The protests that swept through Tibet in early 2008 and the subsequent wave of self-immolations amongst Tibetans raise familiar questions about state-society relations and the ways in which China treats its ethnic minorities. Oppositions between state and society, officials and citizens, and Chinese and Tibetans are highlighted by such events; they apparently justify the paradigm of domination and resistance, used to characterize state-society relations in much writing on China's minorities (Bulag Reference Bulag, Perry and Selden2003; Harrell Reference Harrell2001; Rossabi Reference Rossabi2004). This paradigm is not, however, wholly satisfactory to describe the complex relations found in eastern Tibet.Footnote 1 At most times, strategic interactions and cooperation characterize relations between the minority populations and government officials here, blurring the boundaries between state and society, and government and citizen. This paper explores these dynamics by focusing on the issue of order and how it is maintained, a matter of particular concern to the Chinese government in its Tibetan areas.
My case study is based upon ethnographic fieldwork in Gansu and Qinghai Provinces, presenting a perspective from both the Tibetan pastoralists and members of local government.Footnote 2 It highlights the limits of the state's capacity to provide a basis for order and stability and, indeed, to control processes of local administration. It is not so much that state domination is successfully resisted by the minority populations: despite the profound mistrust with which many of them regard the state and its agents, Tibetan tribal and religious leaders negotiate constructively with local officials over processes of conflict resolution. Rather, the state's ideological position is undermined by its own government officials: the complex and conflicting demands placed upon them mean that state sovereignty has to be negotiated and shared with alternative, historic sources of authority amongst the Tibetan populations.
State, Government, and Party in China
Abrams (Reference Abrams1988, 68) has famously pointed out that the state is “an ideological thing; a device in terms of which subjection is legitimated; it presents politically institutionalised power to us in a form at once integrated and isolated; it creates an acceptable basis for acquiescence.” In China the state as an ideological entity is ubiquitous. It is symbolized by the national flag and anthem, in public ceremonies, and the prominence of state officials and glorification of public projects in the media, and it is reflected in the organization of urban space, with its ranks of imposing, glass-fronted public buildings (Saich Reference Saich2001, 13). According to Shue (Reference Shue, Hays Gries and Rosen2004), the ideals of truth, benevolence, and glory, which characterized the imperial ideology of government, found the ideals of the modern party-state: the ideal government preserves a stable social order in which people enjoy the benefits of economic progress and development, share a nationalist spirit, and live in social harmony.
As Lieberthal (Reference Lieberthal2004, 203) describes, “China's leaders have in the past taken it as an article of faith that they understand the needs of the population better than the citizens themselves.” Government and party accordingly retain a great deal of organizational control (Lieberthal Reference Lieberthal2004, 232–33; Perry and Selden Reference Perry and Selden2000, 7; Saich Reference Saich2001, 204–5; Yang Reference Dali2004), and in the last decades of the twentieth century they have promoted remarkable economic progress. The modernizing agenda has done little for China's minorities, however, instead increasing their experience of inequality and relative poverty (Fischer Reference Fischer2004). The Western Development Initiative, launched in 1999 to address such concerns, treats the problems of these areas as those of underdevelopment, but despite massive state investment and apparent economic growth, this has failed to quell separatist sentiment in either Tibet or Xinjiang (Shih Reference Shih2004; Wong Reference Wong, Bird and Ebel2007). Potter (Reference Potter2005, 294) suggests that in China's peripheral areas “the continued influence of Confucian patrimonial sovereignty along with concerns over national unity, wealth, and power has supported ideas and practice of governance that remain hierarchical and authoritative.” Despite the ideals of truth, benevolence, and glory, state sovereignty is apparently exercised through authoritative and dominating force.
As Hansen and Stepputat (Reference Hansen and Stepputat2005, 3–7) maintain, the sovereignty of any state is inevitably enacted by “internally fragmented, unevenly distributed and unpredictable configurations of political authority, [which] exercise more or less legitimate violence in a territory.”Footnote 3 Sovereign power, by their account, is “made real and reproduced through ritualized, everyday confirmations of . . . violence”; it is not based on forging consent, but on securing subjection, order, and obedience through the performance of paramount power and the suppression of competing authorities.
In Amdo, as the northeastern part of Tibet is locally known, considerable violence was used to bring an end to the 2008 protests, coercive power in the form of prison sentences is used to stifle dissent on sensitive topics, and violence is commonly used during everyday police activities. Policies on birth control and fencing of the pasturelands are forcibly enacted and deeply resented by the local populations. Such dynamics recall the exercise of “dominance without hegemony,” which characterized British colonial rule in India (Guha Reference Guha1999). As I describe in this paper, however, government officials at county and township levels in Amdo are constantly engaging with local populations and promoting relations of nonviolent cooperation.
Such observations are not without precedent. In a study of the state's policies on religion, Potter (Reference Potter2003, 327) describes the state's dependence on compliance: “Party policies grant limited autonomy for acceptable practices while attempting to repress activities that challenge political orthodoxy.” As Hillman (Reference Hillman2005) also found, local officials find it prudent to coopt religious (Buddhist) leaders in local policy making. This enables them to build legitimacy among the local populations and to deliver economic growth and social stability to their superiors. However, attempts to coopt religious leaders present a double-edged sword to government officials because, as he describes, they can increase the local monastery's autonomy and assertiveness. As Potter (Reference Potter2003, 332) also notes, “to the extent that its enforcement of policies . . . appears to contradict the accepted balance between autonomy and loyalty the regime may undermine its own legitimacy.” This paper examines similar, seemingly contradictory dynamics in eastern Tibet and asks about the implications for the Chinese state.
