Stay informed: Sign up for eNews Subscribe

Colonialism in Question Theory, Knowledge, History

Read the Introduction


Colonial Questions, Historical Trajectories

The burst of scholarship on colonial studies in the last two decades—crossing the disciplinary boundaries of literature, anthropology, and history—has begun to fill one of the most notable blind spots in the Western world's examination of its history. Yet there is something strange about the timing: scholarly interest in colonialism arose when colonial empires had already lost their international legitimacy and ceased to be viable forms of political organization. Earlier, when colonialism was an object of mobilization, scholars and intellectuals were most captivated by the drama of liberation movements and the possibilities of "modernization" and "development" for people whom colonialism and racism had excluded from the march of progress.

Part of the impetus behind the recent research and writing on colonial situations has been to ensure that this past is not forgotten. But the colonial past is also invoked to teach a lesson about the present, serving to reveal the hypocrisy of Europe's claims to provide models of democratic politics, efficient economic systems, and a rational approach to understanding and changing the world, by connecting these very ideas to the history of imperialism. Such concerns have led some scholars to examine thoughtfully the complex ways in which Europe was made from its colonies and how the very categories by which we understand the colonies' past and the ex-colonies' future were shaped by the process of colonization.

Yet a significant part of this body of work has taken colonial studies out of the history whose importance has just been asserted, treating colonialism abstractly, generically, as something to be juxtaposed with an equally flat vision of European "modernity." This side of the field has focused more on stance—on critical examination of the subject position of the scholar and political advocate—than on process, on how the trajectories of a colonizing Europe and a colonized Africa and Asia shaped each other over time. Not only does such an approach obscure the details of colonial history and the experience of people in colonies, but the aspirations and challenges posed by political movements in the colonies over the course of history disappear beneath the ironic gaze that critics have directed toward claims for progress and democracy.

The refusal to leave the "colonial" as a neatly bounded, excisable dimension of European history marks an important challenge to historical analysis. Yet unbounding colonialism risks leaving us with a colonial project vaguely situated between 1492 and the 1970s, of varying contents and significance, alongside an equally atemporal "post-Enlightenment" Europe, missing the struggles that reconfigured possibilities and constraints across this period. This is why a reconsideration of colonialism's place in history should both engage deeply with the critical scholarship of the last two decades and insist on moving beyond the limitations that have emerged within it.

Europe's ambivalent conquests—oscillating between attempts to project outward its own ways of understanding the world and efforts to demarcate colonizer from colonized, civilized from primitive, core from periphery—made the space of empire into a terrain where concepts were not only imposed but also engaged and contested. From the very moment of the French Revolution, rebels in the plantation colony of Saint Domingue raised the question of whether the Declaration of the Rights of Man and of the Citizen applied to the French empire as well as the French nation, and in so doing, they, as Laurent Dubois puts it, "'universalized' the idea of rights."1 Ever since, political activism in and about empire has posed not only possibilities of accepting or rejecting the application to colonial worlds of ideas and structures asserted by Europe, but also the possibility, however difficult, of changing the meaning of the basic concepts themselves.

Conceptual issues are the focus of this book. How can one study colonial societies, keeping in mind—but not being paralyzed by—the fact that the tools of analysis we use emerged from the history we are trying to examine?

Interdisciplinarity and the Conformism of the Avant-Garde

Historians' quite recent interest in colonial situations owes much to the influence of literary studies and anthropology; scholarly work on colonial issues gave rise to a cutting-edge interdisciplinary field of scholarship. Yet the basic problem with interdisciplinary scholarship is the same as that within the disciplines: conformism, gatekeeping, conventions that one should publish in the "right" journals—whether the American Political Science Review or Social Text—and cite the right people, be they Gary Becker or Homi Bhabha. The economist—to take the most theoretically monolithic of the disciplines within the American academy—generally has to write within the confines of neoclassical theory and to devise and test abstract models; he or she gets little credit for fieldwork into the complexities of actually experienced economic relations. In cultural studies, the assistant professor is required to decenter, destabilize, and disrupt socially constructed categories and to empower subaltern discourse. To transgress the norm of transgressivity is to be unaware of one's own positionality. The cultural critic may relish her disciplinary hybridity yet have a great deal in common with the economist who thinks that more work within neoclassic models has a higher marginal utility than an excursion into anthropology. Interdisciplinary studies can be impoverished by once provocative constructs that have become clichés, just as a discipline can be narrowed by professional hierarchies, required methodologies, or theoretical conservatism.

The urge to conform is evident in some favorite phrases of scholars charting trends: the "cultural turn," the "linguistic turn," and the "historical turn." These expressions imply that scholars in history, cultural studies, or the social sciences take their intellectual curves together, and anyone who does not is off on a tangent or has entered a dead end. The cultural turn of the 1980s and 1990s corrected to a significant extent the excesses of a previous turn, toward social history and political economy in the 1970s, but after a time scholars were told that we were "beyond the cultural turn," which meant—as some of the more thoughtful participants in these discussions frankly put it—bringing back questions of social and economic history. Excellent research and valuable reflection came out of the cultural turn, as from previous and subsequent turns.2 Meanwhile, however, a generation of graduate students experienced pressure from their mentors and peers to focus their work in one direction, just as a previous generation had been influenced to conform to a different trend. In African history, my generation avoided colonial history for fear of being thought to do "white history"—and contributed thereby to the doldrums of imperial history of which many later complained—whereas now the history of Africa before the European conquests is neglected. Scholars' openness to new ideas and directions is one thing, taking "turns" together another.3

Interdisciplinary studies have their own pitfalls, in particular credulity toward other fields that do not apply to one's own, such as the historian's belief that a quotation from Geertz means doing anthropology or that a reference to Bakhtin means mastery of literary criticism. One is likely to fall for conventional wisdom in another discipline, miss internal debates, and pick up tidbits without exploring their relationship. The remedy for these difficulties of interdisciplinary work, however, is not disciplinarity but discipline: a more thorough and critical engagement with other fields, a more rigorous and wider reading of social theory that both reconfigures and deepens methodological understandings.

Writing on colonialism in the last two decades has had a double—and positive—impact in regard to established verities: calling into question a narrative of progress radiating from Europe that ignored how deeply this history was entwined with overseas conquest, and rejecting the consignment of "non-Europe" to static backwardness regardless of how those regions' fates were shaped by interaction with Europe, including the sidetracking of other modes of change and interaction. The bandwagon effect within colonial studies or postcolonial theory is probably no more severe than in other areas of academic inquiry, but rather is illustrative of a wider problem in intellectual life. Like other new fields, colonial studies has been the object of a dismissive backlash that ignores the insights and the healthy debate within the field—indeed, the considerable heterogeneity that characterizes writing on colonial subjects.4 I hope in these pages to steer between the conformism of the avant-garde and the dismissiveness of the old regime in the study of colonization, colonial history, and decolonization by focusing on specific conceptual and methodological issues.

Bashing the Enlightenment and criticizing modernity have become favorite activities within colonial and postcolonial studies. Such positioning has been answered by a defense of modernity and Enlightenment against the barbarians at the gates who threaten the universal principles on which democratic societies are based.5 Debate at such levels of abstraction is unedifying, not least because both sides are content to treat Enlightenment rationality as an icon separated from its historical significance. There is a delicious irony here, for Europeans become the "people without history," a notion once reserved for the colonized. Both sides are content to let unchanging and unmediated images of reason, liberalism, and universality stand in for a much more convoluted trajectory, in which the status and the meaning of such concepts were very much in question.6 The not-so-delicious irony is that the critique of modernity aimed at destabilizing a smug, Europe-centered narrative of progress has ended up preserving this category as a defining characteristic of European history to which all others must respond. Only a more precise historical practice will get us out of the involuted framing of such a debate.

In chapter 2, I take up the paradox noted at the beginning of this essay, that scholarly interest in analyzing colonialism peaked at a time when it was no longer a political issue. Its starting point is Georges Balandier's article of 1951, "The Colonial Situation," which was a call for analysis of colonial rule using tools perfected in studying indigenous groups but now directed at the "totality" of coercive, structural, and ideological mechanisms of colonial power. This call—timely as it was—went largely unanswered, because scholars, including Balandier himself, were more fascinated by the possibilities of modernizing societies that had been held back and by the liberation movements themselves. My essay surveys the changing focus of scholarship on colonial societies in the half-century since Balandier's intervention, not as a succession of turns, but as overlapping and often conflicting perspectives, all in relation to the shifting politics of decolonization.

Part 2 of this book turns to key concepts that epitomize the current direction of scholarship—in colonial studies and other interdisciplinary endeavors. The use of these concepts has provoked new thinking and important research, but they deserve a scrutiny that the bandwagon effect of scholarly trends has to a large extent repressed. I will examine in detail three concepts—identity, globalization, and modernity—and later in this introduction raise questions about concepts like coloniality, postcoloniality, and post-Enlightenment rationality. In questioning the analytic value of such concepts, my intent is not to step away from the objects of inquiry envisaged by those who use these concepts, but rather to ask if they are adequate to the work at hand.

Identity, globalization, and modernity occupy a large and growing place in scholarly fashions. Figure 1 shows how often these terms have appeared as keywords in a leading web-based index of scholarly articles over the past decade, while references to the buzzwords of a prior era, like industrialization, urbanization, and modernization, have stagnated at lower levels.7 Identity wins the prize, and if modernity isn't as "in" as identity, it passed modernization—a related concept with a different valence—in 1995. [Figure 1]

The use of such concepts addresses important subjects: subjectivity and particularity in people's collective vision of themselves, the apparently increasing importance of cross-border interaction in today's world, and the apparent power—for good or for evil—of a view of historical change as moving in a forward direction. In all three cases, I argue, the concepts are important as indigenous categories, as terms used in today's politics and culture. They need to be understood in the often conflicting ways in which they are deployed. The problem comes with scholars' widespread use of these terms as analytic categories, as tools for description and analysis. This usage does more to obscure than to illuminate the problems of social connection, cross-border interaction, and long-term change that they are thought to address. There is nothing inherently wrong in using the same term as both an analytic category and an indigenous one, but there are two problems that need to be confronted if one does so. First, the usefulness of an analytic category doesn't follow from its salience as an indigenous one: such concepts must perform analytic work, distinguishing phenomena and calling attention to important questions. Second, the academic's endeavor to refine and sharpen analytic categories may obscure the ways in which historical actors deployed similar terms, thereby complicating the task of understanding forms of discourse in their own contexts.

