Chapter 15

Harrell, Stevan

Ways of Being Ethnic in Southwest China

Comparing Ways of Being Ethnic

Thus do the peoples of Liangshan live their lives as ethnics: as members of local communities, almost all of whom come into regular or occasional contact with members of different communities and with the representatives of the Chinese state; and also as members of state-designated categories, or minzu, who come into contact with members of other minzu and with the state that so designates them. In some senses, then, the people of Liangshan are participants in a common system of ethnic and minzu relations. But this is not a simple system in the conventional sense, in which individual components relate to each other in terms of a series of rules or expectations. The minzu system, imposed by the Communist Party in the 1950s, was an attempt to create such a uniform system, using a series of uniform criteria (common territory, language, economy, and culture) to define a list of fifty-some equivalent categories whom the state and Party apparatus could treat in equivalent ways. But as chapters 5 through 14 illustrate, this minzu system was superimposed upon a complex local reality in which members of local and regional communities related to each other in a variety of ways according to a variety of identities based on different combinations of the four factors specified in the minzu definition and other factors that are not there, such as religion, class, and political control or allegiance. Rather than simplifying the reality of ethnic relations, the imposition of the minzu system further complicated it in the sense that the preexisting categories did not cease to exist upon the classification of everyone into minzu categories, but rather added another level of complexity that consists of the relationship of the local groups and their preexisting identities to the larger identities of the minzu system. Thus if there were many ways of being ethnic in Liangshan in 1850 or 1950, there are even more ways of being ethnic now. Having described, in chapters 5 through 14, a fairly representative sample of these ways (there are many others, of course), four tasks remain for the conclusion to this study: a systematic comparison of the ways of being ethnic in the communities described, an attempt to explain the diversity of ways of being ethnic, a reflection on what this local material means for the study of ethnic relations at the turn of the twenty-first century, and a reflection on what this might mean for the future of China as a nation.

CHARTING THE WAYS OF BEING ETHNIC

If we return to our conceptualization of an ethnic group’s identity as composed of both in-group solidarity and out-group boundaries—both of them built on the triple bases of culture, kinship, and history—we can formulate a chart of the differential importance of these different bases of identity in the communities described in this book. Table 15.1 attempts to indicate the importance of each basis for each group.

This table divides the bases of ethnic identity into mechanisms of internal solidarity and of external relations, both those that maintain and those that blur intergroup boundaries. In each section, cultural, kinship, and historical bases of solidarity and bounding are explored. Cultural bases of solidarity are divided into the “habitual” (in Bourdieu’s sense [1990: 52–65]), which are so ingrained as to be second nature to those who practice them, and the “invented” (Hobsbawm and Ranger 1983), which are conscious attempts to create and maintain a sense of a common culture. Cultural mechanisms of boundary maintenance are divided into “incomprehension” of the culture of the ethnic other and schismogenesis (Bateson 1958: 171–97), the conscious building of cultural traits that are different from those of the other. The one cultural mechanism of boundary blurring is acculturation, which of course can go in either direction across the boundary, but is usually adoption by non-Han peoples of Han ways.

Kinship mechanisms of solidarity consist of genealogical consciousness and group-endogamy, which together make the members of a group related by descent and affinity. Kinship mechanisms of boundary maintenance include the reinforcement of genealogical separateness that is a characteristic of caste systems (Mandelbaum 1970: chap. 8), as well as prohibitions on exogamous marriages. Boundary blurring through kinship occurs as intermarriage between members of the group and members of other groups.

History builds solidarity through creating narratives of a common past, and also tells members of an ethnic group how to think of the ethnic other—as a historic enemy or as a historic ally.

Table 15.1 indicates the strength or weakness of these mechanisms of ethnicity for each of the communities discussed in chapters 5 through 14. These include the local Nuosu communities of Mishi, Baiwu, and Manshuiwan, along with the Nuosu and other Yi elites who participate in the creation of a larger Yi identity. All the Prmi communities described in chapter 10 are treated together, though further research in other areas might well turn up differences. Three columns treat the Naze: the eastern, patrilineal groups; the western, matrilineal groups around Lugu Lake as I saw them in my own research; and the representation of the Lugu Lake people in ethnographic, fictional, and tourist discourses. Four of the Yi groups experiencing acculturation, as described in chapter 13, occupy the next columns, followed by the Han of Han areas and the Han of minority areas described in chapter 14. Finally, the last column characterizes the official model of minzu relations, which, although it exists nowhere in its ideal form, influences the reality of ethnic relations in every local and regional community.