Amdo
The area referred to by Tibetans as Amdo is roughly the size of France and occupies the northeastern part of the Tibetan plateau, comprising the majority of Qinghai and parts of Gansu and Sichuan Provinces. The Amdowa (people of the region) follow Tibetan Buddhism, long regarding Lhasa as the ultimate source of religious authority, and consider themselves to be distinct from their Han Chinese and Hui Muslim neighbors to the east. They were never incorporated within the political regime of the Dalai Lamas, however. The majority of Amdo consists of grassland, occupied by nomadic pastoralists who have historically traded their animals with the farmers of the valleys along the northern and eastern edges of the plateau. Although economic developments have allowed many to obtain education and jobs within the new towns, most of the Tibetan population still depends on livestock for both subsistence and income.Footnote 4
Amdo was dominated by Mongol forces from the mid-thirteenth until the early eighteenth centuries, when the Manchu Qing dynasty rose to power and established Xining, a town to the north of Amdo (now the capital of Qinghai Province) as its administrative base for the area. The rule of their senior official, the Amban, was light, however, and allowed relative autonomy to be retained by the monasteries and other local rulers (Goodman Reference Goodman2004). The power of the Qing waned in the early twentieth century, and their control over Amdo was replaced by that of the Hui Muslims, whose administrative influence was likewise largely confined to the north and east of the region.
By this stage, a number of local rulers had emerged in Amdo. The Tibetans were largely organized into tribal groups, often consisting of several thousand nomads, but monasteries such as Labrang, Rongwo, and Taktsang Lhamo appointed headmen, gowa (’go ba), to the tribes within their areas (Ekvall Reference Ekvall1939; Rock Reference Rock1956).Footnote 5 The most powerful of these monasteries, Labrang (bla brang), was founded in 1709 by the reincarnate lama (bla ma) Jamyang Zhepa (’jam dbyangs bzhad pa), and the monastery sent gowa to its tribes for periods of three years. Their duties, according to my informants, were to approve the headmen of the smaller groups, to arrange for monks to carry out religious rituals in the area, and to “sort out the problems” among the nomads. By this they meant resolving cases of theft, fighting, and killings. It does not appear that any substantial taxes were levied, although the nomads offered substantial tributes to the monasteries.
In some areas, secular leaders achieved the status of kings, similarly presiding over a group of tribes (Carrasco Reference Carrasco1959; Ekvall Reference Ekvall1939), while others had their own hereditary gowa (Hermanns Reference Hermanns1949). In Golok, a large area to the southwest of Amdo, several tribes formed a powerful confederacy under hereditary ruling families. The Golok nomads were notoriously independent of centralized control, and numerous accounts tell of their pride, their warlike qualities, and the threats they posed to travelers and neighboring tribes alike (d'Ollone Reference D'Ollone and Miall1912; Kozloff Reference Kozloff1908; Namkhai Norbu Reference Norbu1997; Rock Reference Rock1956).
Robert Ekvall, who lived in Amdo throughout the 1930s and ’40s, describes the overall pattern as one of alliances, warfare, feuding, and overlapping spheres of political, military, and religious influence (Ekvall Reference Ekvall1939, Reference Ekvall1964, Reference Ekvall1968). He also describes well-established practices for the mediation of conflict between tribal groups. Once a death had been inflicted, in response to a raid for example, retaliatory killing could only be avoided by the payment of appropriate compensation or blood money. Large feuds were mediated by high-status individuals, whose power was based on their positions as chiefs, their skills of oratory, or, in the ultimate case, their charismatic authority as reincarnate lamas (Ekvall Reference Ekvall1968, 78–79). Nevertheless, as he also emphasizes, even the lamas’ power was persuasive rather than coercive, and these processes were by no means invariably successful.
In the majority of Amdo, therefore, save those areas closest to Xining, the Tibetan people had no direct experience of state-like control. Kozloff, a Russian explorer who traveled through the area in 1900, recalls the boast of a Golok nomad in the following terms: “You obey the laws of strangers, the laws of the Dalai Lama, of China, and of any of your petty chiefs. . . . We N'goloks, on the other hand, have from time immemorial obeyed none but our own laws, none but our own convictions” (1908, 522–28). These Tibetans felt antipathy towards the surrounding polities and their rulers and laws. While certain individuals, such as the Qing Amban at Xining and the family of Jamyang Zhepa at Labrang, therefore, exercised a limited administrative authority in some areas, there is no evidence of the exercise of close control over the lives of their subjects or the imposition of rules.Footnote 6 Religion and tradition—in the case of hereditary leaders—legitimized the exercise of power, and nothing approached the control now exercised by the Chinese state. A different and well-established set of expectations and practices, including principles of revenge and compensation, generated instead a precarious form of order.
Tibet and China
When China seized control of eastern Tibet in 1958, the power of the monasteries was destroyed, all local leaders were deposed, and all pastoral activities were collectivized.Footnote 7 Like others, I found that local Tibetans were almost universally reluctant to talk about their experiences at this time and during the Cultural Revolution (cf. Makley Reference Makley2007). They generally referred to a period of fighting, during which many perished, to the closure of the monasteries, and to the ban on religious activities: many monks were forced to take jobs and wives or else were imprisoned or executed, while many laypeople were also persecuted for religious practices, often simply disappearing in police custody. Their families continued to herd yaks, sheep, and horses on the grasslands but within the structure of new communes, only being allowed limited food rations. Many talked of the hunger they endured. Other researchers describe the “struggle sessions,” during which the Tibetans were forced to “speak bitterness” against each other (Makley Reference Makley2005), and the persecution of former leaders as “class enemies” (Goldstein Reference Goldstein, Barnett and Akiner1994, 95–96). Two generations later, the memories of this period remain a barrier to the state's attempts to establish legitimacy among the Tibetan populations.
