Is There a Place for Anthropology in Social Science History?

Imagine a fourfold table in which one dimension is “present versus past” and the other “exotic versus home.” Traditionally, social and cultural anthropology’s domain has been the exotic’s present and history’s domain the home’s past. A third box, the home’s present, has been occupied by sociology, while the fourth, the exotic’s past, has usually been the province of anthropologists too because other disciplines—with the exception, perhaps, of ethnohistorians—are usually even less interested in exotic peoples’ past than in their present. These domains are now in flux. I argue, in what follows, that only when the oversimplified ideas about time and space that have created them are seriously questioned will anthropology find a secure “place” in social science history. ¹

At a time when social and cultural anthropologists felt that their obligation was to document the present lives of disappearing peoples, my husband, John W. Adams, and I made an unusual decision to embark on a historical study of the Yankees, a society that moved to the core of the world system during the period we were studying, 1650 to 1880; moreover, the Yankees were a society with writing, thus squarely within the domain of history. Other anthropologists had studied the present-day United States using participant observation, but few had studied its past. Thus, my scholarship is based squarely in the core, but my teaching is very often about the periphery. In most of my classes, I draw examples from classic “ethnographic cases.” This gap has been only partially filled by a new course titled “History’s Strangers: Ethnohistory as World History,” in which I teach about the histories of people who are usually marginalized in accounts centered on “the rise of the West.” Thus, I am continually roaming among at least three of the four domains. I am thankful to social science history for providing a home for such a vagabond as I.

It is in the tensions between a local that is not at all exotic—the American North from 1650 to 1880—and that huge unit encompassing all humans past and present and, for the sake of this essay, the future as well that my intellectual life resides. Unlike many in this association, therefore, I continue to have a stake in representing the entire world. When I came to social science history, in about 1981, I thought this task would be made easy through the use of science, which postulated a universal theory of what it meant to be human. Social science history would be a way of bridging the four intellectual domains and bringing them together under one universal theory. Today this view seems utterly naive. At the very least, it must be acknowledged that it is impossible to have strictly comparable information about each, since the methods needed to study them and the information available are so very different. But I am not ready to give up on the universal, and I do not think social science history can or should give up on the world. The very categories underlying the domains “exotic versus home” and “past versus present” must be questioned. I suppose I still partake of an anthropological vision that is all-encompassing and capable of seeing around those very categories that academics or members of today’s American culture see as so necessary and enduring.

Recently anthropology has been much concerned with redefining its place, now that colonialism is largely over and it can no longer so easily claim the exotic as its domain (Gupta and Ferguson 1997). So it is not surprising that I have been thinking about the future of social science history not only in terms of how the world is divided into “fields” or academic disciplines but also in terms of how those disciplines conceptualize their units of analysis, which inevitably includes ideas about place. The local community study that anthropologists carried out in living communities through participant observation has served as a model for what might be accomplished through written records for communities in the past. But now that model seems more and more obsolete, whether for the past or the present.

In what follows I will discuss the rather marginal “place” anthropology has had within social science history and discuss how it might become more fully incorporated in the future, but I will also raise questions about the concepts of place itself in social science history. These concepts have implicitly reflected ideas about society and the world that have too often been unexamined. For anthropology to become more central to social science history, new and rather different units of analysis must be explored. So after tracing a brief history of anthropologists’ participation in the Social Science History Association, I provide one example of how this might be done in my particular area of interest, historical demography: I argue that the field should move from the analysis of places to the analysis of flows. I introduce some of the ideas anthropologists are discussing for studying such flows and speculate about how they might relate to the uneasy domain of social science history. I conclude with a recipe that I hope will lead to a fuller incorporation of anthropologists into social science history.

Anthropology in the Social Science History Association

Social science history has its roots in modernism, a way of thinking in which anthropology’s subject was not the Other, as many see it today, but “natives.” Each culture was rooted in a locality, or, conversely, every locality had its natives, and it was the job of comparative theory to “explain” the differences and similarities between them. This view was taken into social science history in two ways: the very local community study, modeled after the ethnographic study, and the idea of the functional relationship, the notion that all aspects of every unit studied, even a nation, or a region composed of several nations, fit together organically into a complex whole, one where economy and demography played determining roles.