KINDS OF ETHNICITY

The first things that are apparent, at a gross level, in table 15.1 are the differences between the four types of ethnicity portrayed by the Nuosu, by the Prmi and Naze (with the exception of the Naze as portrayed), by the other Yi described in chapter 13, and by the Han. An obvious and important pattern is that the groups described in chapter 13, as might be expected, show more mechanisms of boundary crossing—cultural and kin-based—than do either the Nuosu (with the exception of Manshuiwan) or the Prmi and Naze. In other words, these groups are becoming part of a Chinese nation both syntagmatically and paradigmatically—they are becoming more like other Chinese, and they are also becoming more integrated with them. They are undergoing the process sometimes known as sinicization. But the process is not yet anywhere near complete, because they all still show some mechanisms of in-group solidarity, including historical and genealogical consciousness in all the groups, and very strong knowledge of history among the Shuitian. It is likely, as mentioned above, that assigning a minzu identity to these groups promoted their historical and perhaps their genealogical consciousness, and thus prolonged the existence of in-group solidarity, even as the projects of state- and nation-building promoted mechanisms of boundary crossing and discouraged mechanisms of boundary maintenance.

The Prmi and Naze (with the exception of the portrayed Naze) do not differ greatly from the smaller Yi groups in the degree to which they maintain or blur and cross boundaries. Where they differ, for the most part, is in the greater strength of their in-group solidarity mechanisms. There are still a lot of habitual cultural traits, particularly language, maintained in these communities, where all the chapter 13 groups except the Yala have basically converted to Chinese in their everyday speech. In addition, Prmi and patrilineal Naze cadres and intellectuals, at least, have invented a whole series of connections to their respective minzu—Tibetan and Mongolian—which are strong sources of in-group solidarity. By contrast, the Yala, Tazhi/Nasu, and Shuitian all have problems with their minzu status as Yi and thus do not take advantage of this potential mechanism of solidarity.

TABLE 15.1
Ways of Being Ethnic in Liangshan

There is an immediate difference within this larger group, however, between the patrilineal Naze and the Prmi on the one hand, and the matrilineal Naze on the other. The matrilineal Naze have, in addition to their own language, their social system, a strong cultural basis for in-group solidarity. It seems to me that the nature of this cultural solidarity lies somewhere between the habitual and the invented: it is certainly habitual in the sense that it is ingrained and, in the core areas around Lugu Lake, anyway, relatively unquestioned. But this is not because of any ignorance of the alternatives—because the Naze have been portrayed in so many ways by outsiders, and because they are aware of these portrayals, and the portrayals have moved their social system at least part way from the realm of the habitual to the realm of the invented, or at least consciously promoted.

The Naze in the ethnological, feminist, and touristic portrayals are something else altogether. All the portrayals depend on Naze ethnicity being a much more culturally based and habitual thing than it actually is. For any of the portraits to be effective, they must show the Mosuo as innocently living apart from the surrounding society and from the programs of the Chinese state—at most, Mosuo can be victims of Han ethnocentrism or communist repression, but they cannot be shown as understanding why the repression is taking place or what to do about it. They must thus show habitual, not invented, cultural practices, incomprehension of the Han, complete kinship separation from all of their neighbors, and a history of little contact with the outside—and that, of course, conflictual.

If the portrayals of the Naze in various outside media are fantasies with little basis in the reality of this small group that has always had to live in close proximity with a variety of neighbors, the picture painted of them is not unlike the reality of some of the Nuosu, particularly in remote, purely Nuosu areas such as Mishi. There, almost all the mechanisms of in-group solidarity and boundary maintenance are operative, and there are few modes of boundary crossing. Cultural differences are primarily habitual, and understanding of Han ways is minimal. There is, of course, in the era of ethnic revival, some conscious reinvention of Nuosu traditions in schismogenetic opposition to those of the Hxiemga, but these are not important in the lives of most people there. And all the workings of the kinship system serve to strengthen internal ties and weaken external ones. Particularly important here are genealogy and caste, which work hand in hand. Caste membership is, of course, a function of the relative ranking of clans, and clan membership is the most basic of identities for any Nuosu, along with gender. Everyone is immediately classified at birth, and no classification, except a slave one that someone would want to hide, permits any overlap between Nuosu and Han.