The reforms that swept China after Mao's death under Deng Xiaoping reached Amdo, along with central Tibet, in the early 1980s (Goldstein Reference Goldstein, Barnett and Akiner1994, 96–100). The major change was the introduction of the Household Responsibility System, under which livestock was redistributed amongst the nomads’ tents (Horlemann Reference Horlemann and Huber2002, 252–57). Once restrictions on the sale of animals were lifted, the nomads became almost fully autonomous again in their pastoral practices. In 1999, following two decades of what was regarded as poor economic performance, a new development strategy, the Western Development Initiative, was introduced (Goodman Reference Goodman2004; Shih Reference Shih2004). This has led to considerable infrastructure development throughout China's western provinces. A network of roads has been established on the Amdo grasslands, linking the cities and major towns with smaller administrative centers. The area is divided, for administrative purposes, into provinces (Qinghai and Gansu), prefectures, counties, and townships, and even the smallest contains a handful of shops, restaurants, and government offices. The townships were, however, largely mapped onto the tribal groups that existed before 1958 (Clarke Reference Clarke, Ihara and Yamaguchi1989; Levine Reference Levine1999), and Tibetans on the grasslands insisted to me that their social organization, including their patterns of feuding and mediation, follow substantially the same patterns and principles as they did previously.Footnote 8
I spent ten months in Amdo, around eight of these living with Tibetan nomads on the grasslands, largely based in one tribe, but traveling around to conduct interviews throughout a wide surrounding area.Footnote 9 While there are significant local variations, the general patterns, attitudes, and views that I describe in this section seem to be widely shared on the grasslands.
The nomads’ tribal groups, dewa (sde ba), may consist of several thousand people and are, themselves, divided into sub-tribes or encampments, which might consist of forty tents, that is, around 200 people.Footnote 10 Each is under the charge of one or more headmen, gowa, selected by the people. Their duties are to coordinate pastoral movements, allocate summer grazing land, negotiate with local authorities, organize ritual events, and resolve local fighting and disputes. A council of these gowa takes charge of the affairs of the dewa; they carry out similar tasks here, organizing religious activities, liaising with the county administration, and resolving conflicts. Some are recognized as representatives of their township and required to join the Chinese Communist Party, although this affiliation is regarded as sufficiently meaningless not to undermine their authority within their own groups.
The introduction of education and health services and, more recently, developments designed to stimulate economic development are largely regarded as beneficial. The network of roads brings markets and consumer goods—affordable trucks and motorbikes, new clothing in place of cumbersome sheepskins, and convenient household goods, such as milk separators—within easy reach of the nomads’ tents and has allowed some to engage in more extensive trading activities. Education opportunities have allowed a few to find jobs in the governmental administration, as officials, clerks, teachers, and health workers, and even in the local police force. The biggest changes have occurred in the urban areas, where the new educated classes now find themselves competing for jobs and educational opportunities with their Han and Hui neighbors (Fischer Reference Fischer2004), but on the grasslands the material changes are, for the most part, welcome.
At the same time, government officials from a bewildering variety of different departments and party units maintain registers, levy taxes, devise vaccination programs, draw up and enforce fencing strategies, and enforce birth control policies. The livelihoods of the vast majority who still practice nomadic pastoralism have, for example, been profoundly affected by a policy to divide the pastureland among individual families on fifty-year leases. The official aim is to promote ecological sustainability and resolve the perceived problems of overstocking and pasture degradation (Breivik Reference Breivik2007; Horlemann Reference Horlemann and Huber2002; Yeh Reference Yeh2003). However, as I have described elsewhere (Pirie Reference Pirie2005a), most nomads deeply resent the new policy for interfering with the flexibility of their pastoral movements and creating divisions within their tribes. They also begrudge official insistence on livestock vaccination programs, which they do not regard as having any practical benefit.
The nomads are indignant about the policy of restricting family size to three children, which is enforced unevenly amongst the rural Tibetan population. Another ambitious policy was introduced in the 2000s to sedenterize nomadic families, particularly in Golok, which has forced a number of them to move to newly constructed houses on the outskirts of the towns (Goodman Reference Goodman2004; Horlemann Reference Horlemann and Huber2002, 258–60). Here they are supported, for a few years, by government grants, but there are next to no opportunities for employment, so drinking, gambling, and fighting have come to occupy their time.
For most of the nomadic pastoralists I talked to, restrictions on freedom of information and speech do not impinge daily upon their lives. However, almost all either have memories of the violent way in which religious practice was suppressed during the Cultural Revolution, or they have heard detailed accounts of these events. Moreover, there are still considerable restrictions on the numbers of monks; all contact with the Dalai Lama is banned, and Tibetans are regularly jailed if they are thought to have breached this rule. Young people still make the arduous and illegal journey to India and, if they return, remain the object of suspicion by the authorities, with restrictions on their employment possibilities. The more educated are well aware that their access to information in the media, including the Internet, is severely restricted.
The Experience of the State
Tibetans use a number of different terms to refer to the state and its agents and representatives. Jia (rgya), an abbreviation for rgya nag (the Tibetan term for Chinese people) is frequently heard. The period before the watershed represented by the events of 1958, for example, is referred to by the phrase jia ma yong na (rgya ma yong nas): “before the Chinese came.” The term jia (sometimes the Chinese zhong guo) is also used to express a general distinction between us, wod (bod), Tibetans, and them, Chinese (Makley Reference Makley, Pirie and Huber2008). The memories of the communal period and the atrocities of the Cultural Revolution remain strong, and this phrase is a constant reminder of their legacy.
