm in the introduction to Cultural Encounters, mentioning the Hui as an especially difficult case at 33–34.
(12) . No minority ethnic category encompasses as many contradictions and distinctions as the vast and variegated majority Han themselves, a subject addressed by several of the chapters in this book.
(13) . These range from administrative units (autonomous regions, prefectures, counties, etc.) to research institutes to funding for minzu-specific public education, textbooks, and folksong contests. So successful has this state-sponsored “ethnic mapping” been that Muslims counted among the hui minzu (usually abbreviated to huizu) rarely even write about Muslims assigned to other minzu living inside the borders of China (e.g., Uyghurs, Salars, Uzbeks, etc.). Rather, they publish books entitled Biographies of Famous [Members of the] Huizu, Essays on Huizu History, and Modern Literature of the Huizu.
(14) . We must also note here the importance of non-Muslim hostility as a device for the maintenance of solidarity among the Sino-Muslims. It is always easier for minorities to stick together if the surrounding majority has proved itself to be ethnocentric and dangerous. The linguistic and cultural adaptation of the Sino-Muslims to Chinese ways does not in itself distinguish them from Muslims in India, West Africa, Indonesia, or other regions to which Muslims have migrated and then settled, intermarried, and acculturated.
(15) . The Sino-Muslims themselves also disagreed over what their proper place in China might be, with options ranging from self-segregation to thorough immersion in the surrounding culture. Lipman, “Hyphenated Chinese.”
(16) . Wakeman, The Great Enterprise, vol. 2, 825–26. The original text may be found in Xie Guozhen, Qingchu nongmin qiyi ziliao jilu (Historical materials on peasant uprisings in the early Qing), 282.
(17) . Leslie, Islam in Traditional China, 122, citing an unpublished translation by Joseph Ford.
(18) . Kataoka Kazutada, “Shinchō no kaimin seisaku no saikento” (Further investigation of Qing policy toward Muslims), 64.
(19) . Fu Tongxian, Zhongguo huijiao shi (A history of Islam in China), 116.
(20) . Here the emperor might be confusing Chen Shiguan’s Shandong jurisdiction with the “Western Regions,” for the Muslims about whom Chen complained were all sinophone.
(21) . Leslie, Islam in Traditional China, 123–24, citing Ford’s translation.
(22) . Leslie, Islam in Traditional China, 125, citing Ford’s translation.
(23) . Chen Hongmou, PeiYuan tang oucun gao, 30.13b. The text of the covenant extends from 30.13a to 30.22a. I am very grateful to William Rowe for introducing me to this important document.
(26) . Leslie, Islam in Traditional China, 126, citing Ford’s translation.
(27) . This is certainly not the first distinction between Muslim and non-Muslim in Qing law, for the regulations regarding frontier areas had long distinguished between Turkic-speaking Muslims and other Qing subjects in Xinjiang. But this change applied to cultural China, not the frontier territories.
(28) . Qing renzong shilu (Veritable records of the Jiaqing emperor), 362.9b–10a.
(29) . Qing xuanzong shilu (Veritable records of the Daoguang emperor), 23.20b-21b.
(30) . Qing wenzong shilu (Veritable records of the Xianfeng emperor), 107.16b–17a. Li was governor of Anhui and Yuan a very successful and outspoken censor on temporary assignment as an anti-Taiping militia organizer. He was also the adoptive great-uncle of Yuan Shikai.
(31) . Like similar perceptions of minorities elsewhere (of African Americans, for example, in the United States), the notion that Sino-Muslims are violent people has some basis in reality—some Sino-Muslims actually were—but does not similarly judge non-Muslim Chinese (the majority) to be inherently violent despite the fact that many of them were violent, too. In the rescript cited above (note 24), the Jiaqing emperor recognized the contradiction.
(32) . Qing gaozong shilu (Veritable records of the Qianlong emperor), 676.22a. Min Eyuan did not regard only Muslims with distrust, for a few years later (as provincial treasurer of Hubei) he voiced similar sentiments regarding wandering Buddhist and Daoist clergy. Kuhn theorizes analogies among fear of “the uncontrolled movement of rootless people,” fear of sorcery, and fear of strangers. The same can certainly be said of fear of “different” people, such as Muslims, especially when those Others have a diabolical reputation as violent people and a propensity for sticking together in a brawl. He credits much of the soul-stealing panic of 1768 to the particular fear of aliens/outsiders in Chinese society, a fear that could attach as easily to Muslims, even Sino-Muslims who had been neighbors for years or generations, as to sorcerers, wandering monks, or beggars. Philip A. Kuhn, Soulstealers, 44–45, 114–15.