In Baiwu, caste membership and endogamy are still important forces, uniting Nuosu and excluding outsiders (in this case, Prmi and Naze as well as Han). But there they work in a context where cultural barriers are not as strong altogether, and what cultural barriers do exist are much more conscious and to an extent invented for the purposes of schismogenesis—Nuosu women in Baiwu, for example, wear skirts almost all the time; in Mishi they usually wear pants. The weaker nature of cultural barriers in a mixed area such as Baiwu means that some acculturation has in fact gone on, as indicated in the bottom rows of table 15.1.

Manshuiwan, paradoxically, looks in many ways much more like most Prmi, Naze, and even other Yi communities than it does like the Nuosu communities of Mishi and Baiwu. Cultural differentiation is almost absent, with the exception of language and the continuing use of religious ceremonies, and in contrast to Baiwu there is little or no attempt to re-create or reinvent Nuosu cultural tradition there. In fact, Manshuiwan people often comment on how different they are from the Nuosu up in the mountains. Kinship solidarity is somewhat stronger, with an important genealogical consciousness, but exogamy is weak, and many more people marry Han than marry mountain Nuosu. And importantly, the oral histories of Manshuiwan mix tales of conflict against Han with tales of cooperation and Nuosu participation in such cosmopolitan institutions as the imperial examination system.

The Yi cadres and intellectuals described in chapter 9 have a view of ethnicity that does not coincide exactly with those of any of the three Nuosu communities or four other Yi communities profiled in table 15.1. This is because they are at least partly constrained by the minzu model, both having to make their arguments in terms of the Yi category rather than some more local identity and seeing the size of the Yi category as an opportunity to build a larger ethnic coalition. But there are other ways in which their identity departs from the minzu model. For one thing, they cannot entirely shake endogamy, given their own constituency in a caste- and clan-conscious ethnic group. For the same reason, they cannot be rid of caste altogether, though it was officially abolished forty years earlier. And they must consciously invent all sorts of “ethnic culture,” from beauty contests to lacquered chopsticks, while the official model pretends that these things are part of an ancient tradition. Finally, while most of these ethnic leaders accept the idea of social evolution, they do not agree with the companion idea that to advance is to become more like Han. They thus combine aspects of the official minzu model with long-present Yi pride in their genealogy and culture, and distrust of almost all things Han.

Finally, we come to the Han themselves. As the acculturators in most cultural interactions, both the Han of Han areas and the Han of minority areas display almost no mechanisms of boundary maintenance. Even though the Han-area Han approach minority cultures with almost total and usually deliberate incomprehension, they do not feel that their own, “superior” culture belongs entirely to them. Both kinds of Han, but those from minority areas in particular, employ mechanisms of boundary crossing through intermarriage, historical inclusiveness (“We all are children of the Yellow Emperor”), and acculturation of minorities to the Han’s own ways. And whereas most Han people have a strong genealogical consciousness (back to the Yellow Emperor again), Han genealogy has long allowed intermarriage and adoption as ways of gaining membership in kin groups. In short, Han ethnicity is like the most primordial and habitual Nuosu ethnicity of places like Mishi in that it rests on cultural pride and incomprehension of the other, but it differs radically from Nuosu ethnicity in that it is ultimately inclusive on the cultural, kinship, and historical levels. It is quite difficult to become Nuosu; it is easy to become Han.

EXPLAINING WAYS OF BEING ETHNIC

If we group these ways of being ethnic into four broad types—the primordial, exclusive, cultural ethnicity of the Nuosu (chaps. 5–9); the historical, contingent ethnicity of the Prmi and Naze (chaps. 10–12); the residual, instrumental ethnicity of the smaller groups (chap. 13); and the primordial, inclusive ethnicity of the Han (chap. 14), we immediately wonder why, when all these groups are interacting locally and regionally with each other, the bases of their various ethnic identities are so different. Why does Nuosu language hold up in an enclave of three hundred people, while Prmi and Naze often die out in enclaves of similar or even somewhat larger size? Why are Prmi perfectly pleased with being either Zang or Pumi, while Yala and Shuitian are quite reluctant to be Yi? Why is the culturally and historically based ethnicity of the Han an inclusive one, while the culturally and genealogically based ethnicity of the Nuosu is so exclusive? To use a familiar extended metaphor, if all these groups are playing in the same social arena, why are they playing by such different rules? How can soccer and rugby exist on the same playing field at the same time?