In their daily lives, on the other hand, the Tibetans are more closely affected by the activities of government bureaucracy, which they generally refer to as zheng fu, the Chinese word for government—it is significant that there is no obvious Tibetan equivalent—or by referring to the names of the towns in which their offices are located. The police, go nyen chi (or jing cha, the Chinese term), are largely treated with suspicion and mistrust given their propensity, as my informants said, for beating people up and taking money.
As regards the more visible and obvious symbols of the state, flags, speeches, and slogans are ubiquitous, particularly in the schools and towns. Children have to sing the national anthem, and their textbooks are decorated with images of children saluting the national flag; in the towns there are often public gatherings presided over by government officials. Under the political reforms introduced in the 1990s there is more support for cultural autonomy, and the government now sponsors horse racing—an activity that is very popular throughout the pastoralist areas—and performances of the Tibetan epic, “Gesar of Ling,” which concerns a local hero who becomes a great Buddhist leader. Such events are prefaced by speeches from officials, who have processed from the nearest town and appear on specially constructed stages, which could be regarded as a deliberate attempt to undermine the potential for these performances to stir up nationalistic sentiment. Nevertheless, the performances are enthusiastically attended by the local population.
At some events, local Tibetans from government offices are required to participate in the dancing or to appear in traditional dress. Occasionally, including at the music festival mentioned later in this article, the audience is made to wear matching colored clothes. This is a format often seen on televised festivals, where an audience dressed in identical shirts waves identical light-sticks at appropriate moments. “The people” are, thus, made to appear as a unified body, whether simply as Chinese citizens or as members of the Tibetan minority. As Hansen and Stepputat (Reference Hansen and Stepputat2005, 8–9) describe it for eighteenth-century Europe, the development of a notion of popular sovereignty prompted attempts to represent “the people” in mass spectacles, “as an objectification of the general will.”
Such symbols of the state's supremacy over an obedient and compliant populace are, however, largely meaningless for the Tibetan nomads. The state ideal in its most obvious manifestations, the flags, parades, and organization of space, represents, for them, the power and presence of the Chinese, jia. Moreover, although the government, zheng fu, is credited with improvements in lifestyles, education, health care, and job opportunities, its agents constitute the same body of officials who enforce the resented restrictions on their livelihoods. Even more importantly, these same officials are not regarded as being responsible for maintaining social order, primarily because the Tibetans regard them as powerless to do so.
Maintaining Order
The nomads talk frequently and readily about both actual and potential violence, and most men prominently carry knives. They talk as if their neighbors were always just about to attack or steal from them, and everyone could tell me stories of fighting and killings that had occurred within their families or encampments within the last few years. They explained that they “have” to take revenge in the event of theft, and if a member of their family has been killed they must retaliate against the perpetrator's family. When such a conflict occurs within one tribe, or between two that normally maintain good relations, there are relatively informal resolution procedures and considerable social pressure is placed upon the disputants to agree to a settlement. Once retaliatory killings have begun between persons or groups from different dewa, however, a feud can only be resolved through procedures of mediation. Monks may intervene to establish a temporary truce, and gowa from a neighboring tribe or senior monks from a local monastery are called in to act as mediators, zowa (gzu ba). A meeting is set up in a neutral place, now often the local town, with different groups in different rooms and the mediators acting as go-betweens. Their task, they told me, is to determine the appropriate level and nature of compensation. Injuries and death must always be compensated for by the payment of the value of the life, mnyö rtong (mi stong), apologies may be made by offering religious books, and the mediators may establish where boundaries should run. Crucially, they have to convince the parties of the justice, jömdri (rgyu bras), of the proposed solution.Footnote 11
The most intractable disputes require zowa with “good speech,” and in Golok certain xhombo (dbon po), men from the local ruling families, are particularly noted for their skills in oratory and diplomacy. In other areas, today as previously, appeal is made to the lamas. The highest reincarnates from Labrang, Jamyang Zhepa and Gongtang Rinpoche (gung thang rin po che) (until his death in 2000), were seen as able to resolve the most problematic conflicts. My informants said they would always tell the truth in front of such figures and that the lamas suggest just solutions (jömdri), taking into account the history of the case.Footnote 12
The police always attempt to track down and imprison those responsible for murder and, if they can, they arrest and punish those involved in fighting. The state's criminal punishments are, however, regarded by the nomads as wholly inadequate to resolve the resulting feuds. In one relatively typical case, two brothers had become embroiled in a fight with some men from a neighboring group after their yaks had strayed. In the course of the fight, one of the neighbors was killed. The police imprisoned both brothers, one for a few months, the other for eight years. Mediation by the local gowa was still necessary, however, in order to avert further fighting: the killers’ family had to pay compensation to the victim's family and was expelled from its group for three years.
As far as the Tibetans are concerned, the government's system of criminal punishment (the “Chinese” system) is completely inadequate for settling their disputes. “They cannot produce just solutions (jömdri). They do not care about the history,” people told me dismissively. Once fighting has erupted, the norms of retaliation must be followed and the police, if they refuse to engage with these norms, have no authority to restore order. The nomads, in this way, deny legitimacy to the state's criminal justice system. It was apparent from these accounts that local officials are effectively forced to acquiesce in and support the nomads’ forms of dispute settlement. They are often present at settlement meetings, they are certainly aware of them, and they may even call upon senior lamas or xhombo to become involved. The xhombo in Golok expect to receive invitations from the authorities after a serious incident, they told me.Footnote 13 Thus, a dual system operates, whereby a wrongdoer may have to suffer double punishment, both imprisonment and payment of a large fine. The injustice of this situation was explicitly recognized by one of the Golok mediators, who told me that disputants have to accept that they are subject to two systems. Using tamxhwe (gtam dpe), a form of speech for which the Golok mediators are famous, he explains to them that a man subject to two laws is like a horse with two saddles.