(33) . Da Qing lüli xinzeng tongzuan jicheng (The revised compilation of the Qing code) (1875 edition), 24. qiedao. 6b-7a.
(34) . Kataoka Kazutada, “Keian shiryo yori mitaru shincho no kaimin seisaku” (Qing policy toward Muslims as viewed in criminal cases), 4. Chen Hongmou had made the same arguments in 1751, including the necessity of more severe penalties for Muslim criminals, but they had not been adopted into the code.
(35) . Da Qing lüli xinzeng tongzuan jicheng, 25. daozei wozhu. 9b.
(36) . Chen Hongmou had noted this possibility in his covenant, at 30.16a–17a. He also recorded the local name for Muslim cat burglars—Hezhou ghosts (Ch Hezhou gui)—making explicit the connection between “strange and different” Muslims from distant Hezhou (a Muslim center in southwestern Gansu), the spirit world, and antisocial behavior, exactly as Kuhn connects these same elements in Jiangnan. Chen’s investigations revealed that many of the thieves who bore this Shaanxi sobriquet were in fact local men feared as if they came from elsewhere.
(37) . Kuhn, Soulstealers, chap. 10, concludes his study of the soul-stealing episode(s) of 1768 with dark foreboding, even in the midst of the Qianlong period “age of prosperity.” This essay’s analysis of anti-Muslim legal and political action, beginning in the 1750s, can amplify and diffuse that same sense of insecurity and anxiety in mid-Qing north and northwest China, as well as Kuhn’s Jiangnan. Instead of wandering beggars, Daoists, and Buddhists, officials such as Chen Hongmou and Min Eyuan focus their loathing and legal remedies on Muslims—criminals, members of the New Teaching, huihui in general.
(38) . Tusanga’s memorial is summarized in the Qing gaozong shilu, 928.27b-28a.
(39) . Gu Jiegang noted this distinction in “Hui han wenti he muqian yingyou gongzuo” (The Hui-Han problem and our current work).
(40) . Kataoka, “Keian shiryō yori mitaru Shinchō no kaimin seisaku,” 13–14. Citing a personal communication rather than a published source, Kataoka notes elsewhere that the tattooing of Muslims criminals specifically as Muslims was sometimes done in Manchu as well as Chinese characters (“Keian shiryō yori mitaru Shinchō no kaimin seisaku: Hosetsu,” 140).
(41) . Xue Yunsheng, Duli cunyi (Questions on statute law), 3.21b, where this “Muslim” crime is included in a long list of sentences that may not be overturned by filial mitigation.
(42) . Xing’an huilan (Collected criminal cases), 3.1b-2b, describing an 1823 case from Shaanxi.
(43) . Xing’an huilan, shou. 36a refers to the former, and shou. 52a to the latter.
(44) . Xing’an huilan xubian (Supplement to ‘Collected criminal cases’), 3.13a-b.
(45) . Duli cunyi, 33.13a; and also Da Qing lüli xinzeng tongzuan jicheng, 27. dou’ou.6b, 8a, 8b-9a.
(46) . Chen Hongmou, PeiYuan tang oucun gao, 30.20a.
(47) . As noted above, Chen Hongmou had by 1751 already found such internecine feuding to be one cause of the Shaanxi Sino-Muslims’ criminality. I have not yet located any sources that identify the leadership, nature, or even the names of the solidarities to which he refers.
(48) . The rule by which a plaintiff may receive the penalty which the defendant would have received if guilty is called fanzuo. Its makers intended it as an incentive against false accusation and spurious litigation.
(49) . Xunhua zhi (Xunhua gazeteer), juan 8, cited in Qinghai minzu xueyuan minzu yanjiusuo (ed.), Salazu shiliao jilu (Collected historical materials on the Salars), 91–92.
(50) . Ma Tong, Zhongguo Yisilan jiaopai yu menhuan zhidu shilue (The history of China’s Muslim solidarity and menhuan system), 223–24.
(51) . Sino-Muslims customarily use the term ahong (from Per. akhund) to denote an ordinary religious professional or teacher.
(52) . Mu Shouqi, an important chronicler of the Northwest, does not hesitate to call the motivations of the litigants entirely economic, though he does also mention the pleasure that Gansu Muslims took in the innovative practices associated with Ma Laichi’s tariqa: “The newness [fell on their] eyes and ears, and in a moment, with one accord, they all followed it.” Gan Ning Qing shilue (Outline history of Gansu, Ningxia, and Qinghai), 18.37b-38a.