These chapters address not just the words as such—although in all three cases academic language adds confusion to ordinary English definitions—but the conceptual questions to which writing about them gives rise. To question the analytic usefulness of the category identity is not to presume that people's particularistic and subjective concerns—about gender, ethnicity, or any other form of affinity—should be downplayed in favor of the great universalisms, be they the liberal idea of a citizenry of equivalent individuals or the Marxist idea of class. But understanding how people conceive of commonality, belonging, and affinity does require a precise and differentiated set of concepts.

Much recent scholarship on identity uses the same word for something that is claimed to be general but soft—that is, everybody seeks an identity, but identity is fluid, constructed, and contested—and for something that is specific and hard, that is, the assertion that being "Serbian," "Jewish," or "lesbian" implies that other differences within the category should be overlooked in order to facilitate group coherence. This contradictory usage leaves us powerless to examine what scholars most need to understand and explain: why some affinities in some contexts give rise to groups with a hard sense of uniqueness and antagonism to other groups, while in other instances people operate via degrees of affinity and connection, live with shades of grey rather than white and black, and form flexible networks rather than bounded groups. In chapter 3, written by Rogers Brubaker and myself, we do not argue for a more refined or precise word to replace identity, but rather for the use of a range of conceptual tools adequate to understand a range of practices and processes.

With globalization and modernity, we again encounter two words and two bodies of scholarships that confuse normative and analytic categories and reinforce the metanarratives that they pretend to take apart. It is hard for anyone who lived through the modernization debates of the 1970s to read the globalization and modernity debates without a sense of déjà vu. The idea that people were being liberated from the stultifying edifice of colonialism or the backwardness of tradition—producing a convergence toward the social practices and living standards of the West—was the hallmark of modernization theory in the 1950s and 1960s. More recently, some pundits and scholars insist that globalization is inevitable as well as desirable. Critics again decry as malignant what advocates insist is beneficial, while some scholars accept the narrative of ever-increasing interaction but deny that it is producing convergence. My argument is neither for nor against globalization; rather, I attempt to reframe the issue, pointing out that the globalization story claims as new what is not new at all, confuses "long-distance" with "global," fails to complement discussion of connections across space with analysis of their limitations, and distorts the history of empires and colonization in order to fit it into a story with a predetermined end.8 The alternative to the concept of globalization is not to reify the state or any other container of interaction, but to detach mechanisms of connection from the artificial notion of globality and to study the marking of territory and the crossing of territorial boundaries in more specific ways than those implied by the linear concept of globalization.

The critique of modernization theory that emerged in the 1970s brought out the teleological and Eurocentric nature of the theory. But if the teleology is gone, the telos remains in the form of a burgeoning literature on modernity, colonial modernity, and alternative modernities, the former two with a negative valence instead of a positive one, the latter as the positive, non-Eurocentric reflection of the others. In chapter 5, I argue that the modernity now in question is hopelessly confused by the divergent meanings given to it and that any effort to refine the analytic concept would result in loss of the ability to understand the meanings of modern as an indigenous category—where it was in fact used. The appeal of the modernization concept in the 1970s was above all that it constituted a package, pulling together such changes as urbanization, the growth of market economies, and achievement-oriented status systems. Modernity in the 1990s was still a package, sometimes decried instead of celebrated, sometimes repackaged as "alternative modernities," but still assuming that the alternatives must be modernities. When Partha Chatterjee talks about the "bitter truth" that no one in Europe believes that Indians "could be producers of modernity," he concedes that modernity is what Europe produced.9 The package is still on its pedestal, and debate about a wide range of issues—from the equality of women in society to the desirability of free markets—will be conducted in relation to a presumed distinction between modern and backward rather than in more specific and less teleological terms.

As scholars, we need to understand what people mean when they engage in identity politics, when they argue for the inevitability and desirability of the global marketplace, or when they articulate aspirations for clean water and better education. We also need to develop a precise and incisive vocabulary for analyzing affinity, connections, and change. We should try to explain why such concepts evoked passions at some moments but not at others. Colonial elites—sometimes—claimed legitimacy on the grounds that they were remaking Asian or African societies in the image of Europe's self-proclaimed modernity, and at other times they insisted that colonies could never be modern, that they would only go astray if their status hierarchies were undermined, and that European rule was necessary to preserve this conservative order. Such arguments are best analyzed as debates within the history of colonization rather than as a "colonial modernity" located vaguely between the Enlightenment and the present. Understanding indigenous categories—be they those of a French colonial minister, an African trade unionist, or an Islamic religious leader—requires asking how people put their thoughts together; in other words, scholars must make an effort to get out of their own categories.

Part 3 develops alternatives to the flattening of time, space, and interaction in the concepts considered above, first via a general argument and then through a case study. Chapter 6 argues that instead of telling a story of the inevitable rise over the last two centuries of the nation-state and the national imagination, one can tell a more revealing story by looking over a longer period of time at a more varied set of political forms. For imperial rulers from the Roman Empire through the Ottoman and Austro-Hungarian Empires to the French Community and the British Commonwealth, governing an imperial polity produced a different set of structures and a different way of imagining political space than did a nation-state. Empires should not be reduced to national polities projecting their power beyond their borders. They always had to balance the incorporation of people and territory and differentiation that maintained the power and sense of coherence of the ruling elite. The chapter puts in a single framework continental and overseas, "modern" and "premodern," European and non-European empires, for all participated in the calculus of balancing incorporation and differentiation, and interacted and competed with each other for resources—but did so in different ways.

Similarly, there is much to learn by looking at political mobilization within and against empire not just in terms of a community or nation rallying against an intrusive, distant power. Political movements developed more varied repertoires, including deterritorialized forms of affinity—pan-Africanism, pan-Slavism, pan-Arabism, Islamism, Christian humanitarianism, proletarian internationalism—as well as attempts to reform and restructure the imperial unit itself, often by turning imperial ideology into a claim on the rulers of empire. It was only with the collapse of the last empires in the 1960s that the nation-state became the generalized form of sovereignty. Up to the very end of those empires, some people within them were trying to turn the empires' incorporative needs into demands for imperial resources and for political voice. Empire is, unregrettably, no longer in the political repertoire, but appreciating the recent roots of the nation-state might help to foster more precise discussion of different forms of political organization and their consequences, without falling into the teleology of nation-building, sweeping denunciations of all forms of state power, the use of empire as an epithet for any form of power, or the sentimental reimagining of past empires as models of stern and responsible governance of the fit over the unfit.

Chapter 7, based on my research in Senegal and France, provides an example of how both the makers of empire and the leaders of social movements operated within an imperial framework and by using that framework changed it. Labor and political movements in French West Africa in the 1940s and 1950s seized the language of postwar French imperialism—in a moment when France needed more than ever for colonies to be orderly, productive, and legitimate—and turned it into demands for equality of wages, benefits, and ultimately standard of living among all people whom the government asserted to be French. This impeccable logic of equivalence—backed by well-organized protest movements and in the context of worldwide debates over self-determination and anticolonial revolutions in Vietnam and North Africa—presented the French government with the dilemma of either giving up the idea of Greater France or facing its metropolitan citizens with never-ending demands and an unpayable bill. The national conception of France was born of the same process that give rise to nation-states in North and sub-Saharan Africa.

Critical History and Ahistorical History

The arguments presented here are historical. They do not, however, imply a polarization between a domain that might be called colonial studies—or more generally, interdisciplinarity—and another called history. Such a division would mask the extensive differences and debate within all such designations, as well as the cross-fertilization across whatever lines scholars use to mark their territory. My goal is not to criticize any scholarly field as a whole, or even to pin down exactly what such field labels signify, but instead to focus on key concepts themselves, to assess the work they do, the blind spots as well as insights they entail, and the difficulties of using them to examine change over time.10

The historical profession has without doubt been reinvigorated by challenges to it, coming from new entrants into the academy—not least of all, scholars from Africa and Asia—by ferment in other disciplines, and by the tense but frequently crossed frontier between academic history and diverse people's interest in the past. In my experience and that of many of my generation of professional historians, the study of colonial empires had by the 1970s become one of the deadest of dead fields within history. Students interested in pushing the frontiers of historical research looked to Africa, Asia, or Latin America, or they sought to look at Europe and North America "from the bottom up." The revival of interest in the colonial world a generation later reflects the influence of literature and anthropology and, most importantly, wider intellectual currents that threw into question the most basic narratives and the most basic ways in which knowledge is configured. Historians were having to face the fact that the new challenges were not simply to add an African or Asian component to a previously Europe-centered curriculum, but to think about what we mean by Europe, Africa, Asia, and how they shaped each other over time (see chapter 2).

But it is now the interdisciplinary domains of colonial and postcolonial studies that could use a new sense of direction, particularly a more rigorous historical practice. These fields of inquiry have introduced a large and transcontinental public to the place of colonialism in world history. Yet in much of the field, a generic colonialism—located somewhere between 1492 and the 1970s—has been given the decisive role in shaping a postcolonial moment, in which invidious distinctions and exploitation can be condemned and the proliferation of cultural hybridities and the fracturing of cultural boundaries celebrated.

Meanwhile, historians can at times be faulted for treating own engagement with sources from the place and time in question as unproblematic, as if sources spoke on their own. The outsider's characterization of academic history as one damn thing after another has a grain of truth. Historians' narratives are built on conventions of narrativity that are not always examined. Nevertheless, the historian's displacement in time itself generates a bias against the homogenization of categories; while some historians narrate the past as if it inevitably led to the present, they still distinguish past from present, and another historian in the same present might interpret that past differently. Historical practice suggests that however varied the impetus and context for the actions of men and women, interactions unfold over time; contexts are reconfigured and shape future possibilities and closures.