There is, therefore, a major clash of expectations about how order can be maintained in these areas. The Tibetan nomads I talked to were insistent that their system of compensation was morally superior to the government's punishments (“more progressive” was one phrase), which do nothing to help the victims. “What is the use of taking another life as a means of punishment?” they ask. There are also opposed views about who has authority to dispense justice and how. The power exercised by the mediators is primarily persuasive, and my informants stressed the precariousness and difficulty of the processes of mediation. The mediators have to bring the parties to agreement, not having the power and authority simply to impose a solution upon them, as Ekvall (Reference Ekvall1964) also noted in the premodern period. The ideal is someone who can “take responsibility for the safety of others without presuming on those others’ prerogatives,” as Dresch (Reference Dresch1984, 45–46) also noted of the Shayks among Yemeni tribes. This ideal is quite different from that of the Chinese legal system, under which punishments are imposed for deviant behavior.
The Position of the Government Officials
The lower-level officials in the county and township governmental bureaucracy are often Tibetans from local families, as are the local police. Nevertheless, for the nomads on the grasslands the government bureaucracy, and those who staff it, are “them,” part of the state's powerful and controlling edifice. In a similar way, such officials talked about how they deal with “the nomads.” Structurally, the most important dichotomy is between officials, those who work in government offices and are paid government salaries, and those who do not.
Such officials are part of an immense bureaucracy, within which the need to satisfy the demands of his or her superiors is crucial to any individual's career prospects. Educated, urban Tibetans were scathing about the culture of patronage that permeates all levels of the administration. Sitting in one of the hotels in a county town, for example, some of them identified a group of government officials entertaining one of their superiors and, with contempt, described how visiting officials, arriving with a large retinue, would expect constant entertainment; an evening's hospitality could easily cost ¥10,000 (€1,000), including payment for expensive spirits and the sexual favors of local girls. They told me of an education official from Beijing who promised an extra ¥60,000 for the county at the end of a visit that had already cost his hosts ¥150,000. Of course, not all officials are ambitious enough to engage extensively in such activities, but the culture of patronage permeates the administration. As Liu and Tao (Reference Mingxing, Ran, Shue and Wong2007, 177) point out, the system of administrative appointments and political accountability remains highly centralized and the performance of local government officials is continuously monitored from above:
In less developed regions, lower factor mobility, together with the relatively tougher-to-meet central development mandates and higher dependence on upper-level transfers work systematically to distort local government behavior and draw local governments in these regions into a style of political performance that caters to higher levels of government instead of local people and encourages unhealthy competition for upper-level transfers.
The maintenance of peace and harmony is a very important aspect of the political performance demanded of local officials in Amdo.Footnote 14 In order to achieve it, however, such officials frequently have to allow, or even encourage, Tibetan forms of mediation. In the 2000s, for example, tensions arose between the Tibetan and Muslim populations of Amdo, leading to boycotts and, often, to violence (Fischer Reference Fischer2004). When I visited the area in 2008, local Tibetans told me how, a year previously, rumors that a fingernail had been found in a bowl of soup in a Muslim restaurant in Dawo, the capital of Golok Prefecture, had led to a violent attack on the restaurant. A crowd of Tibetans had gathered, shouting, throwing missiles, and eventually setting the restaurant alight. Government officials, powerless to control the crowd, eventually sent an envoy to ask Lama Chöyang (bla ma chos dbyangs) to intervene. This charismatic reincarnate has rejuvenated the local monastery and commands great respect among the local population. He drove down to the scene, opened the window of his vehicle, and said, twice, “You must not do this.” The crowd then dispersed. The outbreak of such violence causes great embarrassment to local officials. As my informants explained, the inability of the county and prefecture leaders to maintain order would severely damage their prospects for promotion in the forthcoming reorganization of government positions.
These officials are, however, faced with an impossible state ideal—to maintain order as a matter of command and control, through the state's criminal justice system. Tibetans from Golok described a bloody fight between men of two local tribes in which several were killed. Local officials would normally have called in the xhombo and arranged a settlement meeting for them in the town. In an unusual move, however, the incident was reported to the provincial government in Xining, whose representatives directed local officials not to resort to Tibetan mediators but to arrest the offenders and deal with them according to the criminal justice system. This was never going to resolve the problem: because people had been killed, the only solution was mediation and compensation and in the meantime further violence was almost inevitable. As far as local residents were concerned, by preventing mediation the provincial officials had simply created a more severe problem, and for over a year they, and doubtless local officials too, remained in fear of further violence. Eventually, however, the problem was resolved: “the [government officers] arranged a mediation meeting,” one of the town's residents told me, and compensation was paid. “Of course,” he said, “they had to do it this way, or else the problem would never have been resolved.”
In such cases, local officials, with limited room to maneuver, must wait until the attentions of their superiors have turned elsewhere and then revert to the pragmatic solutions that they know will restore order. It should not be assumed, however, that in doing so they are merely trying to advance their own interests and careers. In the conversations I conducted with local officials, a significant number described their activities and attitudes to me in strikingly moralistic terms.
The Local Moral Order
I conducted a series of interviews among government officials in Amdo about how they regard and deal with such problems. Most of my informants were Tibetans, many of them officials at township and county levels, while some had reached high positions in the prefecture and provincial governments. Several interviewees gave me a straightforward account of state control, asserting that officials “resolve problems” and “decide matters” in problematic cases, refusing to be drawn on the possibility of alternatives. One responded to my questions by acknowledging the existence of “old customs” among Tibetans, but insisted that once the police became involved in a criminal matter, everything went through the courts, with no need for traditional compensation. It was an expression of the state ideal, that it has the power and authority to exercise control and maintain order.