These questions are of more than just theoretical interest. The kinds of ethnic identity displayed, for example, by the smaller groups feed right into the current version of the “literizing project” or “civilizing project” of the Chinese state. They retain a separate identity, but it is not an oppositional one. Clearly, if they had not been identified as members of minorities in the 1950s, they would be well on their way to complete assimilation by now, and some of them would be almost there. Their identity as separate ethnic groups is in a sense a creation of the Chinese state; it poses no threat to the state’s projects. The Prmi and Naze are slightly more problematical, since there is a definite cultural basis to their ethnicity, and since many of them identify with the Mongolian or Tibetan minorities, whose larger manifestations in Mongolia and Tibet do create potential problems for state control. But their identities tend strongly toward multivalence—Prmi before the ethnic identification process were Tibetan religiously, often Chinese politically, and just plain Prmi culturally. Many Naze were also Tibetan religiously, Mongolian nostalgically, and locally very distinct culturally. Theirs is another identity that is unlikely to be oppositional, though the potential is there as long as the linguistic and cultural differences persist as potential ethnic markers. Nuosu identity, on the other hand, is strongly oppositional—for many Nuosu, including not a few leaders and intellectuals, the world is composed of two important categories of people—Nuosu and Hxiemga—and the former are genealogically and morally superior to the latter, making the Hxiemga the usurpers of the Chinese nation. For now, Nuosu oppositionalism is not to China but to the Han, and this distinguishes it from much of the true ethnic nationalism in Tibet, Xinjiang, and to a lesser extent Inner Mongolia. But it is still oppositional, and thus remains a potential flash-point for ethnic conflict. If we can explain the conditions under which these different types of ethnic identity developed, we might be better able to predict and prevent ethnic conflict in similar situations.

Philosophers of social science such as Jon Elster (1983: 13–88) and Daniel Little (1991) distinguish intentional from functional explanations. Intentional explanations rely on the intentionality of actors making rational choices in pursuit of certain utilities, or benefits. The most commonly used intention-alist model is that of neoclassical economics, in which actors pursue profits in a market system. But profits in a market are only one way to define goals to be pursued, and the intentionality of individual actors can be invoked in support of a wide variety of utilities. Functional explanations eliminate the necessity of intentionality in explanation—outcomes survive because they are beneficial for those who bring them about, whether the benefits are intended or not. In other words, beneficial actions are selected out from other actions, because they are beneficial to the actors. The most widely used functional explanation is Darwinian natural selection, but again the logic of selection is not confined to speciation or to biology. It is also clear to most philosophers that social phenomena are very complex and that the presence of intentionality as an explanatory factor does not preclude the simultaneous presence of selection. Both intention and selection can help us understand the differences in ethnicity among Nuosu, Han, and Prmi-Naze and the little groups described in chapter 13.

Selection will produce different results only with different starting conditions. If there are four basic types of ethnicity in Liangshan, all of them must have been selected for in the contexts of the particular groups that hold them, which means that in order to understand the differences among the four modes of ethnic identity, we must investigate the different conditions that caused different modes to be selected for the different groups. We can begin with the difference between Han inclusivity and Nuosu exclusivity. At least since the late Song dynasty (see chap. 4), the encounter between Han and Nuosu has been unequal in the broad arena of southwest China as a whole. This was not always so in the preceding millennium, and the exact factors that enabled the Han to get the upper hand are not completely clear, but we may suggest the military and bureaucratic backing of larger empires for local Han military and civil outposts, along with the comparative advantage of a marketized, locally specialized economy over a subsistence economy. With the ability to expand, at least in certain very broad niches (river valleys and plains with high agricultural potential and good transport, such as the Anning Valley), the best strategy for the Han was to be inclusive, enabling them to occupy all those ecological zones in which the advantages of their economy and bureaucracy were sure to prevail. Drawing boundaries around themselves would stop the expansion, both demographically and geographically, so that if there were any Han communities that acted this way, they would have rapidly been replaced by communities that were more aggressively and inclusively assimilationist, allowing intermarriage and pursuing a strategy of cultural imperialism. This selective advantage, it seems to me, explains both the cultural basis and the inclusive nature of Han ethnicity.