At least some of the criticism has had a positive effect. The June 2004 congress of the once staid and nationally-focused Society for French Historical Studies included seventeen panels on topics related to colonial history, with nearly four dozen presentations, mostly by young historians with fresh material from archives and other sources that enlarged from colonial vantage points the meanings of citizenship, law, social welfare, and "France" itself. In the following pages, I will point both to the importance of the critique of the historical profession and to its limitations, especially when ahistorical methodologies are deployed to answer questions that are unavoidably historical.

Ashis Nandy argues that history is inseparable from its imperialist origins, that it necessarily imposes the imperialist's understanding of people's past over their own. To some scholars, history confines the zigzags of time into linear pathways, privileges state-building over other forms of human connection, and tells a story of progress that inevitably leaves Africans or Asians on the side, lacking some crucial characteristic necessary to attain what is otherwise universal.11 Such arguments are valid criticisms of many histories, but do they amount to an indictment of the study of history itself? In fact, the indictment of history is itself historical. To trace history to imperialism is to give power to a phenomenon that is historically located. The question such an observation leaves is whether it is enough to name imperialism as the dark side of modernity, or if understanding it requires a more searching examination, which in some form is historical. Meanwhile, the practices of many historians may well suggest an "irrevocable link between History and the Nation-State," but the evidence that the nation-state is not so universal is another sort of history, which documents more varied sorts of political imagination.12 Academic history, like all others, has its particularities, and the argument that other visions of the past are more diverse and lively is valid only if one aggregates them—itself a quintessentially academic exercise.

Historians' complacency about the European boundaries of their field was shaken up by Edward Said's Orientalism (1978). Said showed how certain visions of Asiatic societies are deeply woven into canonical European literature. Colonization was no longer out there, in exotic places, but in the heart of European culture. Said was soon criticized for presenting such a closed view of the colonized "other" that there was no room for alternative constructions, including those by Arabs, Africans, or South Asians. In his subsequent book, Culture and Imperialism, Said tried to restore balance by emphasizing not the stark separation of European and indigenous discourses but the efforts of colonized intellectuals to work between them and to develop crosscutting languages of liberation.13 Such an argument, too, is a historical one.

The Saidian view of Europe constructing itself and its others in relation to each other has had wide influence in many disciplines and has fostered an examination of those disciplines. The categories used by social scientists from the nineteenth through the twenty-first century to examine colonized societies have been shown to be less a neutral means of analysis of bounded societies located elsewhere than part of a process of intellectual pacification and ordering of the world. Vocabularies and methods of elite control for handling distinctions of gender, class, and race—of the respectable and the civilized in contrast to the unruly and the dangerous—were developed in both metropoles and colonies. Esthetics and science helped order an imperial world. The scholarship on such subjects in the last quarter century adds up to a impressive reconsideration of intellectual and cultural history. The question it leaves is that which Said faced after Orientalism: whether such work will be read as a solid edifice of colonial modernity or colonial governmentality imposed from Europe, or whether it will be seen as a framework for contestation and debate over the nature of social distinctions and social knowledge across the colony-metropole divide.14

To some postcolonial theorists, the goal has been no less than to overthrow the place of reason and progress as the beacons of humanity, insisting that the claims to universality that emerged from the Enlightenment obscure the way colonialism imposed not just its exploitative power but its ability to determine the terms—democracy, liberalism, rationality—by which political life the world over would be conducted from then on. By contrasting this universalizing modernity with the ugly particularity of colonialism, postcolonial theorists attack head-on a historical metanarrative that shows Europe repudiating step by step the oppressiveness of its own past and making itself into a model for the rest of the world. Some hope to persuade us to "give up the seemingly powerful corollary presumption that liberalism and indeed democracy (even a purportedly radical one) have any particular privilege among ways of organizing the political forms of our collective lives."15

Before we give such ideas up, we would do well to examine carefully not only what they are, but how they have been used—and perhaps, in being used by people in colonies, given a new meaning. We should be careful about what else we might be giving up: perhaps the tools with which to analyze and critique various forms of oppression, from local patriarchies to global capitalism?16

My focus is on the double occlusion that results from turning the centuries of European colonization overseas into a critique of the Enlightenment, democracy, or modernity. First is the obscuring of European history, for the counterpart of reducing non-Western history to a lack of what the West had is to assume that the West actually had it. All the debate and conflict within post-1789 European history is reduced within the critique of the post-Enlightenment to an essence of modernity, producing a label attached to an entire epoch, and this abstraction is assigned causal weight in shaping what happened in colonies in the nineteenth and twentieth centuries. Second is the occlusion of the history of the people who lived in colonies. Positing a colonial modernity (see chapter 5) reduces the conflicting strategies of colonization to a modernity perhaps never experienced by those being colonized, and gives insufficient weight to the ways in which colonized people sought—not entirely without success—to build lives in the crevices of colonial power, deflecting, appropriating, or reinterpreting the teachings and preachings thrust upon them. Within this line of argument, resistance might be celebrated or subaltern agency applauded, but the idea that struggle actually had effects on the course of colonization is lost in the timelessness of colonial modernity. The Haitian Revolution—and especially the possibility that the Haitian Revolution actually affected the meanings of citizenship or freedom in Europe and the Americas—is as strikingly absent in prominent postcolonial texts as in conventional narratives of European progress.17 The result is that ownership of notions like human rights and citizenship is conceded to Europe—only to be subjected to ironic dismissal for their association with European imperialism.

The "colonial" of postcolonial studies is often the generic one, what Stuart Hall sweeps together in a single phrase—"European and then Western capitalist modernity after 1492." It is spatially diffuse and temporally spread out over five centuries; its power in determining the present can be asserted even without examining its contours.18 But might not this generic colonial history produce an equally generic postcolonial present?19

I agree with the postcolonial critic's insistence that the evils of nineteenth- and twentieth-century colonialism lie firmly within the political structures, values, and understandings of its era; colonialism was not an atavistic holdover from the past. Less convincing is the juxtaposition of post-Enlightenment universality and colonial particularity isolated from the dynamics ensuing from the tensions within any ideological formation and from the tensions produced by efforts of empires to install real administrations over real people. Such an approach privileges the stance of the critic, who decodes this transhistorical phenomenon; hence the label Gyan Prakash and others have attached to their project: "colonial critique."20

Such a critique has had its value, above all in forcing historians—like anthropologists or other social scientists—to question their own epistemological positions. The question is how one understands and gets beyond the limits inherent in the stance of the critic. Let me turn now to a brief analysis of modes of writing that can be called ahistorical history, which purport to address the relationship of past to present but do so without interrogating the way processes unfold over time. I will mention four modes of looking at history ahistorically: story plucking, leapfrogging legacies, doing history backward, and the epochal fallacy. My purpose is not to defend one discipline or condemn another, for some of the most searching historical questions have been asked by literary critics or anthropologists. Historians are familiar with many ways of doing history ahistorically, not only from criticizing the shortcomings of other disciplines but from engaging in such practices themselves. Nonetheless, theoretical perspectives that operate in vaguely specified temporalities and that give explanatory weight to agentless abstractions—like coloniality and modernity—both depend on and reinforce the methodological shortcomings described below.

Story Plucking

The "colonial" has itself become an object of study, literary and otherwise—a phenomenon appearing in many places and times. The weighty -ity in such widely used words as coloniality or postcoloniality implies that there is an essence of being colonized independent of what anybody did in a colony.21 One can pluck a text from Spanish America in the sixteenth century, a narrative of the slave colonies of the West Indies in the eighteenth, or a description of moderately prosperous African cocoa planters in the twentieth-century Gold Coast, and compare it to other texts. This gives rise to the question of how far we can go in discussing coloniality when the fact of having been colonized is stressed over context, struggle, and the experience of life in colonies. Colonial power, like any other, was an object of struggle and depended on the material, social, and cultural resources of those involved. Colonizer and colonized are themselves far from immutable constructs, and such categories had to be reproduced by specific actions.

Leapfrogging Legacies

Here I refer to claiming that something at time A caused something in time C without considering time B, which lies in between. African political scientist Mahmood Mamdani, in his Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism,22 draws a direct causal connection between a colonial policy—important in the 1920s and 1930s—of ruling through African chiefdoms given authority under colonial auspices and the brittle politics of authoritarianism and ethnicity in Africa in the 1980s and 1990s. Mamdani has a point at either end of his leapfrog, but he misses what lies in between. His book says almost nothing about the 1950s and 1960s, and thus does not consider another dimension of Africa's malaise: that there was indeed effective mobilization in those years that cut across ethnic divisions and urban/rural distinctions. Through such mobilizations, Africans made strong claims to citizenship. African politicians built a powerful challenge to colonial regimes—either to make good on the implied promises of imperial citizenship or to give way to governments that could truly represent their citizens (see chapter 7). But once in power, such leaders understood all too well how dangerous such claims were. The explosion of citizenship in the final years of colonial rule appears nowhere in Mamdani's book. He thus misses not only the sequence of processes in the decolonization era, but the tragedy of recent African history, people's heightened sense of possibility and the thwarting of their hopes.23

Doing History Backward

Trying to illuminate present issues is a fine motivation for exploring the past, but as one looks backward, one risks anachronism: confusing the analytic categories of the present with the native categories of the past, as if people acted in search of identity or to build a nation when such ways of thinking might not have been available to them. Even more important is what one does not see: the paths not taken, the dead ends of historical processes, the alternatives that appeared to people in their time. Two common, and in many ways meritorious, approaches to historical analysis can easily fall into backward-looking history. One is the idea of social construction, a useful antidote to claims that race, ethnicity, or nationality are primordial characteristics of given groups, and which is also helpful in recognizing that race or any other category may be no less important for having been constructed historically. The trouble with constructivism, as it is most often practiced, is that it doesn't go far enough: we talk of the social construction of racial categories, but it is rare that we even ask about categories that are not now important, and we thus lose sight of the quest of people in the past to develop connections or ways of thinking that mattered to them but not to us.24 The study of nationalism in colonial societies is a case in point: because we know that the politics of the 1940s and 1950s did indeed end up producing nation-states, we tend to weave all forms of opposition to what colonialism did into a narrative of growing nationalist sentiment and nationalist organization. That the motivations and even the effects of political action at several junctures could have been something else can easily be lost.25

At a more abstract level, seeking the genealogy of concepts or ideas also easily turns into a backward-gazing approach to history. Just as an ordinary genealogy starts with "ego" (the person looking backward) and produces a tree of connection, genealogical approaches to ideas look backward to find their roots, sometimes finding them in a discredited colonial past. What gets lost here is the historical context in which concepts emerged, the debates out of which they came, the ways in which they were deflected and appropriated. Genealogical and constructivist approaches when done in a historically grounded way—that is, by working forward—become other words for doing . . . history. To the extent that such approaches both call attention to the non-neutral position of the present-day observer and see the conceptual vision of that observer in historical terms, they are valuable, albeit hardly new.26 Good historical practice should be sensitive to the disjunctures between the frameworks of past actors and present interpreters.