By contrast, many were, like my nomad informants, wholly dismissive of the powers of government officials to control violence. “They cannot resolve problems,” several officials told me. “They have to refer to the lamas; the lamas have more respect than the police.” The issue of respect was constantly raised as a reason for the inability of either officials or the police to maintain order. “The Buddhist lamas know the history and often have useful documents; they also have high status and command respect; they can help the officials persuade people to follow their decisions,” admitted one retired county leader, who had initially described procedures to me in official terms. He then added:
It is true, if we just relied upon the law then we could not resolve all these problems. We need the social influence of local tradition, which is often more effective than law. Historically, the Tibetans had a society without law; after Liberation there came an end to the old system of compensation, but it is still sometimes reasonable to use such methods. This is the traditional way to ensure long-term peace.Footnote 15
Several made similar attempts, more or less sophisticated, to rationalize the clash of norms and expectations at the heart of their duties. Some described their relations with traditional leaders as if they were employing agents. One even described Tibetan practices of compensation by reference to Chinese principles of harmony. Another told me that he thought it was better for Tibetans to use their own lamas because the Chinese laws did not suit them, maybe because they were “not developed enough.” On the other hand, he said, some do not know enough about the law, particularly in the realm of marriage and gender relations, when it can act to their advantage. One relatively junior member of the county's Si Fa Suo, a department responsible for the local administration of justice, told me enthusiastically about the education work his department was carrying out, which was giving people more knowledge of the “protection” that the law can give them. He and his colleagues were able to act as mediators in minor cases, threatening to report a violent husband to the police, for example. Similar attitudes were expressed about the People's Mediation Committees, Tiao Jie Wei Yuan, an initiative launched in 1989 under the Regulations on People's Mediation, which seem to work particularly well in the farming areas, where existing village structures can be transformed into mediation committees.
The head of a township police force, by contrast, told me that the problem with police methods was that they often did not resolve the underlying problem. If a fight arose out of a land dispute, village leaders needed to call a meeting with elder men in order to reach a solution that would be respected by everyone, he said. A township leader elsewhere told me that he and his fellow leaders were arguing that a boundary dispute should not be taken to court as a case of theft but resolved through a meeting between both sides. High lamas would be able to restore good relations, he explained. One county leader, making a distinction between law and procedure, told me that Tibetan custom, so jön (srol rgyun), and government law had to work together in order to produce justice, jömdri. Ideally, he said, mediation would combine Chinese law and traditional Tibetan procedures.
The activities of local officials are, therefore, influenced by a number of contradictory factors. First and foremost is the need to maintain peace and order in the areas under their control. The state's emphasis on stability was underlined by one county-level judge, who explained that it is only after stability has been achieved that law can be employed. There is, however, a suspicion of Tibetan lamas and other authority figures, especially at the higher levels of government, leading them to make unrealistic demands of those in local offices about the imposition of law and order. At the same time, the expectations and practices of the Tibetans, amongst whom they need to do this, require careful management and the employment of locally recognized authority figures. In the face of the clash between these ideas and expectations, officials pay lip service to the importance of official law and state control, while engaging constructively with the expectations and dynamics of the Tibetan communities and those who wield power and authority within them. Thus, they construct a local moral order, subverting the state's ideal of sovereignty, by engaging with other, more effective, sources of power and authority.
The Negotiation of Order
Two contrasting cases illustrate the delicate nature of these dynamics and the ambiguous and imperfect extent of the state's power and authority. The first concerns a long-running pasture dispute between Ngulra (dngul ra), a large Tibetan tribe in Machu County, in Gansu Province, and Arig (A rig), a tribe in the neighboring Henan County, in Qinghai Province.Footnote 16 The Arig nomads, who claim descent from the Mongols who formerly ruled the whole area, occupy pastureland on the north side of the Machu River, but also claim the right to a large area of grassland (around 220 square kilometers—sufficient to support around one hundred tents) on the southern side. The Ngulra tribesmen claim that all the land to the south of the river is theirs. Even before 1958, in a conflict that spanned several years, forty-four people were killed in fighting between the two tribes. The dispute was suppressed during the collective period, but in the meantime the provincial boundary was settled, placing almost the entirety of the disputed land into Qinghai, thus under the control of the Arig nomads. During the reform period, the Ngulra tribesmen again claimed the land and, when the Arig nomads refused to move, they launched an attack and killed four of them. Jamyang Zhepa and Gongtang Rinpoche, the highest reincarnates from Labrang, intervened, investigated, and declared that around two-thirds of the land should belong to Ngulra and one-third to Arig. Arig, however, refused to accept this and, in an astute move, petitioned the state government in Beijing, arguing that such an important matter should not be left to the decision of Tibetan lamas, who, they said, represented an outdated regime. Beijing was, unsurprisingly, sympathetic and sent representatives to confirm the established boundary, declaring that all property belonged to the state. Inevitably, further fighting ensued, and in the 1990s a further twenty-nine people were killed.
Eventually, in the early 2000s, Beijing directed the provincial governments to resolve the matter, again confirming that the boundary was not negotiable. Local officials, therefore, set up a mediation committee, with six zabshad (gzu bshad). In Tibetan mediation processes zabshad are appointed by each party to make representations to the zowa. This committee spent around four months consulting with local monks and gowa in order to establish appropriate compensation for the deaths, injuries, and loss of property on each side. The settlement was approved by Jamyang Zhepa, acting as zowa, and accepted by both tribes as appropriate compensation for the losses incurred on both sides. Ngulra, however, continued to argue that the correct boundary was that earlier determined by the lamas.