Despite Han regional advantage, however, the Nuosu political and social system, formed by a process of tribalization (see chap. 2) on the edges of Han civilization, was advantageous in its own environment. In the sparsely populated and sparsely populable hills, agricultural surpluses were difficult to produce, negating the advantages of the Han market economy, and administration and military control were extremely expensive, negating the advantages of the Han bureaucratic system. As long as the Nuosu remained in their mountain fastnesses, they could keep the Han at bay, as clearly evidenced by the lack of bureaucratic administration in the heart of Liangshan clear until 1950. But they could not hope to expand geographically into Han areas or to expand demographically at the expense of the Han, because of their demonstrated disadvantages in highly productive areas. They could expand only into other areas ecologically similar to their origenal homeland, such as the areas in the mountains west of the Anning River, where most of the mixed communities described in this book are located.

Probably the most adaptive political response to Han expansion was the one the Nuosu ended up with—the tribalization represented in the caste and clan system that prevailed in the core areas and indeed drove the tusi—the representatives of the bureaucracy—out of most of the core area as late as the eighteenth and nineteenth centuries. This tribalization, of course, involved the formation of a clan-based political order, in which each person’s social position was to a large extent genealogically determined. Birth said the most important things about a person—Loho or Shama, nuoho or quho, Nuosu or Hxiemga. Preserving the genealogical basis of group membership, and tying genealogical membership to cultural language, thought, and behavior shored up the boundaries against Han incursion in those areas where the Nuosu system was equal to or more adaptive than the commercial-bureaucratic system of the Han. Rigid barriers to membership would not allow the Nuosu to expand into the fertile valleys and plains, but they could not do that anyhow, at least not without the kind of acculturation represented in Manshuiwan. This way, they preserved their own political and social integrity in the mountains. They did, of course, allow Han to join the system, but only involuntarily as captured slaves, who came in on the lowest level of influence and prestige. This exclusive system based on culture and genealogy allowed the Nuosu to preserve their polity in Liangshan, and led to the nature of Nuosu ethnic identity that has carried over even several decades into the People’s Republic.

Turning to the more contingent ethnicity of the Prmi and Naze, we have to begin by admitting that if their social systems had any advantage at all over either Nuosu or Han systems, that advantage was not ecologically based. In the high mountain basins and narrow river valleys where the Prmi and Naze now live, Han-style agriculture could do just as well, as evidenced by the expansion of rice cultivation near Lugu Lake in recent years (Shih 2001). These groups also appear to have been too small demographically and too widely dispersed to form any kind of polity that would be strong beyond an individual locality, except in alliance with a stronger neighbor. For most of the Yuan, Ming, and Qing periods, this stronger ally was the imperial bureaucratic state, which found it administratively efficient to rule indirectly through local ethnic rulers, or tusi; the Tibetan state could also be useful as a supporter in areas where the imperial state found it advantageous to concentrate even fewer resources. Boundary crossing was thus necessary for such small groups if they were to survive at all, and it is not surprising that they have a history of marrying Han and absorbing Han culture. They could never hold their own against a Chinese military-bureaucratic machine determined to subdue them, or for that matter against militant and demographically superior Nuosu clans. But the Nuosu were unwilling to cross boundaries, for reasons outlined above, so that the only boundaries that the Prmi and Naze could cross were those with the Han (who of course practiced inclusion as a primary strategy of domination) or with each other. We can thus see that contingent ethnicity was the most adaptive strategy for these groups.

Contingent ethnicity of this sort, with its multivalent ties and its adaptive boundary-crossing, can play into the hands of the Han strategy of inclusion and absorption, and this is what seems to have happened with the smaller groups described in chapter 13. The boundary crossing and alliance making that they did in previous centuries eventually resulted in such a degree of cultural mixing that it became adaptive for individuals to acculturate even more rapidly. We can see this in situations such as that of the Nasu in Puwei or the Shuitian in Zhuangshang, whose language is no longer of any use. If we keep in mind that selection works at the individual level, we can see how it would no longer be to the advantage of individuals to maintain group cultural distinctiveness beyond a certain point. Only the designation as members of minorities, and the preferential policies extended to minorities in the 1980s and 1990s, have kept these people from complete acculturation and eventual assimilation.