The Epochal Fallacy

Historical analysis can point to moments of uncertainty—when stabilizing institutions were weakened and expectations of change heightened—and to moments of stability, and it can point to change. But to see history as a succession of epochs is to assume a coherence that complex interactions rarely produce. Whatever makes an era distinct should not only be present but be its defining feature; otherwise, the identification of an epoch says little. It is ironic that postmodernists, who distinguish themselves by a refusal of high theory and grand narrative, have to jimmy modernity into an epochal straightjacket in order to claim to have moved beyond it.27 A more nuanced approach involves assessing change in whatever dimensions it occurs and analyzing the significance and limitations of conjunctures when multidimensional change became possible.

The term postwar has a clear meaning if the war in question has ended, and postcolonial is meaningful if one accepts—as I do—that the decolonizations of the postwar era extinguished the category of colonial empire from the repertoire of polities that were legitimate and viable in international politics.28 The post- can usefully underscore the importance of the colonial past to shaping the possibilities and constraints of the present, but such a process cannot be reduced to a colonial effect, nor can either a colonial or a postcolonial period be seen as a coherent whole, as if the varied efforts and struggles in which people engaged in different situations always ended up in the same place. One is not faced with a stark choice between a light-switch view of decolonization—once independence was declared, the polity became "African"—and a continuity approach (i.e., colonialism never really ended), but one can look at what in the course of struggle before and after that moment could or could not be reimagined or reconfigured, what structural constraints persisted, what new forms of political and economic power impinged on ex-colonial states, and how people in the middle of colonial authority systems restructured their ties within and outside of a national political space.29

Skepticism is especially in order in regard to the modern epoch. Modernization theory was justly criticized for claiming that a certain societal form came to define a modern era.30 Era labeling has been given a new interdisciplinary lease on life, in part through the work of Michel Foucault, which locates modern governmentality in a space that is amorphous in time and amorphous in agency and causality, and provides a blueprint for a wide range of scholars to attribute practices and discourses to the fact of modernity, often elided with post-Enlightenment rationalism, bourgeois equality, and liberalism.31

Dipesh Chakrabarty, for example, justly criticizes versions of Indian history, colonialist, nationalist, or Marxist, which measure the colonized by how well they succeeded in class formation and state-building—where Europe supposedly led the way—and attribute their failures to certain lacks on their part (of a proper working class, of a proper bourgeoisie). He instead calls for the "provincialization" of Europe, its history seen as particular rather than as a universal model.32

Then he proceeds to do the opposite. Post-Enlightenment rationality, bourgeois equality, modernity, or liberalism become not provincial ideologies but a grid of knowledge and power, forcing people to give up diverse understandings of community in favor of a one-to-one relationship of the unmarked individual and the nation-state, at best seeking "alternatives" to a modernity that is decidedly singular and decidedly European. European history is flattened into a single post-Enlightenment era. A reference to Hegel stands in for a European history reduced to the claim of progress.33

Yet nineteenth-century Europe was immersed in struggles within and among many parochialisms and many universalities. Secularism was more often beleaguered than triumphant; ancien régimes and aristocracies didn't die out on the guillotine.34 The balancing of the universalized rights-bearing individual against questions of "difference" was a vital debate within and after the Enlightenment. Intellectuals who called themselves modernists between 1890 and 1930 were in "revolt against positivism, rationalism, realism, and liberalism," something lost in the stark opposition between Enlightenment reason and the "posts" in vogue today.35

Sankar Muthu has brought out the debate over empire among Enlightenment thinkers. For Diderot, most notably, coming to grips with the humanity of non-European people also meant confronting their subjection. Rather than seeing universal values as effacing difference, Diderot insisted on the fundamentally cultural nature of humanity. Others, like the Abbé Grégoire, were deeply sympathetic to slaves and the other victims of imperial oppression, but assumed that people, once liberated, would abandon their particularity. Still others—those most emphasized by critics of Enlightenment reason—advanced taxonomic structures, which in some (but not all) hands implied distinctions that put some outside the realm of the rights-bearing individual and made them a potential subject for colonization. "The" Enlightenment implied no one view of race or difference. It provided no clear basis either for legitimizing the subordination of certain non-European societies on the basis of universalistic criteria or for claiming that cultural difference precluded criticizing different political practices, in Europe or elsewhere.36 What Enlightenment implied in its time—and since—was the necessity of having the debate. The historians' contribution is not to decide which Enlightenment was the authentic one, but to point to the responsibility of those who advanced particular arguments and the consequences of their interventions.

Too ready identification of an actual Europe with post-Enlightenment rationality not only leaves out the conflict and uncertainty within that continent's history, but also the extent to which even such constructs as bourgeois equality were not some essence of the West but products of struggle. The ascension of a liberal idea of a rights-bearing individual over the equally liberal idea of rights as earned by the civilized behavior of a collectivity reflected the labors not only of a Toussaint L'Ouverture or a Frederick Douglass, but of unnamed ex-slaves, dependent laborers, and colonized peasants who revealed the limits of colonial power and defined alternative modes of living and working in the crevices of authority.37

Doing history historically does more to challenge the supposedly dominant narrative of Western-led progress, of nation-building, or of development than an approach to the past based on story plucking, leapfrogging legacies, doing history backward, or the epochal fallacy. Criticisms of historians for writing everything into a linear history of human progress are often accurate and appropriate, but understanding different forms of temporality is not assisted by positing a flattened modern era against the linearity of a history of continual Western-centered progress. Historical temporality, as Williams Sewell puts it, is "lumpy": the tendency for innovations and breaks to be reabsorbed into ongoing discursive and organizational structures is sometimes broken by a cascade of events that reconfigures the imaginable and the conceivable.38 Historical time is lumpy in another sense—across different conceptions of temporality held by different people at the same moment. But if time is plural, it is not divided into self-contained compartments. One circles back to the problem that in order to understand how ideas of history were shaped by colonialism, one has to understand colonization and challenges to it over time. The critical insistence that historians examine their own concepts of time is valuable, but so too is the historian's insistence on attention to process, on how what happens at one moment in time configures possibilities and constraints on what can happen the next.39

Imperial Space

Can one really provincialize Europe? One way to do so is to dig more deeply into European history itself, and there is no more central myth to be dissected than that of narrating European history around the triumph of the nation-state. Much recent scholarship has exaggerated the centrality of the nation-state in the "modern" era, only to exaggerate its demise in the present.40 Post-Revolutionary France, as I will explain in chapter 6, cannot be understood as a nation-state pushing into colonies external to it. The Haitian Revolution of 1791 revealed how much questions of slavery and citizenship, of cultural difference and universal rights, were part of debate and struggle across imperial space.41 This complex, differentiated empire, expanded into continental Europe by Napoleon, did not produce a clear and stable duality of metropole/colony, self/other, citizen/subject. Political activists in the colonies, until well into the 1950s, were not all intent upon asserting the right to national independence; many sought political voice within the institutions of the French Empire while claiming the same wages, social services, and standard of living as other French people. If one wants to rethink France from its colonies, one might argue that France only became a nation-state in 1962, when it gave up its attempt to keep Algeria French and tried for a time to define itself as a singular citizenry in a single territory.

A fuller version of the story of European colonial empires in the nineteenth and twentieth centuries can also come from telling it alongside the histories of the continental empires with which they shared time and space, the Habsburg, the Russian, and the Ottoman, and those empires that lay outside Europe, notably the Japanese and the Chinese, not to mention two powers with wide reach and an ambivalent sense of themselves as imperial powers: the United States and, after 1917, the Soviet Union. At times colonialism was layered: late-nineteenth-century Sudan, for instance, was colonized by Egypt, which was part of the Ottoman Empire but itself experienced heavy British intervention.42 The sharp separation of a certain kind of empire—which produces colonial and postcolonial effects—not only precludes the posing of important questions about critical historical moments and interrelated processes, but reproduces a form of Eurocentrism. Central Asian Muslims conquered by the tsars and subjected to the violent and modernizing project of the Soviets do not receive the attention devoted to North African Muslims colonized by the French; 1989 is not marked in postcolonial circles as a milestone of decolonization.43

The narrowing of the range of inquiry is based on certain assumptions: that these empires were not truly colonial, and above all that they were, except for the Soviet case, not "modern." The latter argument reads backward the collapse of the Ottoman, Habsburg, and Russian Empires in 1917-23 into a thesis of the inevitable transition from empire to nation-state. But excellent historical research has shown that far from being beleaguered holdouts against claims to the nation, these empires produced a strong empire-centered imagination that captured the minds of many self-conscious minority activists within their territories until World War I, a theme developed in chapter 6.

At the heart of colonialism, Partha Chatterjee has argued, is the rule of difference.44 It might be more useful to emphasize the politics of difference, for the meanings of difference were always contested and rarely stable. As broad comparative study suggests, all empires, in one way or another, had to articulate difference with incorporation. Difference had to be grounded in institutions and discourses, and that took work. "Modern" empires were in some ways more explicit about codifying difference—and particularly codifying race—than aristocratic empires, for the giving way of status hierarchies to participation in a rights-bearing polity raised the stakes of inclusion and exclusion. Just where lines of exclusion would be drawn—in terms of territory, race, language, gender, or the respectability of personal or collective behavior—was not a given of the "modern state," but rather the focus of enormous and shifting debate in nineteenth- and twentieth-century Europe. The openings and closures of such debates deserve careful examination.45

New imperial endeavors confronted the dilemmas of older ones: geographic dispersion, extended chains of command, the need to make use of regional economic circuits and local systems of authority and patronage. The most technologically sophisticated, bureaucratized, self-consciously rational empires were compelled to give elites of conquered and subordinated people a stake in the imperial system and to produce subordinates and intermediaries who also had a stake in the system, a problem also faced by the Romans and the Ottomans. The most powerful empires were often in danger of being hijacked by their agents, by settlers, or by indigenous collectivities in search of alternatives to cooperation with an imperial center. Within empires, Enlightenment thought, liberalism, and republicanism were neither intrinsically colonial nor anticolonial, neither racist nor antiracist, but they provided languages of claim-making and counter claim-making, whose effects were shaped less by grand abstractions than by complex struggles in specific contexts, played out over time.