Anthropology would study these organic wholes in exotic cultures, mostly without writing and in the present. Historians would study the cultures with writing, especially those in Europe and North America, at the core of the world system. Yet anthropology’s “place” in the Social Science History Association has never been very secure. In contrast to the sociologists, who have become more and more numerous, the number of anthropologists has been small, though persistent—perhaps no more than 10 individuals. The association has most often attracted those anthropologists interested in the nuts and bolts of the family and demography, particularly those studying people closer to “home,” and those interested in quantification or in grand schemes, such as world systems theory. Yet there are other anthropologists and historians working at the intersection of anthropology and history who do not participate in the association. I will attempt to sketch this intellectual terrain that does not fit neatly into the fourfold box and describe some of the interdisciplinary efforts of anthropologists and of historians who borrow from them but who do not usually participate in our association. ²

When I got my degree, I was required to study a “non-Western culture,” though no one really knew anymore what that was. Some said the language had to be non-Indo-European; others allowed Jamaica. But “peasant societies” in Europe were considered exotic enough to be suitable subjects for doctoral dissertations, and many of the anthropologists who have been regularly active in the Social Science History Association made such studies through participant observation and then reconstructed the demographic history of “their” villages. It was through their interest in European “family history” that they came into the association (see, e.g., Brettell 1986; Halpern 1972; Kertzer 1989; Schneider and Schneider 1996).

Historical demography and family history is alive and well, but these small-scale anthropological projects of the past have been eclipsed by much larger, well-funded attempts to study historical demography in several countries, projects resulting in large, complex data banks that usually go back only to the beginnings of industrialization because of the lack of documents. This set of projects, though, is not very anthropological, in part due to the unit of analysis. In several of these projects, it is the entire country (as in the Netherlands, where a 1% sample of the population in the early nineteenth century is being drawn) or a large region (as in Sweden, where the Demographic Database at Umeå University has compiled information on a large region in the North, and Lund University has collected similar information on areas in the South).

My husband and I came into the association not from that exotic locale, Europe, but because of our desire to test theories and generalizations about how migration and the family changed over time. A physical anthropologist named Alan C. Swedlund (1990) came for similar reasons. It seemed to us useful for comparative purposes that this type of study be made at the core as well as in the peripheral societies that were anthropologists’ more usual domain. Furthermore, it was much more feasible given the written sources and might even lead to new theories, theories that could not be tested in the ahistorical domain of anthropology. But this avenue into the association has also dried up.

Several physical anthropologists are working with Richard H. Steckel on his large comparative project “A History of Health and Nutrition in the Western Hemisphere” (for a summary of this project, see the American Journal of Physical Anthropology, supplement 26, 1998). Responsible for the detection of pathology and health indicators in bones and teeth, they are paired with historians or historical demographers who have expertise related to the history of the populations whose skeletal remains are available for analysis. Another pocket of anthropologists in the association are those who are interested in world systems theory. The anthropologists’ contribution is to show that the processes creating the modern world system were at work before the development of modern capitalism historically and in prehistoric times, as well as in parts of the world far away from Europe and its direct influence (see, e.g., Chase-Dunn and Hall 1991). The creation of a new macrohistory network may portend a strengthening of interest in this direction that will draw more anthropologists in the future.

These are all examples of “grand schemes,” which at least for a while were social science history’s stock in trade. The students of peasant demography in Europe had an implicit interest in testing ideas about modernization. But there are many scholars working at the intersections of anthropology and history who are not at all interested in grand schemes. It is these scholars who do not usually participate in this association. Most of them see anthropology’s contribution as ethnography and find an easy harmony between the joint particularisms of history and anthropology. They are interested in the meaning ordinary people make of their lives, in holistic descriptions that are nevertheless grounded in particular material circumstances, usually in small localities. Many take their inspiration from the interpretive school of anthropology initiated by Clifford Geertz. What follows is a description of a set of scholarly networks; these groups are not really distinct in terms of their intellectual interests. ³

For example, one active group of anthropologists and historians works in the Marxist tradition, with ties to German-speaking historians; its members have rarely presented at SSHA meetings (Ludke 1995; Medick and Sabean 1984). Their undertakings quite easily cross the “home versus exotic” divide, and most of their publications are edited volumes that contain essays on both types of locations. They have published several books and meet periodically away from any association (see Cohen 1994: chap. 1 for a brief history of this group). But in contrast to our association’s, their meetings are, as far as I know, by invitation only.

The Melbourne School is another such group. Its members have written mainly about clashing or changing worldviews—for example, the history of encounters between colonists and colonized (see, e.g., Clendinnen 1987 and Dening 1996), though the scholar who is best known, Rhys Isaac, has written about eighteenth-century Virginia (1982). But his theme is modernization, in any case, as the old world of planters was disturbed by a new and more individualistic religious ideology. Charles Joyner (1984, 1999) also interacts frequently with this group. These scholars are quite interested in performance. They are a good example of how the interpretive approach can unify scholars studying the exotic and those studying home.

