Where is the field of African history going these days? Every which way, it would seem. And that is rather a good thing. Younger scholars are delving into a wide variety of topics—from studies of early African societies based on linguistic reconstruction to studies of politics and culture in independent Africa in the 1970s. A new wave of studies of the slave trade, focused on the interaction of trading networks on sea and on the African continent, has come ashore, and economic history, for some time in the doldrums, is being revived by scholars vigorously debating questions about change in the longue-durée. Religion has for some time been of interest, enriched more recently by several studies of the ways in which mission converts used literacy to preserve their own historical memories, to develop cultural syntheses, and make claims on missionaries and colonial governments.

It wasn't always this way. African history has been as subject to fads as any other field of history. The field came into being in the American and British academies—even more so in the new universities of Africa—at a time when nationalist movements appeared to have triumphed in most of Africa, and historians saw themselves writing a history that would be useful for nation-building. Two subjects were privileged: precolonial history—providing a precedent for a new period of self-rule in Africa—and the history of resistance to European conquest. In the late 1960s and early 1970s, colonial history or the history of South Africa attracted less interest; they were too "white." By the 1990s, or even earlier, that tendency had been almost completely reversed: very few doctoral students wanted to study precolonial history. Colonialism was in. Once it became clear that nation-building projects were not providing an ideological basis for a new Africa, scholars became more interested in the constraints: on the institutional and ideological constructs that colonial rule imposed on Africans and on the particular ways in which Africa's incorporation into the world economy subordinated it to outside forces, before, during, and after colonial rule. That fad passed too, and younger Africanists, and some older ones too, seemed mostly to want to study cultural history. Labor history had been a hot field in the 1980s, two decades later it is virtually gone, although some scholars in South Africa and the Re:Work program in Berlin are trying to revive it.

The push to conformacy is a common problem in scholarly circles. Scholars have celebrated something of which they should be skeptical: "turns"—be it the "cultural turn," the "imperial turn," or any other. Why should we all turn at the same time and treat those who fail to do so as if they were driving down a dead end? Do we need to turn to cultural history to correct a bias toward social history and then back again? History, nonetheless, has been less stultified by conformist pressures than political science or economics, helped by our theoretical eclecticism and the lack of pressure to prove that we are really scientists. That history departments usually divide their turf by time and place has allowed for both a degree of flexibility and a certain tendency to let each specialty do its thing and not really to foster conversations among them. That weakness actually helped African history get going in the 1960s: hiring an African historian implied an addition, not a rethinking of the nature of the discipline.

The prevalence of conformism in American academia is striking to an Africa specialist, because it is so obvious how privileged we are, at least those of us who teach at research universities. Our colleagues teaching at universities in Africa have not had things so easy–both because of financial constraints, especially after "structural adjustment" was imposed on most African countries by international financial organizations in the 1980s and 1990s, and because oppressive governments have cracked down on university autonomy. Many historians have struggled nobly in the face of such constraints, but they have not had the luxury of following different trends. Even people teaching in North America and Europe have been adversely affected by Africa's academic crisis, for the intellectual stimulus which we got from the university seminars, conferences, and journals in Africa during the 1970s was quite fragile by the 1990s, and with it was lost some of the excitement of studying and talking about subjects where they were of immediate relevance. The big exceptions have been CODESRIA and several of the universities in postapartheid South Africa; meanwhile American universities have benefitted from the increasing presence of African scholars on their faculties—many of them concerned with doing as much as possible to reconstitute connections between American and African institutions. Research funding in Africa is scarce, and in the United States it is less than secure: we need sources, like the SSRC's international dissertation research program or Fulbright grants that let students make a case for the importance of their topic, rather than impose their elders' agendas on research possibilities.

One way or another, the field of African history has for some time been concerned with Africa's place in the world—as part of a long-term dialectic of subordination and liberation. At times, this engagement has taken an inward turn: to emphasize the distinctiveness of an "African perspective," to deploy oral history as a methodological alternative to the domination of written European languages, to zero in on the particularity of each African kingdom or community, to insist, stridently at times, that Africa really does have a history. At times it has looked outward, and one can hark back to 19th-century African and African American writers or, more recently, to W. E. B. Dubois's The World and Africa (1946), which insisted both on Africa's historical integrity and the need to analyze the effect of the slave trade and colonization on it.

To consider Africa in relation to "globalization" or "global history" is only the latest such manifestation, one which both suggests valuable lines of connection to other fields of history and raises the usual concerns about hopping on bandwagons. The globalization fad began with claims, post-1989, that the world was opening to flows of capital, culture, and people across continental and national barriers—it often confused an assertion of fact with a normative claim. But in a literal sense, very little that happens or has happened is truly global—climate change perhaps. Much happens over big spaces and long distances—but not everywhere. By adopting the globalization conceit, we risk encouraging our students to ask poor questions, about limitless, boundaryless flows instead of about mechanisms of connection and their limits. But African historians have been among those asking good questions about long-distance connections and large-scale processes. The work of slave trade scholars is a case in point—a study of networks coming out of the Mali Empire in the 15th century intersecting with networks coming out of Iberia. One can make similar arguments about trans-Saharan networks over several centuries. These networks were long but finite, and those who controlled them often tried to keep them narrow. Colonization forcefully integrated Africa into certain kinds of large-scale spatial structures, but suppressed other networks and imposed different kinds of boundaries.

Ideas about liberation circulated widely—from Lagos to London and Moscow to Delhi—but a dedicated pan-Africanist in the 1930s might have had closer linkages to people 2000 miles away than to somebody 20 miles from the port city in which he lived. The limitations of the flow of ideas and political connections should not be lost in a celebration of their "global" reach. With the decline of colonial rule, Africans liberated themselves from one set of constraining structures but ran into others—elites with a national power base, restrictions by European or American states on immigration, pressure to conform to certain political, social, and cultural norms, international power politics. One could go on: a dialectic of territorializing and deterritorializing processes provides a better conceptual apparatus to understand African—or other histories—than the notion of globalization. As for "global history," thinking in such terms may have the beneficial effect of giving history departments a category in which to hire people outside of continental or national categories, but it still leaves us with the problem of finding the level of analysis which is appropriate for the problem we are trying to solve–and that is rarely the planet as a whole.

The study of empires, with which I have been very much engaged for the past 25 years, takes as its unit of analysis something that is big, expansionist, perhaps with pretensions to global domination. But running a durable empire also meant understanding the limits of one's power and working specific channels of authority. Once again, the concept can be useful but also misleading. Some scholars have posited a generic "coloniality" that extends from 1492 to the late 20th century or an equally generic "colonial modernity" that covers the 19th and 20th centuries. Such notions easily become reductive, not least in obscuring the fact that even in the most repressive colonial empires, people pushed back. We need to study geographies of power, not assume them. African historians have, on the whole, been attentive to such processes. Not the least of the benefits of studying the mechanisms of imperial power and their limits—and the networks of commercial and political connections that crisscrossed imperial lines—is that it moves historians across and beyond the geographic fields on which the profession has been organized, calling on us to give as much attention to the particularities of other places as we Africanists wish our colleagues would give to the specificities of African history.

Frederick Cooper is professor of history at New York University. He is the author of several books, including the anthology of essays, Colonialism in Question: Theory, Knowledge, History.