Ideologies of imperial inclusion and differentiation were challenged by people acting within the ideological and political structures of empire, as well as by people who tried to defend or create a political space wholly outside. At certain moments, empires needed to soften differentiation and enhance incorporation, when the need for colonial soldiers rose—in the French Caribbean of the 1790s or European campaigns of 1914—or at many other moments when people in the middle of relations of authority proved too important to making colonies work, too reflective of the actual ambiguities of colonial societies. At other moments, sometimes in reaction to activism in the colonies, rulers became more intent on articulating a colonizer/colonized dualism, a more national conception of the polity. But such conceptions were as hard to sustain in practice as the fiction of belonging to a unified polity. And colonial elites did not always agree on which direction they should lean. Among colonizing elites—even if they shared a conviction of superiority—tensions often erupted between those who wanted to save souls or civilize natives and those who saw the colonized as objects to be used and discarded at will. Among metropolitan populations, colonized people sometimes provoked sympathy or pity, sometimes fear—as well as the more complex sentiments that emerged during the actual encounters and political struggles in the colonies themselves.

Imperial Space and the Varieties of Political Imagination

The backward projection of the post-1960s world of nation-states into a two-century-long path of inevitability affects our understanding not only of the relationship of national and imperial regimes but of the diversity of opposition to them. Pan-Arab, pan-Slavic, and pan-African movements put political affinity into a nonterritorial framework. Territory-crossing politics today, far from being a new response to a new "globalization," have a long pedigree—and, beginning with antislavery movements, a record of some effectiveness.

As I will argue in the final essay in this volume, there is a danger that ahistorical history encourages an apolitical politics. To write as if "post-Enlightenment rationality" or "the cunning of reason" or the "insertion of modernity" were what shaped the political possibilities of colonial situations is to give excessive weight to the determining power of agentless abstractions and offer little insight into how people acted when facing the possibilities and constraints of particular colonial situations. We lose the power of their example to remind us that our own moral and political choices, made in the face of the ambivalences and complications of our present situation, will have consequences in the future.

The view of an atemporal modern colonialism goes along with a notion of resistance as heroic but vain. Only at the end, in some views at least, could it have much effect—in an anticolonial moment in which iconic figures like Nkrumah and Fanon stand in for an epoch. But the heroic moment proved ephemeral, and much of the impetus behind postcolonial theory has been the failure of decolonized states to fulfill an emancipatory project—a disillusionment that then turns its critique toward the emancipatory project itself, now seen as fatally linked to its imperial genealogy.46 The view expounded in these pages acknowledges the impetus behind this version of postcolonial theory but takes a different view of the history. I argue that colonial regimes and oppositions to them reshaped the conceptual frameworks in which both operated. Struggle was never on level ground, but power was not monolithic either. The intersection of locally or regionally rooted mobilizations with movements deploying a liberal-democratic ideology, with attempts at articulating a Christian universalism, with the mobilization of Islamic networks, with the linkages of anti-imperialist movements in different continents, or with trade union internationalism helped to shape and reshape the terrain of contestation. Collaborators and allies of colonial regimes—or people simply trying to make their way within empire—also pushed rulers of empire to change the way they acted. Subtle and dramatic changes at critical conjunctures are both part of the story.

The conjuncture of the post-World War II era indeed produced a situation in which longer-term political processes, with diverse goals, focused on fundamentally transforming the colonial state. Revolutionary mobilization, especially in Indonesia and Vietnam, as well as the climax of India's nationalist movement, had effects well beyond the immediate territories involved. But attempts at change within empires had a profound effect too (see chapter 7), for the danger that social movements operating within imperial frameworks could effectively make demands upon colonial states for resources equivalent to those of the other—metropolitan—members of the polity raised the question of whether a postwar empire could aspire to legitimacy without taking on an impossible burden of social and economic expenditures, with the threat of violence lying behind the demands. That such demands were phrased in a language of citizenship, progress, democracy, and rights both reflected social movements' serious engagement with the categories of colonizers and profoundly changed the meaning of those categories because of who was speaking. At the same time, movements outside of such frameworks—sometimes denounced by colonial rulers as atavistic, demagogic, or antimodern—raised the stakes for colonial regimes to contain tensions within familiar institutions and allowed African political movements room to maneuver between different visions of the future. One needs to appreciate the sense of possibility of these years and to understand what ensued not as an imminent logic of colonial history but as a dynamic process with a tragic end.

Empires, Colonies, and the Politics of Naming

For many postcolonial theorists, the naming of the colonial makes a point with relevance beyond the specificities of bygone regimes. Doing so links the history of the West and its identification with civilization and progress to its colonial genealogy. The colonial evokes above all the marking of certain people as distinct, in need of special forms of surveillance and supervision, and unable to participate fully in the projects of a modernizing society. The colonial phenomenon is thus located broadly—it may appear within "national" territory as well as in institutions of empire.47 The use of such a general conception has its costs: a diminished ability to make distinctions among the various forms of discrimination and exclusion and a tendency to look away from the actual histories of colonization toward a homogenized coloniality. Politically as well as analytically, a more precise use of categories may be enabling.

Hence the potential value of leaning away from a dilute use of the concept of the colonial and toward a focus on the institutionalization of a set of practices that both defined and reproduced over time the distinctiveness and subordination of particular people in a differentiated space.48 Hence the importance of concepts that bring together a range of polities across time and space all sharing basic characteristics, all the while emphasizing distinctions among them and change over time. We can set out a family description of empire, if not a precise definition: a political unit that is large, expansionist (or with memories of an expansionist past), and which reproduces differentiation and inequality among people it incorporates. The extent to which difference across space is institutionalized is important to constituting empire. Empire could be a phase in a polity, for if incorporation ceased to entail differentiation, it could result in a relatively homogeneous polity that becomes more nation-like and less empire-like—sometimes as the result of extremely brutal tactics of coerced assimilation or extermination, possibly a more gradual (if still asymmetrical and at times violent) process of mixing.49 Nation-states and empire-states are, first of all, states, and power is unevenly distributed in all kinds of states.50 In empires, power does not necessarily cohere in a core collectivity or a "people," for all members of the polity might be subordinated to greater or lesser degrees to a monarch, dictator, oligarchy, or lineage. An empire-state is a structure that reproduces distinctions among collectivities while subordinating them to a greater or lesser degree to the ruling authority.51

How starkly should colonial empires be separated from other types of empire? At stake in such a question is how one thinks about an institutionalizion of distinction that it is collective, invidious, and spatial, the marking of particular people as subject to distinct regimes of discipline and exploitation. But let us back up a moment. The spatial referent of colonization goes back to Greek and Roman meanings of the word—the bringing of new territory into use by an expanding society, including settlements for trade and agriculture. Such a referent remained part of the word's significance into the twentieth century, so that French officials, for example, could write about—without deploying an oxymoron—indigenous colonization in Africa, that is, the movement of African peasants onto new land.52 But the principal meaning of colonization has come to involve people rather than land: coercive incorporation into an expansionist state and invidious distinction. The political salience of the colonial has been sharpened by the addition of an "ism": either an accusation—set against the alternative of a more inclusive, more consensual polity—or a defense of the legitimacy of a polity in which some people ruled over others. The power of both accusation and defense lay in bounding the colonial phenomenon to make it appear to be an exceptional form of political organization. Here definitional exercises need to enter the historical realm. Maintaining the colonial required coercive and administrative work and cultural work—to define hierarchies and police social boundaries. Such work was always subject to contestation, by those who sought to exit from the colonial polity or to make the polity less colonial.53

Was empire in the nineteenth and twentieth centuries more colonial than its predecessors? Brutality, enslavement, land grabbing, the denigration of indigenous cultures, and coerced religious conversion are not unique to any era or place. The more profound argument lies both in a supposedly post-Enlightenment penchant for classification—and hence invidious distinction based not on the give and take of relations between unequals but on systemic rankings of peoples (see above)—and, more persuasively, in the contention that as European publics claimed rights and citizenship for themselves, they defined a sharper division between a metropolitan polity for which such claims were relevant and an external sphere for which they were not. Subordination was no longer a fate to which anyone might be subject, but a status assigned to specific people, whose marking therefore became an issue. Overcoming such marking required evidence of acquiring the prerequisites of inclusion, hence the importance both of civilizing missions and of tightly controlling the passage from one status to another. There is something in this long-term view of a shift toward sharper distinction between a potentially democratic imperial core, located in Europe, and a colonial periphery, where access to rights, if attainable at all, required evidence of personal transformation. Even some of the old empires—the Russian and the Ottoman, for instance—began to act more colonial in the late nineteenth century, trying to impose an imperial civilization along the edges of empires, although constrained by the practical necessity of working with local elites.54

But if empires could become more colonial, could they get away from the dilemmas of still being empires? I will argue that they could not, because of both the old problem—administrative and political constraints intrinsic to the vastness and diversity of imperial spaces—and the ambiguities of the spatial referents of new ideologies of rights and citizenship. The old problem would not go away: colonial rulers needed to co-opt old elites and generate new collaborators, but such ties might soften the colonizer-colonized distinction and strengthen the indigenous social and cultural practices colonial ideology was trying to denigrate; rulers hoped at times to profit from indigenous trade networks and productive systems without fostering the autonomy of indigenous economic elites; they needed to raise levels of exploitation without fostering rebellion or undermining local authorities vital to the maintenance of order. The new problem was a question not simply of the ambiguity of rights discourse, but of struggle. Could concepts of rights, human dignity, and participation be confined to national units and be kept from contaminating imperial ones?55 The Haitian Revolution in the French Empire, the combination of slave revolts and antislavery mobilization in the British Empire, and the tensions between creole elites and peasants and slaves in the era of revolution in Spanish America all point to the possibility that politics in metropoles could not be neatly segregated from colonies. Somebody might take imperial space seriously. When British or French rulers wanted Africans or Asians to be soldiers as well as workers and cash crop producers, they appealed to a notion that subjects had membership and a stake in an imperial polity. I argue in chapter 7 that the French government after 1946, faced with challenges to the legitimacy and security of its colonies, explicitly effaced the colonial nature of the regime in favor of an imperial vision of Greater France as a differentiated unit of belonging, in which all people were now considered rights-bearing citizens, but with a range of political relationships to the state. Such initiatives from above provided openings to more demands for equivalence—economic and social as well as political—from below.