A third group, largely composed of anthropologists, seeks to study colonization and its aftermath in the anthropologists’ usual exotic domain. It might be dubbed the Chicago School, because its leading members, Marshall Sahlins (1981) and Jean and John Comaroff (1992), are at that university. Here again, the interest is away from grand schemes and toward “historicizing” the experiences of the peoples of Africa or the Pacific and understanding how they produced cultures from encounters with Europeans that were to some extent unique. The desire to bring these peoples into the present as equal actors on a stage often occupied only by “the West” has led these scholars to question the old functionalist paradigms in much the same way I will be doing in what follows. Despite the fact that our association meets in Chicago every third year, neither these scholars nor their students attend. I would place some of the scholars at the University of Michigan who work on the anthropology of colonialism in the same group (see, e.g., Cooper and Stoller 1997). The barrier to the attendance of these scholars interested in colonialism is probably our association’s steadfast interest in home. A fourth group that participates fitfully in the association is composed of ethnohistorians—those who work, for instance, on the history of Native American groups.

Yet the ideas anthropology has been concerned with continually enter the association and influence social science history as they are discovered and introduced by others in the association. The idea of the local community led to family reconstitution; the longue durée in the Annales School was a clear analogue to the anthropologists’ concept of culture. The relatively recent “culture” network, however, is largely the province of sociologists and is not often the site of interdisciplinary interaction. This network presents research on what is “left over” after normal sociology and economics do their work—topics such as masculinity, clothing, and material culture, all within the domain of home. This view of culture is different from the more holistic concept anthropologists have. I fear that, as in so many of these missed opportunities, the division between exotic and home is so strong that it has inhibited communication when both sides could learn a great deal from each other. Yet another such missed opportunity is in the study of colonialism: many are realizing that what happened in the colonies was closely linked to the bourgeois cultures being developed at home (Cooper and Stoller 1997), yet this topic has rarely been explored, to my knowledge, within the association.

Nevertheless, these borrowings occur at creative moments. Whether the present time of indecision and soul-searching in anthropology is such a moment is not at all clear. Since the end of the Cold War, the very largest units into which social science historians divide the world have changed. They now divide the world into large geographical entities such as “the Western Hemisphere” or “Eurasia” (the Eurasia Project [Bengtsson and Campbell 1998] compares demographic processes in local communities in Europe with those in Japan and China) and make comparisons only within these units. Also, scholars not from or trained in Europe or North America are now becoming active in the association for the first time, largely, I believe, as demographers, though their role is apparently to be experts on their home cultures. There are important and interesting assumptions behind these decisions, but I believe these assumptions remain largely implicit. A full-scale history of the kinds of units—spatial and cultural—that have motivated research in social science history since its beginning would be quite an interesting piece of intellectual history to explore but is beyond the scope of this essay. Still, I hope to have drawn attention to how research is affected by the way disciplines divide up the world. In social science history there is certainly a humbling visible in the new spatial units employed by new grand schemes that, unlike the previous “modernization” paradigm, are no longer centered on the place where capitalism began, the core of the world system.

It is time to question some of these preconceived ideas. And I would hope that in the future this association would become a meeting place for the groups I have mentioned who are exploring the terrain between anthropology and history but have not participated in the association in the past. Many of these groups do not provide forums for discussion that are as open to new participants as the Social Science History Association. But to assume this role, the association would have to move toward a less scientific image. This would be an obvious way to bring more anthropologists into the association.

Anthropology, of course, has always hovered uneasily between the sciences and the humanities. Historians, however, may look to social science for quantification or formal analysis; to them this may be what “science” offers that history does not. This kind of disciplinary borrowing is certainly useful, but it does not raise the sorts of questions about the very categories defining disciplinary domains that would lead to new forms of cooperation between the disciplines and other sorts of conversations between them. ⁴ Anthropology has a great deal to offer precisely in its questioning of preconceived notions through careful case studies and comparisons; though this questioning is not quantitative or formal, it is still quite necessary for any scientific endeavor. Discussions with anthropologists who do not use these methods would push these discussions further and enrich social science history.

In what follows, I provide an example of how the current questioning by anthropologists of the notions of place could affect one of the most “scientific” endeavors pursued within social science history: historical demography.