Despite this lack of agreement, the government staged a settlement ceremony in July 2005. The ceremony, recorded on a VCD, which I obtained on a 2007 visit, took the form of a music festival. A large stage was set up on the Machu grasslands, under a banner proclaiming in Chinese, Tibetan, and English that this was “the gathering at the first bend of the Yellow River” (as the pastureland in Machu County is known). A succession of Tibetan singers and dancers, in both modern and traditional clothes, took the stage before another banner was unfurled, declaring it to be the “ceremony for the signing of a contract of friendship” between Machu and Henan Counties. Officials crowded the stage, and the leaders of the two counties, each in Tibetan dress, signed an official agreement. Khatag (kha btags), white scarves used by Tibetans as a mark of respect, were exchanged, while it was declared over the sound system that this was an important moment marking the reestablishment of friendly relations, to everyone's satisfaction. In one of the many letters of congratulation that had been received, it was explained, Jamyang Zhepa complimented each party on achieving a settlement and urged them to celebrate the restoration of “national peace under communism.”
Thus, the ceremony represented the official acceptance by two administrative counties of the existing provincial boundary. But for the Tibetans involved it also represented the state's failure to settle a conflict between their tribes. My Ngulra informants told me that Jamyang Zhepa was unhappy with the ceremony; his authority had been undermined when the state government refused to endorse his involvement in the 1990s, and his approval of the official settlement was misrepresented during the ceremony. A subsequent and less public meeting was also arranged between the tribal leaders in order to ratify the compensation payments. During this meeting, one Ngulra gowa refused to shake hands with the Arig representatives, declaring, “Your Chinese Machu and Henan [Counties] are now said to be ‘heart to heart’ and ‘hand in hand,’ but our history knows no such traditions.”Footnote 17 Another declared that, “although appropriate mi stong has been paid, we will never regard this as the settlement of the boundary or of the conflict.”Footnote 18
The official ceremony was clearly designed to impress higher officials in the state system, symbolizing the successful promotion of peace and harmony. It made explicit the national ideal of the harmonious society in declarations of unity and slogans celebrating friendship, surrounded by a display of the acceptable face of Tibetan culture. But for the Tibetans it was a sham. Even in distant Golok they were scathing, one urban resident commenting that the whole ceremony was government propaganda, designed to send a message to other minorities. For the Tibetans, it underlined the gulf between their own concerns and those of the state, and between the authority of Jamyang Zhepa and that of the government officials. It was supposedly a demonstration of the state's authority to determine local boundaries and manage tribal relations. In reality, however, it thinly disguised the fact that the Chinese state had reached the limits of its abilities in this regard and had failed to impose order amongst the Tibetan tribes.
In the second case, a more successful attempt to resolve a similar boundary issue demonstrates the possibility and success of shared and coordinated authority during negotiations set up by government officials. A pastureland dispute was described to me by Tibetans from Xunhua, a largely Hui county in Qinghai. It had arisen between a Tibetan nomad group and their neighbors in the Ganja (rgan gya) region of Gansu. The dispute began when a group of Xunhua nomads let their animals cross onto the pastures of Ganja. In the ensuing encounter, one man was killed and a revenge attack led to a death on the other side. The county leaders asked Jamyang Zhepa for assistance. They also organized a group of zabshad, men who were “honest and respected with high reputations,” according to one of the township officials from Xunhua. Representatives from each side were invited to gather in hotels in Lanzhou, the capital of Gansu, where they stayed for over a month while the zabshad negotiated between them. Unlike the Machu-Henan dispute, Jamyang Zhepa refused to get involved until he had received direct requests from the provincial governments. Again, there was an official stipulation that the boundary line could not be moved. However, this time Jamyang Zhepa was able to point out to the parties that he had been part of the committee that had established the boundary line in the first place, along with the Panchen Lama, representing the state, and Zharong Karpo (zhabs drung dkar po), representing Qinghai. Since the other two were now deceased, he explained, that decision could not be questioned. In this way, he avoided the disaster that followed the refusal by the state government to follow his recommendations in the Machu-Henan conflict.
Eventually the zabshad formulated a settlement acceptable to each side, involving the payment of blood money and compensation for medical fees and lost or stolen livestock. Having secured agreement from all concerned, the zabshad took the agreement to Jamyang Zhepa for his approval and signature. The final agreement was read out at a meeting attended by both sides. Throughout this process, the same informant explained, representatives of the county and township governments, himself included, had been present, acting as servants to the zabshad, organizing their food, accommodation, and transport. This process, therefore, resulted in a negotiated order, achieved through the persuasive authority of the zabshad, backed by the religious authority of Jamyang Zhepa, and supported by government officials. The state could, thus, take credit for establishing peace and harmony, thereby apparently affirming its sovereignty; but at best its authority had been established through accommodation with the Tibetans’ own religious leaders, coopting their more effective forms of power and authority.