(53) . Nakada Yoshinobu, Kaikai minzoku no shomondai (Some problems regarding the Hui people), 88. This “Bright Sand” business, quite common in Chinese-language accounts, derives from the Chinese characters used in transliterating the Arabic name of Ma Laichi’s text, the Mingshale (Ar munshar), sometimes also called the Mingsha-jing, the Bright Sand Sacred Text. Some versions of this title use the characters for “bright sand” (ming sha), while others use a different mingto indicate the sinister “dark sand.” The Arabic name actually means “the saw,” referring to the rhythmic sound some Sufis made while chanting, and the sand, whether bright or dark, derived entirely from the transliteration. Ma Yinghuan clearly took advantage of his non-Muslim audience’s ignorance of Islam and of the Arabic language.
(54) . The statute regarding xiejiao, heretical teachings, may be found in Chinese and in translation in de Groot, Sectarianism and Religious Persecution in China, vol. 1, 137–47, with the main statute (lü) at 137–38.
(55) . Chen Hongmou, PeiYuan tang oucun gao, 30.19b-20a.
(56) . Both types of practice were common in Middle Eastern Sufi communities, and their coexistence did not usually lead to conflict. Naqshbandis generally favor the silent repetition, but many other orders vocalize, chant, sway, and dance. The famous “whirling dervishes” of Turkey are Sufis who dance their devotions.
(57) . Since the Qing based this nomenclature on time of arrival in Gansu, rather than any doctrinal understanding, it could easily change as circumstances demanded. The Jahriyya/New versus Khafiya/Old division remained in Qing official discourse for almost a century, but by the late 1800s, other innovative groups received the xin-jiao label (or its variants, xinxinjiao and xinxingjiao), while the Jahriyya joined the legitimate fold of the laojiao.
(58) . Though the case materials do not make it clear, the first killing was probably also handled under the statutes for non-Chinese, since collective brawling and manslaughter both carry much heavier sentences in the conventional code.
(59) . The texts do not reveal which laws these officials intended to invoke. They seem to have meant a generic public order, which they perceived the New Teaching as opposing, and perhaps also the statutes against heterodoxy.
(60) . Qinghai minzu xueyuan, Salazu shiliao jilu, 100–103.
(61) . Zhu’s memorials may be found in the collection Qingdai wenziyu dang (1938), cited in Ma Ruheng, “Cong Hai Furun anjian kan Qianlong dui Huizu de tongzhi zhengce” (The Qianlong emperor’s policies for control over the Hui minzu as seen in the Hai Furun case), 8–12.
(62) . The edicts may be found in the Qianlong shilu, and are cited by both Ma Ruheng and Leslie, Islam in Traditional China, 128, from which this translation is taken.
(63) . Wen-djang Chu, The Moslem Rebellion in Northwest China, 157–58.
(64) . Yi Kongzhao, “Pingding Guanlong jilue” (A record of pacifying Shaanxi and Gansu), vol. 4, 8. For a general summary of conflict within Muslim communities and its relationship to Muslim uprisings, see Gao Zhanfu, “Guanyu jiaopai zhi zheng zai Qingdai xibei Huimin qiyizhong xiaoji zuoyong de tantao” (The negative functions of factional struggles during the righteous uprisings of the Hui people under the Qing), 245–62.
(65) . Kataoka, “Keian shiryō yori mitaru Shinchō no kaimin seisaku,” constructs a simple, linear periodization for this evolution: the Kangxi and Yongzheng emperors opposed their local officials, while the Qianlong emperor (and, by implication, his successors) accepted their judgment of the Muslims as depraved and perverse. As we have seen above, this masks the complex realities of center-province relations and ethnic politics.
(66) . Zuo’s campaigns are narrated in Chu, Moslem Rebellion; and in Liu and Smith, “The Military Challenge.”
(67) . In Kuhn’s (Soulstealers, 5) narrative of soul-stealing, the fact of common Chi-neseness did not stop Jiangnan peasants and urbanites alike from assaulting (sometimes murdering) suspicious outsiders, even if their speech marked them as neighbors from nearby counties. If a dozen miles and a county line sufficed to constitute “stranger,” how much more so would outsiders’ perceptions of an exclusive, coherent in-group identity such as huihui and its attached reputation for bellicosity.