From Places to Flows

Even the seemingly “hard” science of demography is engaged in questioning its traditional paradigms (see Greenhalgh 1995, 1996; Hodgson and Watkins 1996; Watkins 1993). The nation, the usual unit of comparison within demography, is being questioned from within—on the basis of its internal diversity—while the local community, the usual unit of historical demography and anthropology, is being questioned from without—on the basis of its location in a larger system. Linking my discussion with questions anthropologists and others are asking about their traditional units of analysis, I propose that historical demographers perform an analysis based on migration streams rather than locations.

In questioning the nation-state, historical demographers are leading the way because they have information about a time when vital rates varied much more than they do today. When there is a high degree of variation among the constituent segments of the nation (whether these be villages, counties, regions, or classes), the usual comparison of nations’ means—the type of comparison on which demography has in the past depended—creates a misleading impression. Rather than a common experience of all its citizens, the mean in this case is a compromise between the competing means of the constituent units. Furthermore, the historical series of national means or rates that social science historians have struggled so long to create may not be the story of any one actual entity. Rather than reflecting improvement for some hypothetical average citizen, for example, a declining death rate may reflect the shifting of the population between different regions or between the countryside and the city.

Local variations in mortality rates were considerable during the mortality transitions of the nineteenth century. Furthermore, disease environments were multidimensional, and different spatial units were relevant for different diseases (see Johansson and Kasakoff forthcoming). Waterborne diseases stemmed from certain types of “flows” and resulted in certain areal patterns, while airborne diseases stemmed from different sorts of flows and thus resulted in different areal patterns (Woods and Shelton forthcoming). The two may have overlapped in urban areas, creating pockets of high mortality from both causes there, but at the edges there were places that had high rates of waterborne diseases but not airborne ones, and vice versa, and there was quite a bit of variation among rural areas as well.

The older mode of analysis stressed places, isolated villages that served as a kind of litmus test for the national character that demographic behavior was supposed to reveal. To be sure, people would move in and out, and there may have been flows of water or air between them, too; but there was no means to document these flows, given that the usual source of data, family reconstitution of single locations, the unit studied in historical demography, remained a unit with fixed boundaries, whether the local community or the nation-state. This type of unit was useful enough as a heuristic device when anthropology was making community studies through participant observation, but it is now understood that even the most isolated community is part of a larger system or at least must be seen in terms of some of the other entities to which it may have reacted in its isolation.

Is a locality a useful unit of analysis at all, if people move as often as they do? Should we continue to make our analyses as though the globe were a set of spaces, each supporting its own “community”? This “billiard ball” approach was questioned by Eric R. Wolf, who recognized that even the isolated societies studied by anthropologists were embedded in a world system (1982), but even at the core of that system, it has been difficult to conceive of the connections between the localities that are usually studied. Ideas about processes and linkages that go both from the national to the local and vice versa, probably with a regional level in between, are very much needed. In my earlier example of disease environments, the spatial units were created specifically for the reporting of disease. Boundaries were drawn to isolate groups of about 50,000 persons. But disaggregation of areas into microenvironments can proceed ad infinitum down to the family, in the case of infant mortality. Now that we are able to collect such fine-grained data, the question becomes, how do we build our units up and break them down, for what purposes, and under what circumstances? G. William Skinner (1964) and others who did regional analysis in geography and anthropology several years ago offer models of regional differentiation (Smith 1976). Today these models seem too mechanical, but they are a starting point.

Many local studies in historical demography were based on the reconstitution of the records of single parishes. This method often excluded migrants because it was difficult to obtain information on the birthdates of people who moved into the parish, and it was quite difficult, if not impossible, to follow those who had left to their new places of residence. But the exclusion of migrants also followed from the way spatial units were conceived. A stable and unchanging premodern countryside of interchangeable towns contrasted with “modern” flows to cities. Migration was assumed to be irrelevant because the small units in the countryside were interchangeable—the view put forward by the British in their family reconstitutions—and migrants into a parish could thus stand as a proxy for those who had left. But in any case, it was thought that migration in the countryside was repetitive and occurred only in response to life course events, such as finding a spouse. Thus, migrants were thought to be demographically the same as people who stayed in one place their entire lives.

When an established local community is the focus of study, it is hard to escape a negative image of migrants. Because they left, they are viewed as misfits, or disloyal, or simply as imprudent, given their overreproduction. Migrants may be seen as disruptive—if not actually so, then at least in terms of the scholars’ paradigms—because of the difficulty of even describing their lives. When the focus is the formation of new communities or their growth, as in the study of the “rise of cities,” migrants are often extolled. But even in these cases, studies of migrants’ origins are quite difficult due to the scattering of sources and lack of pointers back to birthplaces, let alone places where they lived between their birth and residence in the community of interest.

