If the expansion and consolidation of state power simply undermined, homogenized, and ultimately destroyed the distinctive societies and ethnic groups in its grasp, as various acculturation or melting-pot theories would have it, the world would long ago have run out of its supply of diverse ways of life, a supply presumably created in the dawn of human time. To the contrary, state power must not only destroy but also generate cultural differentiation—and do so not only between different nation states, and between states and their political and economic colonies, but in the center of its grasp as well. The historical career of ethnic peoples can thus best be understood in the context of forces that both give a people birth and simultaneously seek to take their lives.

A perusal of the contents of any social science journal will indicate that ethnicity has been a popular topic during the past two decades. Yet despite the volume of material produced, this period has not seen a notable increase in theoretical sophistication in the field (Drummond 1983:803; Young 1983). For the most part students of ethnicity remain mired in antique arguments about motivation that obscure as much as they illuminate. The following discussion attempts to break through some entrenched conceptual blinders by drawing on the theory of practice outlined by Pierre Bourdieu (1977). This theory avoids the explanatory gaps and fallacious reasoning and at the same time allows richer interpretations of ethnic phenomena, individual and collective, than do conventional models of ethnicity. Since individual and collective expressions of ethnicity have elicited divergent theoretical treatments, the practice theory of ethnicity promises to give new unity to a fragmented field of study.

Despite its title this article does not pretend to be or to offer a study in comparative ethnography, since my expertise in the cultures to be discussed is limited and uneven. I know somewhat more about the rural Irish of the Dingle Peninsula, among whom I lived during 1974–75, than I do about the Pueblo Indians of New Mexico, among whom I spent the greater part of the summer in 1979 and again in 1985. Nor do these remarks have any pretense to social history, although it is to these peoples' collective responses to aspects of their histories that the article is addressed. Rather, it should be taken as a critical reflection on the vulnerabilities of the human spirit and on the vulnerabilities of a science that tries to pierce the meanings of that spirit. It is a tale of two cultures, and of the ethnographer as a teller of tales, and of the pitfalls of cultural interpretation.

Investigations in social history confront a fundamental methodological difficulty: How is it possible to comprehend and to present the dual constitution of historical processes, the simultaneity of given and produced relationships, the complex interdependence of encompassing structures and the agency of “subjects,” the relationships obtaining among the circumstances of life, production, and authority, and the experiences and modes of behaviour of those affected by these circumstances?

Few questions have galvanized the attention of observers of African affairs in recent years as forcefully as the performance of the state on the continent. The debate on the nature of the state—its capabilities, weaknesses, external and societal connections, and impact—has come to occupy center stage in the field of African political studies. This overriding preoccupation emanates from the underlying assumption that the state constitutes a superior means for the fulfillment of economic and social aspirations; participation in its activities is deemed beneficial, and various sectors of society strive to associate with its institutions and gain access to its resources. Some recent works have cast doubt on this assumption, however, and the trend in the literature has been shifting towards an emphasis on the diminishing role of the state in African social life. However, even in these new studies the focus has been primarily on the state itself, its difficulties, incapacities, and failures, rather than on societal response to its actions.

Beginning around the turn of the twentieth century, the Chinese state launched onto a course of development that seemed to resemble the process in early modern Europe that Charles Tilly and others have called state making (Tilly 1975). The phenomenon of an expanding state structure penetrating levels of society untouched before, subordinating, co-opting, or destroying the relatively autonomous authority structures of local communities in a bid to increase its command of local resources, appeared to be repeating itself in late imperial and republican China. The similarities include the impulse toward centralization, bureaucratization, and rationalization; the insatiable drive to increase revenues for both military and civilian purposes; the violent resistance of local communities to this inexorable process of intrusion and extraction; and the formation of alliances between the state and local elites to consolidate their power (Duara 1983).

Problems of periodization have received but limited attention in Burmese historiography. Precolonial, that is to say, pre-nineteenth-century history, is said to be of a piece, without significant institutional or social transformations. Dynasties and rulers changed, of course, sometimes with stunning rapidity; but it is always assumed that these oscillations occurred within a static framework. Lamenting the failure of the early Kon-baung kings to move their capital to the coast, G. E. Harvey, whose history remains the standard work on the precolonial era, observes, “Their ideas remained in the nineteenth century what they had been in the ninth. To build pagodas, to collect daughters from tributary chiefs, to sally forth on slave raids, to make wars for white elephants—these conceptions had had their day, and a monarchy which failed to get beyond them was doomed.” In the same vein, it has recently been argued that no “significant transformations” occurred between the origins and collapse of monarchical Burma. The entire precolonial royal era “should be viewed as one entity,” for from the mid-ninth to the late nineteenth century “the major features of [Burma's] political, economic, social, administrative, and religious systems were also virtually identical.”