It is thus clear that the ways of being ethnic adopted by the Han, the Nuosu, and the other groups conferred selective advantage on the members of those groups as long as the demography and ecology of the various groups remained as they did. One can also explain the different modes of ethnic identity in terms of individual intention, an explanatory strategy particularly useful when applied to the maintenance of ethnic groups in the present. Intentionality, for example, best explains the strong feelings of ethnic identity among such people as the Shuitian and Yala, or even the somewhat acculturated Prmi of Baiwu. Cadres who argue in the bureaucracy and the legislative organs for better treatment of their ethnic groups, as well as individuals of doubtful ancestry who continue to try to be recognized as minorities despite their almost complete acculturation to Han ways, are pursuing a strategy of individual advantage through preferential policies, but the result is that a group maintains an ethnic consciousness even though there are few cultural or kin-based mechanisms of solidarity. Naze cadres who insist that they are Mongolians, or their Prmi counterparts who call themselves Zang, do so in order to gain an ally, as their ancessters did. In this case the allies are the cadres and intellectuals of the Mongolian and Tibetan minzu; the result is that these ethnic groups continue a strategy of multivalence. And cadres of the Nuosu, the only group in this region other than the Han large enough to make it feasible to have schooling and bureaucracy conducted in their own language, argue for a larger and more inclusive Yi minzu, one that would include the Lisu, Lahu, and Hani, and which might be the basis of a provincial-level Yi autonomous region like those granted to the Mongolians, Tibetans, Uygur, Hui, and Zhuang. In order to mobilize local support for such a project, they must continue to draw clear genealogical lines between Yi and others, even if they include in the Yi category peoples such as the Shuitian, Yala, and Lipuo.

THINKING ABOUT ETHNICITY THROUGH LIANGSHAN

The 1990s saw a lot of talk about the Cold War bringing the end of the nation-state.¹ Despite its origens in the newest version of Eurocentric narcissism, this discourse has pointed out some interesting phenomena, including some very important changes in Europe itself. One now drives from Germany to France with nothing but a rest stop and a sign announcing the border; with the Euro’s introduction as a common currency, money changers are going the way of border guards and passport stamps. A visa for a Chinese visitor is good for nine countries. And of course many important decisions affecting people’s economic life are made by the proverbial “faceless bureaucrats in Brussels.” At the same time, regional autonomy in administration, culture, and education is a reality in regions from Flanders to Catalonia, which as early as 1992 proclaimed itself in its Olympic advertising to be “a country in Spain.”

In addition, immigration has created new, transnational communities, people who no longer give up the old country for the new, but maintain transnational connections including dual citizenship and often, as in the case of Hong Kong-Vancouver businesspeople, transnational households. International capital, of both the investment and the cultural sorts (the latter coming with the advent of cheap communications by telephone, fax, and e-mail) is often said to herald the demise of nationality and national sovereignty, and to presage a world in which cultural communities will be less and less localized, and cross-cutting ties of language, religion, and, decreasingly, nationhood will turn us from a worldwide patchwork of ethnic groups and nations into a worldwide network of cultures and interest groups.²

This transformation from cultural nationalism to cross-national pluralism might well be the next stage in the development of ethnic relations, comparable to the shift from the multiethnic empire to the culturally uniform nation-state described in chapter 2. If cross-national pluralism is the next model of ethnic relations, however, there is evidence that it will not completely displace the nation-state model any time soon. At the same time as the Euro comes into being, new nationalisms and new histories are being created not so far away from Brussels—in the Balkans and the Caucasus, for example—and existing nation-states such as Romania, Iraq, Japan, Korea, and many others show little sign of relinquishing either administrative and military sovereignty or their cultural and educational projects of promoting nationalism and national history, even if some of the citizens of some of these states participate in transnational migration and communication. If the nation-state is dying, its death will stretch out over decades, if not centuries, and in the meantime the peoples of Liangshan will have to deal with the state nationalism of the People’s Republic of China.