But migrants cannot be ignored, especially now that the technology and sources have been found to follow them. The computer, the existence of national registration in at least some countries, and intrepid genealogists (see, e.g., Kok 1997; Pooley and Turnbull 1998; Rosenthal 1999) have made it possible to gather systematic historical data on migrants and to follow them over their life courses. Yet our units and modes of analysis have hardly adjusted to the new materials.

Now that data on the life courses of both stayers and migrants are available, one can see that they are quite different demographically. We have demonstrated how the lack of migrants in demographic studies biases ages at vital events downward (Kasakoff and Adams 1995; Ruggles 1992). Ages at marriage or at death taken only from records of stayers—that is, people born in a locality and remaining there—cannot be generalized to the entire population of a nation or region because people who experienced the event of interest after moving—usually later in life—are not included. Louis Henry (1980) recognized this problem and devised an elaborate set of methods to correct for the absence of migrants in family reconstitution. Such methods, however, have not always been used (see, for example, the English studies).

One would expect to find differences between migrants and nonmigrants, because migration is always selective. When empirical studies do not find such differences, this must result from a complex set of countervailing forces that even out (Wrigley 1994). The bias downward for stayers may be offset by the very real difficulties they might have in setting up new households in crowded villages; thus, their age at marriage may on average be the same as movers’, but for quite different reasons. Explanations must differ in every particular case. The detailed study of those differences, or of the countervailing effects that make them the same, should reveal important forces of social change. But migration and migrants cannot be ignored even in this case; nor can it be assumed that social processes among migrants are the same as those among stayers, or that migrants into places are “the same” as the people who leave them.

If we were to reverse the usual picture of stable “communities” with migrants moving between them and instead put the focus on what used to be negative space—the migrants themselves—how would our ideas change? The individual places that we usually see as so important and so disconnected would appear as temporary confluences of people who all had links to their previous places and whose experiences even hint of their next residences.

This has been accomplished for the very mobile segments of populations, those whose movements have been described as “circulation,” and in the recent recognition of transnational communities (Basch et al. 1994), but these studies do not go far enough. They continue to use the bounded community as their unit of analysis. The “community” in this case simply exists in several locations; the flows in and out of this larger set of linked locations remain unexplored. As long as the unit of analysis is a bounded unit and there is no study of who comes and goes, of the boundaries themselves, the study of flows is incomplete. Further, these kinds of “communities” have been contrasted with groups of citizens with full rights in the countries closer to the core of the world system, who supposedly do not lead such migratory lives. In this case, migrants are again treated as exceptional. Denied full citizenship by the country where they are working, they are seen as forming the lower tier in a dual labor market. Rather than contrasting people with full citizenship rights with people without them, seeing one as stable and the other as migratory, scholars need to let migration take its rightful place in the lives of everyone.

A true “revolution” in ideas about places would include not only flows that turn in upon themselves to form a “community” but also bifurcations and permanent leaving. The units could be conceptualized as migration streams that exist in time as well as space, some of them drying up as new ones diverge. The local community then becomes less interesting than the flows in and out and is subdivided according to the origins of people who live there as well as their potential to leave. Seeing the streams themselves as ephemeral, or at least as existing for only a limited time, allows us to study the rise and fall of communities or parts of them. People do leave cultures; they relocate—a process that has not often been studied systematically. Yet how new cultures arise and absorb old ones in concrete demographic terms is quite important. The time periods that form the basis of historians’ specialties may not capture the full duration of these migration streams and may make it difficult to study their entire course. Using a similar view in his study of West Africa, Jean-Loup Amselle (1998) describes a precolonial chain of ethnic groups; when one of them moved or its “powers” changed, so did the others. He sees history as “a pump” that “constructs state and suppresses or peripheralizes segmentary societies,” the state as “a mass that alternately contracts and dilates.” Each ethnic identity or political unit there exists in a dynamic field.

Further, what anthropologists or community historians may valorize as “wholes” may have been local not by choice but through a process of marginalization; ties to larger groups and to opportunities at the core may have been systematically cut off or denied. National governments have policies that create locality, as Germany did when it decided against having national museums after World War II: the local and the supralocal are always in tension. Localness is not a natural state of affairs.