The empire perspective allows us to appreciate not only the significance of the racialization of difference within nineteenth-century imperial polities, but the instability of that racialization. It gives us more options with which to understand the variety of political forms in the past and present than those of colony, nation-state, and amorphous globality.56

The naming of empire has acquired in the first years of the twenty-first century a political salience it seemed to have lost in the last half of the twentieth, and once again the politics of naming need to be understood. One contemporary use of empire is as a metaphor for the extremes of state power. The Bush administration in the United States has been denounced from the Left for behaving like an empire, and encouraged from the Right to act like an empire in order to bring more order to the world.57 Whether either argument makes effective polemical use of the word is not for a historian to pronounce upon. But one can point to the risks either usage entails for understanding political processes: if every form of asymmetrical power is termed empire, we are left without ways of distinguishing among the actual options we might have. Liberals may be sliding into a denunciation of power that fails to distinguish different motivations and mechanisms for deploying it. Conservatives who evoke the empire analogy seem little interested in an essential dimension of historic empires: the long-term incorporation of territory and people into a polity. Iraqis and Afghanis are not about to become American subjects. Even proponents of the empire analogy doubt that the United States has the gumption to undertake imperial responsibilities—but such responsibilities are not actually at stake.58 The empire word is being used to delegitimize the sovereignty of particular regimes, to mark "rogue" states, to separate the world into the fit and the unfit, the modern and the backward. Conservative empire talk is about domination, not incorporation, and most fundamentally of all, its political purpose is to mark the excluded.59

Empires have a large place in history, but the exercise of power across territorial lines also took other forms and can be described in other words: hegemony as used by international relations theorists, the gunboat diplomacy that was part of American foreign policy, or the "imperialism of free trade" of nineteenth-century Britain.60 We need to consider both the entire range of forms of power and the consequences each one entails. Some scholars argue that the adjective imperial, applied to power, should be separated from the noun empire to underscore the diverse methods by which power is sometimes exercised on a vast scale: by Great Britain in the early nineteenth century or the United States in the twenty-first. One can accept this argument without losing sight of the specificity of actual empires. If we don't pay attention to what empires did—the marking and policing of boundaries, the design of systems of punishment and discipline, the attempt to instill awe as well as a sense of belonging in diverse populations—we will not understand any better the other ways in which powerful states act, and their limitations. Nor, if we wish to study power from "below" (or from in between), can we afford to miss the importance of making claims for resources, rights, or access on an empire on the basis of belonging—a claim that rulers of empire in certain circumstances needed to take seriously. In short, the need to understand the range of forms of imperial power entails appreciating both the general condition and its specific forms, including empire and colonies. Such analysis should be a dynamic one: states could be dragged into colonization when other means of exercising imperial power failed, and they could decolonize without giving up indirect means of authority.61 Thinking carefully about such distinctions in historical terms underscores the misleading nature of discussions about "empire" today.

One should neither avoid the specific trajectories of Western European expansion nor fetishize them. To enlarge empire to include non-Western or ancient empires is not to dilute responsibility for what European empire entailed, but on the contrary to enable a more specific discussion of choice, responsibility, and consequences. To take the story of European colonization out of the metanarratives of globalization, the triumph of the nation-state, colonial modernity, or post-Enlightenment reason is, in fact, to provincialize Europe.

Chakrabarty and others are quite right to point out that historical asymmetry is reproduced in the practice of historians: scholars who examine Asia or Africa refer continuously to European models and European styles of history writing, whereas those who study Europe are free to ignore or compartmentalize the experience of Asians and Africans and need not refer to modes of apprehending the past outside their own imaginations.62 On an imperial scale, Zulus or Bengalis, whatever their political strategies were to be, had a much greater need to learn the English language and frame their projects in relation to European models than Europeans had to learn Zulu or Bengali or envisage the modes of understanding that Zulu or Bengali brought to their histories.63 But scholars have enormous difficulty in separating the asymmetry of power from a totality. They can show that such successful challenges to power as antislavery, anticolonial, and anti-apartheid movements did not fully overthrow the inequalities they challenged or escape the frameworks of social order that imperial expansion produced. Scholars are less willing to acknowledge to what extent asymmetrical power is assailable power, or that the terrain labeled "Europe" might in fact change even as other people seem to be conducting their battles for recognition on "European" terms. Chakrabarty, in the end, contributes to the asymmetry he rightly deplores by focusing his attention on what he calls a "hyperreal" Europe instead of taking on a more historical, more provincial Europe.64

There is no ready formula for analyzing power structures that are neither symmetrical nor dichotomous. The work that has gone under the name of colonial studies and postcolonial theory is both vital and insufficient to such a task, vital because of the fundamental role of imperialism and colonialism in shaping the geography of power, insufficient because discussion at the general level of the colonial does not tell us enough about the ways in which conflict and interaction have reconfigured imaginative and political possibilities. As we address ways in which people of different origins within states or in international fora can interact, our task becomes much more difficult as we recognize that the issue is not difference per se, but rather a history that has placed differences in fundamentally unequal relationship.65 But such relationships are not static either. We are not faced with a dichotomous choice between a universality that is really European and an alternative that can be located within an irreducible "community," and rather than resolve the tensions in favor of one such pole, we are best off using those tensions to think through issues and conflicts in their painful concreteness.

Scholarship in the 1980s and afterward has rewritten French, British, Spanish, and American histories to show that Europe was reshaped in the colonies even as people in Asia, Africa, the Pacific, and the Americas were confronting the categories of colonizers. This work has invested these histories with a moral fervor as well as an expanded horizon of inquiry. We should not lose that fervor, even while taking inspiration from it to explore the historical trajectories of colonial situations. We can examine the constraints imposed by the insinuation of Western social categories into daily life and political ideology in conquered spaces without assuming that the logic imminent in those categories determined future politics. We can recognize the instability and contested nature of colonizing ideologies and ask how political leaders in the colonies sought to reinterpret, appropriate, deflect, and resist the political ideas they gleaned from colonial rulers, their own experiences, and their connections across colonial boundaries.

We do not need to romanticize anticolonial movements in their moment of triumph or treat colonial history as if the actions of the colonized never changed its course up to the final crisis; colonialism was as much threatened by fissures within its modes of action and representation as by the threat that the last might become first.66 We can probe the continued traces today of colonial histories while still acknowledging that these histories are not reducible to a colonial effect. Far from having to choose between examining the complexities of a colonial past and broadening our sense of the opportunities and constraints of the future, a critical and sensitive historical practice can help us retain our focus on the possibilities of political imagination and the importance of accountability for the consequences of our actions.



1. Laurent Dubois, "La République métissée: Citizenship, Colonialism, and the Borders of French History," Cultural Studies 14 (2000): 22.

2. See the considerations of the insights and blind spots in different turns in Victoria E. Bonnell and Lynn Hunt, eds., Beyond the Cultural Turn: New Directions in the Study of Society and Culture (Berkeley: University of California Press, 1999), especially the editors' thoughtful introduction, and William H. Sewell, Jr., "Whatever Happened to the 'Social' in Social History?" in Joan W. Scott and Debra Keates, eds., Schools of Thought: Twenty-Five Years of Interpretive Social Science (Princeton, N.J.: Princeton University Press, 2001), 209-26. For more turns, see Terrence J. McDonald, ed., the Historic Turn in the Human Sciences (Ann Arbor : University of Michigan Press, 1996); Fredric Jameson, The Cultural Turn: Selected Writings on the Postmodern, 1983-1998 (London: Verso, 1998).

3. Antoinette Burton writes of an "imperial turn" in historical scholarship, contrasted to what she considers a "hidebound" insistence on Britain's "national" character. After the Imperial Turn: Thinking With and Through the Nation (Durham NC: Duke University Press, 2003), 9. I agree with much of her argument (see chapter 6), but the contrast of the hidebound and the turners does not seem to me the most effective way to make it.

4. Stephen Howe describes the mutually dismissive stances of people who call themselves postcolonial theorists and imperial historians. "The Slow Death and Strange Rebirths of Imperial History," Journal of Imperial and Commonwealth History 29 (2001): 131-41.

5. For the critique of modernity, one can begin with Dipesh Chakrabarty, Provincializing Europe: Postcolonial Thought and Historical Difference (Princeton, N.J.: Princeton University Press, 2000), while the supposed clash between those who have achieved modernity (the West) and those who are incapable of doing so (Muslims, most notably) is the theme of Samuel Huntington, The Clash of Civilizations and the Remaking of World Order (New York: Simon and Schuster, 1996).

6. See Eric Wolf, Europe and the People without History (Berkeley: University of California Press, 1982).

7. Data are from the ArticleFirst database of OCLC. Data on books from the Stanford University catalogue are similar: modernity, almost never used in book titles around 1980, passed modernization in 1991.

8. A.G. Hopkins and Niall Ferguson differ considerably in how they interpret the economic history of the British empire, but both seek to ratify their positions by associating them with globalization—a position more teleological than historical. Hopkins, ed., Globalization in World History (London: Pimlico, 2002); and Ferguson, Empire: The Rise and Demise of the British World Order and the Lessons for Global Power (London: Allen Lane, 2002).