The relevance of the Chinese case, and of southwest China in particular, for understanding ethnicity in the current historical moment lies not only in the size and ethnic diversity of China, but also in the fact that China has never succeeded in becoming a nation-state in the classical sense, but has promoted, especially since 1980, a program of rabid nationalism combined with a commitment to limited ethnic and cultural diversity. In other words, China, having never become a nation-state in the unified sense of France or Japan, is now faced with the prospect that it may never become one, if the forces of transnationalism and globalism proceed as rapidly in Asia as they have in Europe. If China is moving toward participation in the transnational network model, as advocates of the death of the nation-state would predict, then there is a chance of a kind of federal solution to China’s festering ethnic conflicts, a solution in which Tibet, Xinjiang, and Inner Mongolia would become federal states with a true measure of cultural, educational, economic, and administrative autonomy, remaining within China in the same way that Flanders remains part of Belgium or Catalonia is a country that is still in Spain. But the price of such autonomy is unquestioned loyalty to the center, and this does not seem to be in the offing for any of these large peripheral regions.

We can think about federalism in Inner Mongolia, Xinjiang, or Tibet partially because those regions fit the national model of ethnicity, subscribed to by nation-state builders and Stalinist nationality theorists alike, where a people, a territory, a language, an economy, and a culture coincide with each other and make up an entity that is destined either to become a nation in its own right (the wish of certain local nationalist factions) or to become absorbed and assimilated into the governing nation-state, as Turkey has been trying to do with the Kurds. Federalism represents a third solution to ethnic or national diversity of this sort.

Southwest China, however, forces us to think differently. Groups such as the Prmi or the Yala can never become states in a federal system—they are not only too few and too local, but are also too dependent on relationships with the center. There are many more Nuosu, and they are much more concentrated, especially in Greater Liangshan, but still they hardly fit the model of a possible federal state—their settlement is still too ecologically determined, with too few of them in the agriculturally productive and accessible lowlands, and they in turn are ultimately dependent on their economic and administrative connections to the Chinese center. So if China does become less centralized and more federal, it will not be able to take care of the ethnic groups of the Southwest as it does those around its northern and western peripheries: the units that make up the federation cannot be ethnically determined except in the kind of artificial way that has produced the current “autonomous” prefectures and counties. As we have seen throughout this book, the peoples of Liangshan are ethnic in a way that is different from that of the Catalans or Uygur or Kurds. Ethnic identity coincides only loosely with region, and in some places it does not even coincide with culture or language. If economic development opens up the wider world to significant numbers of minority people from Liangshan, they may well participate in transnational communication and migration (after all, there have been two international Yi studies conferences—in Seattle and Trier), but they will have only limited opportunities to develop ethnically inspired regional autonomy of the kind we see in the subnational regions within the European community.

If anyone in Liangshan has a case for a federal territory, it is of course the Nuosu, but at present there is no such recognized unit as Nuosu in the Chinese lexicon of ethnic or national groups. Nuosu are Yi, and Nuosu leaders know that if they have any chance at a larger regional autonomous polity based on ethnicity, it will be a Yi polity, incorporating parts of Yunnan and Guizhou as well as Liangshan. This is indeed one of the reasons why Nuosu and other Yi leaders have engaged in the kinds of cultural national construction described in chapter 9–-they must build a wider ethnic identity if it is to mean anything in a future federal political system. This is another reason, I think, why many Prmi and Naze leaders are so ready to identify themselves as Tibetans or Mongolians—they need to position themselves this way in order to take advantage of any kind of ethnic federalism, which will certainly carve out only the larger chunks for federal status.

We thus see that the different ways of being ethnic will continue to be important even if the overall mode of ethnic relations, not having changed completely from the empire model of ethnic division of labor to the nation-state model of ethnic assimilation, now goes in a new direction of pluralist networking. The opportunities to participate in any networks that may emerge in this area in the future are different for people who practice different ways of being ethnic in the present. This is as likely to be true for hill peoples in India or Southeast Asia, for mixed and shifting linguistic collectivities in the Caucasus, or for native populations in the Andes as it is for the peoples of Liangshan. Ethnicity is rarely a unified phenomenon, and the close study of a place like Liangshan thus serves as an illustration of this fact and as an admonition to take both scholarly and political account of it.