In their discussions of globalization, postmodernists, like Marc Augé (1995), talk about “nonplaces,” conjuring up images of the resolutely nonlocal by calling attention to what they think is a new type of space, a key attribute of “supermodernity.” These locations—airports, supermarkets, and the media, for example—are the negation of the locality anthropologists used to study. They arise, at least in part, from flows of people; some are marginalizing, such as refugee camps and public housing, but others are the result of privilege, such as tourism. In either case, there is a lack of permanent social ties. Instead of seeing these places as somehow “beyond culture,” I would rather see them as creating new forms of nonlocal cultures. While all are in this sense the same, each is also different and unique.

Even Augé grants precursors of supermodernity, in prophets, messianic movements, and syncretic religions, all of which he calls “contact phenomena.” He has singled out the massive transfer of people in slavery, which had the same characteristics of shrinking time and space. These movements would also have created meeting grounds for people from different local cultures. So rather than being a new invention, as writers about the postmodern world often insist, nonplaces have always existed. There were probably always crossroads where people from different localities could meet, though social scientists, in their desire to assign individuals firmly to a strong local culture, have often ignored them.

Recognizing the ubiquity of flows of people also sheds new light on the ideas about purity that are so much at the heart of today’s ethnic conflicts. Thus, it is important to document the blendings and nonrooted peoples that certainly existed in the past, the movement in and out of nations and ethnic groups, a task for which social science history is uniquely suited—if only the documentary evidence can be found! In anthropology, the idea of “ethnogenesis” (Hill 1996; Moore 1994) has been used to call attention to the creation of new ethnic units out of particular interethnic contacts, diverse remnants that came together rather than remaining “rooted” forever in a particular homeland. This theory thus questions the older anthropological model, in which a “natural” process of local differentiation was thought to have produced cultural differences.

Culture and Flow

What concepts of culture are appropriate for this world of flows? I have found that writings about Africa, surely the continent least represented in the Social Science History Association, are pointing the way. Because it lacked a dominant religion or cultural group, its indigenous empires were in disarray when “discovered” by anthropologists, and writing was a late intrusion, Africa has seemed, from the colonial point of view, most in need of anthropological ideas to explain its condition. Having been colonized so thoroughly and for so long, Africa was perhaps the first continent to have truly experienced the destabilization that is now associated with globalization (Augé 1998: 116). Perhaps it is understandable, then, that in Africa the concepts anthropology devised early on would be exposed as all too clearly lacking.

Amselle, who has done extensive fieldwork in West Africa, critiques the static nature of the classical concept of culture. “Culture . . . can only be considered as a set of incorporated rules by not taking into account how it has come to be, and by erasing the conditions of its very creation. To lay cultures ‘flat’ is to ‘photograph’ them at a given instant and to refuse to consider them as living entities. It is precisely by overlooking the conditions of production of a culture that one can treat a culture in such a way.” One important condition is its relation to other cultures. Cultures are not all equal. Instead, he points out that “there are cultures that have the power to name other cultures and to circumscribe their own field of expression, while others are only capable of being named.” But, returning to his original theme, he stresses that “nevertheless, the system is by no means static; formerly subordinate cultures become dominant while others, like stars, are born only to disappear” (1998: 33). To Amselle, the goal of anthropology is to study the process whereby cultures and the identities they contain are born and disappear. Because cultures and identities change over time, the new view of culture must be temporal. Both anthropology and history will be needed to chart such processes.

As anthropologists in Africa turned from “tribes” to religious and popular movements that attracted individuals from many localities and tribal origins, the idea of a self-contained culture with clear “rules” had to be discarded. The Africans themselves recognized that cultures were prisons that were being used against them. One such Christian movement that Johannes Fabian (1998) studied rejected tribalism and racism, those cherished units of anthropology, along with individual selfishness, because they saw how these units had been used by mine managers to divide workers.

So instead of studying the traditional ethnic unit, a tribe, Fabian writes about popular culture (1996 (1998), which he says always appeals to more than one ethnic group. Popular culture exists only in opposition to the specifically ethnic and local; it is nothing if not cosmopolitan. Thus, it does not seem that the notion of the ethnic and the local needs to be abandoned. Instead, this kind of culture should be viewed as one form of identification that popular culture is struggling against. There are competing and overlapping identities and cultures, some more local, and others, like popular culture, less so.

Amselle (1998: 4) has proposed that each culture be seen as a reservoir of differences, “a collection of practices internal or external to a given social arena that the actors mobilize as a function of one or another political conjuncture.” In West Africa, before the nineteenth century, people could change their ethnicities and other features of their identities that we would consider fixed: their names, their taboos, their personal double, and their interlineage pacts, in which case these seemingly individual changes had political impact (1998: 134). According to Audrey Smedley (1998), this was also the case for racial categories in earlier times. Whether it was the “modern” obsession with identity, which Amselle attributes to the literate bureaucratic nineteenth-century state, or whether there were times and places even before this where such categories were rigidified, is another question that social science historians are well equipped to explore.