9. Partha Chatterjee, Our Modernity (Amsterdam, Dakar: Sephis/Codesria, 1997), 20. Available on the web at

10. Others have attempted to define and examine differing directions within postcolonial theory, most recently Suvir Kaul and Ania Loomba, eds., Postcolonial Studies and Beyond (Durham, NC: Duke University Press, 2005). Subaltern Studies has been the focus of an analysis that has been both appreciative of its accomplishments and critical of some of its suppositions or tendencies. See the forum in American Historical Review 99 (1994): 1475-1545; and David Ludden, ed., Reading Subaltern Studies: Critical History, Contested Meaning and the Globalization of South Asia (London: Anthem Press, 2002).

11. Ashis Nandy, "History's Forgotten Doubles," History and Theory 34 (1995): 44-66.

12. Nicholas Dirks, "History as a Sign of the Modern," Public Culture 2, no. 2 (1990): 25. For a thoughtful discussion of critiques of history and their relation to the state, see Mamadou Diouf, "Des historiens et des histoires, pour quoi faire? L'histoire africaine entre l'état et les communautés," Revue Canadienne des Etudes Africaines 34 (2000): 337-74.

13. Edward Said, Orientalism (New York: Pantheon, 1978) and Culture and Imperialism (New York: Knopf, 1993). For a recent reassessment by historians of the influence of Said, see the contributions of Andrew Rotter, K.E. Fleming, and Kathleen Biddick to a forum on "Orientalism Twenty Years On," American Historical Review 105 (2000): 1204-49.

14. Such approaches have been strongly influenced by Michel Foucault, whom Said both used and criticized. See chapter 2. Africanists have been challenged by Valentin Mudimbe to examine how the "colonial library" shapes the ways in which the concept of Africa has been constructed, while Gaurav Desai asks us to "reimagine the colonial library as a space of contestation." V.Y. Mudimbe, The Invention of Africa: Gnosis, Philosophy, and the Order of Knowledge (Bloomington: Indiana University Press, 1988); Gaurav Desai, Subject to Colonialism: African Self-Fashioning and the Colonial Library (Durham, N.C.: Duke University Press, 2001), 4.

15. David Scott, Refashioning Futures: Criticism after Postcoloniality (Princeton, N.J.: Princeton University Press, 1999), 156, emphasis in original.

16. For sympathetic but critical views of the political implications of postcolonial studies, see Gayatri Chakravorty Spivak, A Critique of Postcolonial Reason: Toward a History of the Vanishing Present (Cambridge, Mass.: Harvard University Press, 1999); and Nicholas B. Dirks, "Postcolonialism and Its Discontents: History, Anthropology and Postcolonial Critique," in Scott and Keates, Schools of Thought, 244, 246; for hostile ones, see Sumit Sarkar, "The Fascism of the Sangh Parivar," Economic and Political Weekly, Jan. 20, 1993, 164-65; Sumit Sarkar, "The Decline of the Subaltern in Subaltern Studies," in Writing Social History (Delhi: Oxford University Press, 1997), 81-108; Arif Dirlik, "The Postcolonial Aura: Third World Criticism in the Age of Global Capitalism," Critical Inquiry 20 (1994): 328-56.

17. The interest of Chakrabarty (Provincializing Europe) in the "post-enlightenment" does not lead him to mention the post-enlightenment of the Haitian revolutionaries; Scott doesn't mention Haiti either (Refashioning Futures). Robert J. C. Young, in Postcolonialism: An Historical Introduction (Oxford: Blackwell, 2001), cites Haiti as a slave revolt, not as a revolt that shaped debates on emancipation. It was C.L.R. James who appreciated the significance of this event. The Black Jacobins: Toussaint L'Ouverture and the San Domingo Revolution, 2d ed. (New York: Vintage, 1963 [1938]).

18. Stuart Hall, "When Was 'the Post-colonial'? Thinking at the Limit," in Iain Chambers and Lidia Curti, eds., The Post-Colonial Question: Common Skies, Divided Horizons (London: Routledge, 1996), 249. More curious is Young's statement in a book about postcolonialism: "The postcolonial does not privilege the colonial. It is concerned with colonial history only to the extent that history has determined the configurations and power structures of the present, to the extent that much of the world still lives in the violent disruptions of its wake, and to the extent that the anti-colonial liberation movements remain the source and inspiration of its politics." His postcolonialism assumes "a common political and moral consensus towards the history and legacy of western colonialism." How one would be able to judge "the extent" without studying the history is not obvious, but what counts for Young is a stance, taken out of the context of whatever that stance is directed against. Postcolonialism, 4-5.

19. Achille Mbembe's insightful and provocative On the Postcolony (Berkeley: University of California Press, 2001) also tends toward a generalized conception of both the postcolony and its colonial antecedent.

20. Gyan Prakash, "Subaltern Studies as Postcolonial Criticism," American Historical Review, 99 (1994): 1475-90.

21. Walter Mignolo sees "coloniality" as part of a singular Western history dating to the sixteenth century: "Coloniality, in other words, is the hidden face of modernity and its very condition of possibility." "The Many Faces of Cosmo-polis: Border Thinking and Critical Cosmopolitanism," Public Culture 12 (2000) 722. For other examples of a colonial phenomenon abstracted from time and place, see Homi Bhabha, "Of Mimicry and Man: The Ambivalence of Colonial Discourse," in Frederick Cooper and Ann Laura Stoler, Tensions of Empire: Colonial Cultures in a Bourgeois World (Berkeley: University of California Press, 1997),152-60; Walter Mignolo, Local Histories/Global Designs: Coloniality, Subaltern Knowledges, and Border Thinking (Princeton, N.J.: Princeton University Press, 2000).

22. Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism (Princeton, N.J.: Princeton University Press, 1996).

23. For another example of the leapfrogging fallacy, see Richard Price, The Convict and the Colonel: A Story of Colonialism and Resistance in the Caribbean (Boston: Beacon, 1998). This book is constructed around the ironic juxtaposition of a violent confrontation of police and demonstrators in the French island of Martinique in 1925 and the trivializing of the memory of colonization in the period of Price's fieldwork, when the people of Martinique were caught up in the tourist business and the French welfare state. By omitting the history in between, Price obscures the seriousness of political mobilization in the 1930s and 1940s, when a strong Caribbean movement pressed the French government to accord this colony the status of a French department, and thereby worked to lay claim to French educational and social resources equivalent to those of other French citizens. One would not know from Price's account that this movement succeeded in 1946—that the victims of the 1925 conflict did not die in vain—or that the noted writer/activist whose authority Price invokes in indicting French colonialism, Aimé Césaire, was the main leader of the departmentalization movement. The missing middle contains the politics.

24. The recent flood of books on the social construction of whiteness is an example of both the value and the problems of this approach. See Eric Arnesen, "Whiteness and the Historians' Imagination," and the debate it provoked in International Labor and Working Class History 60 (2001): 3-92.

25. I have discussed the relationship between nationalist politics and trade union movements in "Dialectics of Decolonization: Nationalism and Labor Movements in Postwar French Africa," in Cooper and Stoler, Tensions of Empire, 406-35. Michel-Rolph Trouillot makes a related point about constructivism: we can point to narratives being produced, but "the reasons why a specific story matters to a specific population are themselves historical." Silencing the Past: Power and the Production of History (Boston: Beacon, 1995), 13.

26. Karl Mannheim, Ideology and Utopia: An Introduction to the Sociology of Knowledge, trans. Louis Wirth and Edward Shils (San Diego: Harcourt Brace Jovanovich, 1985 [1936]).

27. See chapter 5 for more on the confusions of the term modern, in its epochal and other variants. See also Bernard Yack, The Fetishism of Modernities: Epochal Self-Consciousness in Contemporary Social and Political Thought (Notre Dame, IN: University of Notre Dame Press, 1997).

28. Hall, "When Was 'the Post-colonial'?" 246, also argues for a break point with decolonization. I prefer to see this as a change in the political repertoire rather than to make generalizing claims about each epoch, for example between a Manichean coloniality and a hybrid postcoloniality.

29. When Simon Gikandi states that "postcolonial theory is one way of recognizing how decolonized situations are marked by the trace of the imperial pasts they try to disavow," he is opening the door to a valuable analysis of where those traces lie and what course they take over time. When he later states, "My assumption here is that the 'founders' of the new postcolonial nations legitimated their authority by claiming the agency of pure modernity (even as they sang praise songs for precolonial traditions) not because they were attracted to the imperial project per se but because they had no real access to modes of knowledge outside the horizon of expectations established by empire," he is preempting such an inquiry. Maps of Englishness: Writing Identity in the Culture of Colonialism (New York: Columbia University Press, 1996), 15, 18.

30. For a thoughtful discussion of the confusions of social type and historical era in the evocation of the modern, see James Ferguson, Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copperbelt (Berkeley: University of California Press, 1999), 42-43.

31. These remarks should not detract from the illuminating work that some scholars have done using Foucauldian concepts. See especially the valuable reflections of Ann Laura Stoler, Race and the Education of Desire: Foucault's History of Sexuality and the Colonial Order of Things (Durham, N.C.: Duke University Press, 1995).

32. Chakrabarty, Provincializing Europe.

33. Chakrabarty, Provincializing, 237. Hegel serves the same role for Gyan Prakash, Another Reason: Science and the Imagination of Modern India (Princeton, N.J.: Princeton University Press, 1999), 8, 118; and Nicholas Dirks, Castes of Mind: Colonialism and the Making of Modern India (Princeton, N.J.: Princeton University Press, 2001), 52.

34. Arno Mayer, The Persistence of the Old Regime: Europe to the Great War (New York: Pantheon Books, 1981).

35. David Hollinger, "The Enlightenment and the Genealogy of Cultural Conflict in the United States," and Dena Goodman, "Difference: An Enlightenment Concept," in Keith Michael Baker and Peter Hanns Reill, eds., What's Left of Enlightenment? A Postmodern Question (Stanford, Calif.: Stanford University Press, 2001), 7-18 (11 quoted) and 129-47.