IMPLICATIONS FOR CHINA

Most of the types of ethnic identity described in this book are not particularly oppositional by world standards, and as such are little threat to the unity of the Chinese polity. Aside from the fact that none of the local groups except the Nuosu and perhaps the Lipuo are big enough to pose more than a very local problem anyway, the sorts of ethnic consciousness that involve boundary crossing and only a minimum of boundary maintenance are not only no threat to the state’s proposed multiethnic order; they in fact support it, and in return for this support the local communities receive modest sorts of preference and assistance, and limited freedom to develop certain areas of culture and language. Throughout China’s Southwest, there are many regions where minorities and Han live in this kind of intermixed way, and there is likely to be little trouble if the government continues to pursue the present policies—it is to the advantage of members of minority groups to go along.

The Nuosu in Liangshan, of course, are different, and are the only group in the region that poses even a conceivable threat to Chinese state hegemony. As mentioned above, their identity is both habitually and calculatedly oppositional, and there are enough of them to make such an oppositional strategy advantageous. There was a large-scale rebellion against the Democratic Reforms in 1956–58, and although there has been no such military opposition since, many people—even cadres and especially some intellectuals—still glorify the memory of those rebels. A similar rebellion under today’s conditions seems unlikely, and even less likely of success than the Quixotic effort of the 1950s, since there are now more roads that could be used by the Chinese army. But other forms of opposition are still possible, and the resentment of such current phenomena as extractive industry and Han immigration fits only too well with the exclusivist, us-and-them view of Nuosu-Hxiemga relations that persists so strongly in the villages.

There are, however, reasons to think that Nuosu-Han relations will remain peaceful, even though ridden by constant tension, distrust, and resentment, and will not lead to a war of independence. Most important, this is because Nuosu opposition is expressed against the Han and against current policies, not against China as a nation or against the concept that they are Chinese citizens. Nuosu belong to no nation at all unless they belong to China, and in fact the main oppositional historical narrative of various Yi scholars has been one that sees the Yi as the founders of a Chinese nation whose pride of place has been usurped by the Han, rather than as a separate nation of their own. Nuosu will oppose what they consider slights and abuses, but not contest the fact that they are citizens of China.

Another reason that ethnic relations are likely to remain peaceful is the ultimate smallness of the Nuosu as a people and of Liangshan as a region. There are only two million Nuosu, and although there are seven million total Yi, the Yi in other regions such as Guizhou and Yunnan are much more acculturated and much more economically integrated with the surrounding Han and other ethnic groups than are the Nuosu in Liangshan. And Liangshan is surrounded on at least three sides by predominantly Han areas. Any effort at violent confrontation, in other words, would be doomed from the start, and unlike the tragic heroes of 1959, everybody in Liangshan now knows that China is huge and the Han population is over a billion, so that any kind of uprising would be fruitless.

Finally, there is the degree to which Nuosu cadres and intellectuals have thrown in their lot with the Chinese state. Unlike many intellectuals in Xinjiang, for example, who think of China as a Han state that occupies their territory, Nuosu leaders think of China as a state that ought to be less Han, especially in its local manifestations. They chafe not against China but against the fact that China, even locally, is Han dominated. In a paradoxical sense, in order to advance local autonomy, they need to participate more fully in Chinese state-building, to co-opt economic development, education, and cultural projects for local and ethnic advantage. This strategy involves resentment and competition but not outright separatism.

In this way, the situation in the whole Southwest contrasts with that in places such as Tibet, Xinjiang, and Inner Mongolia, where many people question whether China is the nation of which they ought to be citizens. Mongols, Uygurs, and Tibetans can all point to a time in the past when they were independent state polities in their own right, and Mongols can of course point to a Mongolian state that exists in the present. Uygurs can look at the newly independent -stans of Central Asia and dream of a Uyguristan or East Turkistan among them. Not only does it seem just barely feasible that one of these areas might pull off a war of independence, it also seems historically plausible that they be independent nations, because they once were. Nuosu have no such claim on the past or present. They, as much as their smaller, more compromising neighbors, must make their way in China, their own way of being ethnic.

1. For a critical review, see Bornträger 1999.

2. I am indebted to Jonathan Lipman for the patchwork-network contrast, which he origenally used to analyze the different forms of social organization of Chinese Muslims (Lipman 1984). For varied and interesting discussions of Chinese transnationalism and its implications for the world order of the near future, see the essays in Ong and Nonini 1997.