Amselle is at pains to point out that the supposedly timeless comparative categories anthropology constructed during its modernist phase under the guise of science were mirror images of the colonizers’ own societies, suspiciously like the categories used centuries before. For example, just as the Greeks defined people who were not organized in city-states or centralized bodies as ethnos rather than polis, so too did the anthropologist. And this practice continues today. It happens not only in the way the core defines the periphery but also within the periphery; within West Africa itself, the dominant groups characterize the rest in the same way. So the segmentary societies anthropologists claim to have discovered as a form of radically different political organization, one independent of states, could also, in fact, be found on the peripheries of states and existed only in relation to them. Amselle extends the argument to religious typologies: the Muslims at the core of West African states viewed the local religions in exactly the same terms as the British anthropologists did, seeing them as animist, although these supposedly “pure” and pre-Islamic religions had incorporated many Muslim features long before the nineteenth century. Even the Muslims who encountered them years later did not recognize them as such.

Each attempt to find a “pure” culture, whether in the religious realm or the political, flounders because the “locals,” the new conquerers, were themselves interlopers at some time in the past. The primordial concept of ethnicity that formed the basis of anthropology needs to be abandoned as a description of “what happened,” though it is clearly important as a way of constructing ethnicity, which recurs under certain conditions.

Rather than eliminate typologies like “state versus segmentary” or “animist versus non animist,” Amselle suggests that we see these typologies as recurring ways that ethnic groups perceive themselves. Each of the poles of the typologies anthropologists constructed under colonialism are simultaneously present in any set of ethnic identities in a particular place. There is, therefore, a sense of unfolding possibilities, in which ethnic groups are continually redefining themselves instead of being bound by rigid and absolute differences.

Thus, after a long absence, these ideas about fluidity may bring individuals back into social science history—not the heroic great man, the original enemy of social science history, but the notion of what it means to be a person and especially the question of agency, the kinds of choices individuals make within their cultures. Social science history has long advocated a view from below, but in the process it too often created a faceless mass, a set of trends. Augé (1998: 117) describes his work on African healers and prophets as a study of “the multiplicity of individual trajectories that converge there (i.e. among their clients) more or less durably or ephemerally,” an idea quite similar to the migration streams I have sketched above. In the media age, he writes, “with the death of exoticism, a kind of short-circuit is produced that confronts each individual directly with the image of the world. Difficulty in symbolizing relations between people is stimulating a multiplication and individualization of cosmologies” (ibid.: 121). It would seem that today we have a whole set of modern Menocchios (Ginzburg 1980), but they cannot yet be placed at the intersection of an important transformation because we do not know the future.

Dismantling Disciplinary Domains

Social science history has been a meeting ground outside the confines of traditional disciplines. It is here that at least some of the talking across disciplinary domains has occurred in the past, and this makes it uniquely situated to provide a forum for the further disintegration of the fourfold box with which I began. Many of these changes have already occurred within anthropology. Here I am suggesting that they be carried forward within social science history.

So here is my recipe for a new social science history. First, compare adventurously across the dimension “exotic versus home.” Too often “comparison” within the association has meant comparison of the United States with Europe; “others” are lumped together in sessions that seem marginalized, at least judging from the number of people who attend. Studies of ethnic identification in the Balkans should be juxtaposed with those in West Africa and other areas that ethnographers traditionally study. Anthropologists in urban settings away from Europe and America can learn from students of those processes in the core and social scientists in the core from students of these processes in peripheral areas. We need a discussion of how and why certain social groups with good written records have become treated as “isolated closed ethnic groups,” as the classic anthropological monographs treated them. The more we are able to see the exotic in the present and at home, and “home”—world religions, popular culture, nonplaces, and the like—in places that used to be the location of the exotic, the better. I look forward to a future in which these very dimensions that have divided our disciplines are questioned further, when we ask about ourselves why and for what purposes the overly simple anthropological model of the bounded community isolated from all others was appropriated to describe what surely must have been a very complicated situation in the histories of both Europe and North America.