36. Sankar Muthu, Enlightenment against Empire (Princeton, N.J.: Princeton University Press, 2003); Alyssa Goldstein Sepinwall, "Eliminating Race, Eliminating Difference: Blacks, Jews, and the Abbé Grégoire," in Sue Peabody and Tyler Stovall, eds., The Color of Liberty: Histories of Race in France (Durham, N.C.: Duke University Press, 2003), 28-41.

37. On the role of slaves in redefining the meanings of freedom, see Robin Blackburn, The Overthrow of Colonial Slavery, 1776-1848 (London: Verso, 1988); George Fredrickson, Black Liberation: A Comparative History of Black Ideologies in the United States and South Africa (New York: Oxford University Press, 1995); Thomas Holt, The Problem of Freedom: Race, Labor, and Politics in Jamaica and Britain, 1832-1938 (Baltimore, Md.: Johns Hopkins University Press, 1992).

38. William Sewell, Jr., "Historical Events as Transformations of Structures: Inventing Revolution at the Bastille," Theory and Society 25 (1996): 841-81, 843 cited. I use the concept of lumpiness to make a point about space similar to the one Sewell makes about time (chapter 4).

39. One of the main achievements of African history has been to bring out not just the variety of historical trajectories found in Africa before and after colonization, but the variety of ways in which Africans have thought about them. The difficult task is to see how people worked between different modes of understanding time and process, as they came to grips with the perspectives with which colonial schools and missions confronted them. People in the middle experienced neither a smooth blending or hybridity nor a radical incommensurability of different historicities. For examples of the use of texts by people in the middle of colonial relationships in order to reconceptualize history, see Nancy Rose Hunt, A Colonial Lexicon: Of Birth Ritual, Medicalization, and Mobility in the Congo (Durham, N.C.: Duke University Press, 1999); Desai, Subject to Colonialism; and Meredith McKittrick, To Dwell Secure: Generation, Christianity, and Colonialism in Ovamboland (Portsmouth, NH: Heinemann, 2002).

40. The centrality of nationalism to the history of the last two hundred years is emphasized in Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1983); while the demise of the nation-state is announced in Arjun Appadurai, Modernity at Large: Cultural Dimensions of Globalization (Minneapolis: University of Minnesota Press, 1996), 19.

41. James, Black Jacobins; Laurent Dubois, Avengers of the New World: The Story of the Haitian Revolution (Cambridge, Mass.: Harvard University Press, 2004); idem., Les esclaves de la République: L'histoire oubliée de la première émancipation 1789-1794 (Paris: Calmann-Lévy, 1998).

42. Eve M. Troutt Powell, A Different Shade of Colonialism: Egypt, Great Britain, and the Mastery of the Sudan (Berkeley: University of California Press, 2003).

43. The possibilities of putting Central Asia in an imperial perspective are illustrated in Adeeb Khalid, The Politics of Muslim Cultural Reform: Jadidism in Central Asia (Berkeley : University of California Press, 1998).

44. Partha Chatterjee, The Nation and Its Fragments: Colonial and Postcolonial Histories (Princeton, N.J.: Princeton University Press, 1993), 16.

45. Edmund S. Morgan's pioneering study of colonial Virginia brought out the increasing racial polarization that developed as elite Virginians sought to forge a republican coalition, whose class inclusiveness demanded racial exclusiveness. American Slavery, American Freedom: The Ordeal of Colonial Virginia (New York: Norton, 1975). For a contrast, the tsarist empire of Russia did not develop such dichotomies, for all Russians belonged to defined social categories and all were subjects of the tsar. See Jane Burbank and Mark von Hagen, eds., Geographies of Empire (Bloomington: Indiana University Press, forthcoming).

46. David Scott adds to his heroic view of anticolonialism and to subsequent postcolonial disillusionment a third stage, "after postcolonialism," whose hallmark is the revelation that everything that went before was the violence of modernity. Refashioning Futures, 10-14, 16-17, 45, 199. See chapter 7 for a different way of discussing openings and closures in the process of decolonization.

47. Some scholars see both possibility and danger in the fact that postcolonial theory "seems to locate itself everywhere and nowhere." Ato Quayson and David Theo Goldberg, "Introduction: Scale and Sensibility," in David Theo Goldberg and Ato Quayson, eds., Relocating Postcolonialism (Oxford: Blackwell, 2002), xvi.

48. The spatial dimension of colonization is both underscored and complicated by some scholars' use of the notion of internal colonialism. It is sometimes used literally—to denote the consignment of certain categories of people into specific spaces within self-governing territories, such as Indian reservations in the United States or African homelands in South Africa after its independence from Britain in 1910. Sometimes it is used metaphorically to denote extremes of categorization and discrimination of people in a polity that is formally national. The metaphoric usage requires prudence, for it risks diluting the actual experience of people in actual colonies as well as misanalyzing the specificities of categorization and differentiation in national polities.

49. Australia and the western United States in the nineteenth century provide examples of the assimilation of individuals and the destruction of cultures, of extermination, and of extreme marginalization." Scholars of Spanish America have debated since the nineteenth century the extent to which mestisaje is myth or reality; some China specialists see China as an empire with a strong homogenizing tendency—a movement toward becoming a very large nation-state—but others put more emphasis not just on the importance of frontier areas like Tibet or Inner Asia, but on the ethnic distinctiveness of the ruling dynasty in the long Qing period, from the seventeenth to the twentieth century. See chapter 6.

50. Michael Mann, The Sources of Social Power: Volume II. The Rise of Classes and Nation-States, 1760-1914 (Cambridge: Cambridge University Press, 1993).

51. This is the problem with Michael Doyle's definition of empire as control of "a subordinated society by an imperial society": societies are not necessarily what does the subordinating; they may be produced in the building and evolution of empires. Empires (Ithaca, N.Y.: Cornell University Press, 1986), 30.

52. Robert L. Delavignette, "Action colonisatrice et paysannat indigène," Afrique Francaise 45 (1935): 526-30.

53. Ann Laura Stoler and Frederick Cooper, "Between Colony and Metropole: Toward a Research Agenda," in Cooper and Stoler, Tensions of Empire, 1-57.

54. Selim Deringil, "'They Live in a State of Nomadism and Savagery': The Late Ottoman Empire and the Post-Colonial Debate," Comparative Studies in Society and History 45 (2003): 311-42; Jane Burbank, "The Rights of Difference: Law and Citizenship in the Russian Empire," paper for conference "Beyond Europe," School of American Research, Santa Fe, New Mexico, Oct. 25-30, 2003. The civilizers, in these cases, were imposing an imperial rather than a national vision of directed change: Ottoman imperialism, not Turkish in the first instance, an imposition made on Russian peasants as much as on Central Asians in the second.

55. Dubois, Avengers of the New World, 3.

56. Globality is particularly amorphous—and its relationship to nation-state and colonial empires particularly murky—in Michael Hardt and Antonio Negri, Empire (Cambridge, Mass.: Harvard University Press, 2000). For a critique, see Frederick Cooper, "Empire Multiplied," Comparative Studies in Society and History 46 (2004): 247-72.

57. On the left, see Ellen Meiskins Wood, Empire of Capital (London: Verso, 2003); David Harvey, The New Imperialism (Oxford: Oxford University Press, 2003); and Michael Mann, Incoherent Empire (New York: Verso, 2003); and on the right, Ferguson, Empire.

58. Niall Ferguson, "The Empire Slinks Back: Why Americans Don't Really Have What It Takes to Rule the World," New York Times Magazine, April 27, 2003, pp. 52-57.

59. Revivifying empire and enforcing exclusion are at the center of Niall Ferguson's thinking. He compares the supposed dangers of Muslims immigrating to Europe to the decline and fall of the Roman empire, citing Gibbon and evoking the dangers of "a youthful Muslim society . . . poised to colonize—the term is not too strong—a senescent Europe." "Eurabia?" New York Times Magazine, April 4, 2004, pp. 13-14.

60. P.J. Cain and Anthony Hopkins make a good argument in the British case for examining a range of imperial strategies of which empire as a political and administrative concept is a subset, and not always the most important one. They see occasional coercion, diplomatic initiatives, and the establishment of close networks—of "kith and kin" linking Britain with colonies of settlement, of banks building ties of credit and information exchange, of merchants developing interfaces with producers in different areas—as part of an imperial project. One can argue about the strength of these networks and about the prominence of the "gentlemanly capitalists" they see at their center, but the broad view of British naval, political, economic, social, and cultural power is a useful one. It does, however, require specification of the consequences of particular strategies, including the establishment of colonial institutions over much of Asia and Africa and the selective transformation of colonial institutions into other political forms in places like Canada and Australia. British Imperialism, 1688-2000, 2d ed. (London: Longman, 2002).

61. International relations specialists distinguish the "hegemonic" (setting the rules of the game which others must follow) from the "imperial" (intervening in another polity without actually governing it) and the "colonial" (governing the internal affairs of a subordinated polity). The importance of such distinctions—but also the possibility of change from one category to another—are among the issues raised in Craig Calhoun, Frederick Cooper, and Kevin Moore, eds., Lessons of Empire (New York: New Press, forthcoming).

62. Dipesh Chakrabarty, "Postcoloniality and the Artifice of History: Who Speaks for 'Indian' Pasts?" Representations 37 (1992): 2-3.

63. But not entirely, for colonial regimes trying to govern specific territories could be caught up in the historical discourses of their indigenous interlocutors. See Carolyn Hamilton, Terrific Majesty: The Powers of Shaka Zulu and the Limits of Historical Invention (Cambridge, Mass.: Harvard University Press, 1998).

64. Chakrabarty, "Postcoloniality," 17.

65. Seyla Benhabib, seeking to understand how people can address difficult issues across lines of cultural difference, rejects "radical incommensurability" and argues for "a pluralistically enlightened ethical universalism." There is much in her argument, but it is too abstract, for the plural units she hopes to see brought into a "conversation" with each other are left generic and equivalent. That is why—and here I agree with postcolonial theorists—the history of colonialism is so important to such debates, for it brings out the profound asymmetry in global interaction. The Claims of Culture: Equality and Diversity in the Global Era (Princeton, N.J.: Princeton University Press, 2002), 35-39.

66. To cite the famous phrase of Frantz Fanon, The Wretched of the Earth, trans. Constance Farrington (New York: Grove Press, 1963), 30.