Coercion and Consent
A simple model of domination and resistance is inappropriate to describe the state-society relations I have outlined here, which are characterized more by accommodation and acceptance than by resentment and resistance. As Moore (Reference Moore1998, 348–49) points out, there have been many critiques of the resistance paradigm, including of James Scott's (Reference Scott1985) famous account, although analysts do not advocate a common alternative path. To describe the Tibetan dynamics, Gramsci's (Reference Gramsci, Hoare and Nowell Smith1971) concept of hegemony seems more appropriate. As Crehan (Reference Crehan2002) argues, Gramsci's account should not be understood as a description of purely ideological domination, however, implying general acceptance of ruling ideas. His interest was in the ways in which coercion and consent are entangled. Submission and subordination, on his account, generate a type of “verbal consciousness”: although people may be discontent with the ruler or elite, they do not act upon that discontent; rather, their everyday behavior manifests acquiescence and moral and political passivity. The combination of active commitment and passive acceptance leads to “contradictory consciousness” and the widespread, if equivocal, acceptance of the perceptions and values favorable to the status quo (Femia Reference Femia1975, 46–47).Footnote 19
The Chinese regime is, of course, very different from the class systems of Europe considered by Gramsci. A political elite has, within living memory, assumed power through violence, and the lack of political alternatives for the Tibetan populations and the underlying threat of coercion are real. As one might expect, there is a good deal of resistance to the more unpopular policies, such as fencing and birth control, which includes noncompliance and foot-dragging. However, as I have described here, there is also considerable acceptance of and acquiescence in governmental measures and policies, at times active engagement.
As Moore (Reference Moore1998, 352) points out, Gramsci emphasized the agency of the subaltern. My own focus in this article has been as much on the government officials and their agency as on that of the subalterns. Consent and engagement on the part of the Tibetans are largely generated by the creative ways in which local government officials interact with them and their leaders. Although expected by their superiors to exercise direct control through the imposition of law, these officials regard the imperative to maintain peace and order as paramount. Tibetan nomads have very particular ideas about how this may be achieved, and attempts by officials to accommodate such expectations result in a set of local practices, moral attitudes, and ideological contradictions, through which passive acceptance of Communist Party rule is generated. While this might appear to be a skillful adaptation of techniques of government to local situations, it also has the effect, as I suggest in the concluding section, of undermining the state's own ideals of sovereignty.
Conclusion: State Sovereignty Undermined
The Chinese state, as an ideological entity, maintains an image of authority and control in its Tibetan areas, guaranteeing order and taking responsibility for the settlement of conflict. The state is, apparently, sovereign. In the academic literature, too, concerns about a “legitimacy crisis” have given way to recognition of the continuing power of both government and party: Pieke (Reference Pieke2004, 525) has argued of China as a whole that the state has become “stronger, better integrated, more efficient, even more legitimate, than it was ten or fifteen years ago.” The focus has been on how it exercises its sovereignty and builds legitimacy, not on whether it is about to collapse.
The suppression of the 2008 Tibetan protests indicated the ability and willingness of the Chinese state to use violence in the imposition of authority over its minority populations. During more peaceful periods, by contrast, officials readily engage with local populations and coopt their leaders’ authority in processes of conflict resolution. Such dynamics are familiar from colonial situations where regimes frequently coopted indigenous leaders to secure a measure of consent to indirect colonial rule. As Guha (Reference Guha1999) succinctly puts it, two parallel dynamics characterize state-society relations in colonial situations: coercion and resistance, on the one hand, and persuasion and acquiescence, on the other.
A form of quasi-colonial indirect rule is not official policy at higher levels in China, however, where there remains considerable suspicion of the power of local Tibetan authority figures and provincial governments discourage engagement with them. When local officials do engage, it is generally not in accordance with higher-level policy. The ideal of state sovereignty does not involve deference to traditional and religious authority figures. Moreover, a colonial model is not entirely apposite for such situations because, like the paradigm of domination and resistance, it tends to suggest a clear divide between state and society. It is not just that the Chinese state vacillates between coercion and engagement; it is not always clear what or where the state is and to what extent it is exercising power and authority. Should we regard the local officials I have described as skillful state agents, adapting their techniques of government to maintain its sovereignty, or should we see them, rather, as agents of the Tibetan population, resisting and subverting the state's authority? The links between state and society are complex, shifting, and negotiated.
What do these shifts and negotiations mean for the state's sovereignty? My case study has suggested that in eastern Tibet the ideology of the party-state and its ability to guarantee a harmonious society is little more than a thin veneer over a series of complex local dynamics. Beneath the symbols of the state's ideal, as they were played out on the Machu grasslands, lay its failure to provide a meaningful or practical basis for social order, while the far less public negotiations of the Xunhua-Ganja conflict reveal the extent to which local officials have to rely upon Tibetan authority figures in order to do so. At the center of these dynamics, local government officials are caught between the unrealistic, ideological demands of their superiors and the radically different expectations of the Tibetan populations. They are often able to establish a local moral order by negotiating a form of tacit consent, but they have to go well beyond their roles as agents of the sovereign state in order to do so.
To understand the Chinese state in all its complexity, we must transcend the easy dichotomy of domination and resistance suggested by quasi-colonial models and also look beyond the state's own edifice of sovereignty. The state, as an ideological entity, maintains an ideal of unitary sovereignty, but that edifice is undermined, not so much by the resistance of its Tibetan populations as by local government officials, its own agents. It is they who can be credited with negotiating a limited, but important, measure of order and consent amongst the Tibetan populations; but in order to do so they also have to subvert the state's own ideals, recognizing and coopting the power and authority of the Tibetans’ tribal leaders and reincarnate lamas.
Acknowledgments
The fieldwork on which this paper is based was funded by a Small Grant from the British Academy, and I also draw upon research conducted while a fellow at the Max Planck Institute for Social Anthropology, Halle. I am indebted to Frank Pieke for his comments and encouragement and for inviting me to the British Inter-University China Centre's conference on Reinventing the Chinese Party-State, at which the paper was first presented. For further comments and assistance I am grateful to Anna Lora-Wainwright and Lamajabb, while the anonymous reviewers at the JAS provided exemplary comments—detailed, constructive, and well-referenced. Numerous informants and assistants in Amdo, generous with their time and information, must sadly remain anonymous.