Second, make studies more systematic. The new questions related to flows and the terms so much in vogue like globalization, transnationalism, identity, diaspora, and the like need to be grounded. Such supposedly new processes need to find their analogues in the past. A new level of analysis bringing together the various case studies is required. This move would lead to similar measures and a set of “middle-level” generalizations. Arjun Appadurai (1996: 27–47) has called for new analyses based on several sorts of flows: technology, ideologies, media images, and money, in addition to people themselves. We need to take these ideas further, asking about relationships between flows of different types, origins and causes, and the forms these flows take in different social and historical contexts. We need to measure them and bring quantification into these sorts of discussions. Yes, we certainly need studies of the new cultural meanings people create as the result of such flows. But these studies need to be theorized against volume, direction, timing, and scale.

Third, dismantle the “past versus present” dichotomy in the fourfold box. One way to accomplish this is to take timing and temporality seriously. History deals with time, but it does not really conceptualize it. Here the ideas of temporal scale, longue durée, conjuncture, and event were a beginning (Braudel 1980). Too often the ideas of “change” or time period are too particular, and similarities of process and timing, the unfolding nature of events and the way the future is constrained by the past, are ignored. Sociologists within social science history have created formal models and do explore these questions, but their approaches need to spread.

Another way to break down the disciplinary barriers is for anthropologists to do history and historians to do anthropology. This is already happening, of course, and in this sense the anthropologists already active in the association are in the vanguard. But I do see one threat to its continuation. In anthropology’s attempt to rid itself of the exotic, some have suggested that the discipline define itself by its methods. If it does so, anthropology will be forever in charge of the present due to its unique field methods. These may be quite useful in the attempt to compete with sociologists in the attempt to colonize the space of the present at home. ⁵

But this sort of “home work” (see Fricke 1998) has been almost entirely ahistorical. It has become entirely commonplace for anthropologists to call for their local studies to be “historically situated,” but this has not always involved historical investigation using historical sources. Too often a mere description of “what happened,” or an interesting hypothesis, passes for a discussion of history in these accounts. Of course, quite often the “history” in question is difficult to reconstruct, lost due to lack of information. But interaction with historians interested in the same problems would bring to anthropology badly needed information about historical sources and how to use them more fruitfully, as well as how to conceptualize gaps and lost histories. Furthermore, to simply “situate” a given “community” in “history” reifies the community when I have called for a deeper examination of its genesis, if indeed the term community is useful at all. And certainly anyone who participates in social science history comes to a deeper understanding of what “history” means as well. It is a style of argument, of competing interpretations, not just a question of dates.

So I would argue against letting historians do the historical part. If anthropology’s object is no longer to be a bounded local unit and anthropologists are to look at trajectories and movements, then they must do so in time and must theorize time as well as space. This will mean that more and more anthropologists will make the journey my husband and I, and the others who regularly participate in the association, have made.

In this short discussion, I have suggested three major ways in which anthropology can find a more secure place in social science history. First, social science history can concern itself more with the exotic, the usual domain of anthropology. This seems to me to be the least likely route, since the association has dealt so exclusively with the core of the world system in the past. Second, the association can continue its pattern of attracting anthropologists who study close to home by encouraging the growing number of anthropologists studying home to become historians, to study the origins of the groups and patterns they are studying rather than assuming that they are unique features of modern life, producing the kinds of static pictures that Amselle cautions against. This seems to be the most likely way for anthropologists to solidify their place in the association, because it builds upon past patterns, but it will require some active work. The third way is to use anthropology to question received ideas, to talk across disciplines about common questions and projects. I have developed this approach here by way of example in my discussion of ideas about place and spatial units of analysis. It is also, I think, already under way within social science history. It simply needs more encouragement.

In questioning received ideas and fostering debate, social science history is forging the intellectual categories needed to live in a postmodern world. To many, the postmodern moment may seem to be one of particularity and diversity, but I think it is interesting that so many of the anthropologists coming to terms with modern Africa, in their attempts to break away from the typologies of social science that privilege certain peoples over others, have come back to a more universal idea of what it means to be human. So, indeed, have many of the Africans they write about when they attempt to rise above the tribalism that is at the core of much of traditional anthropology, especially those in urban areas.

Is the task of global history to describe a set of historical connections between places on the globe, webs that are particular and historical in the sense that they are unique, or is it to describe a set of repeating processes of development, which has come to be known as comparative history? My answer would be “some of both.” Surely the connections and flows that occur and have occurred between localities have repeated themselves at least partially. They can even be seen as engineering problems—a view that opens the door for a very scientific approach. There can be only so many ways to structure trading routes, for example; their costs and benefits would have to be reflected in prices. On the other hand, the globe has been on a trajectory of greater connectedness over time, which has a unique history that has not repeated itself. Amselle perhaps offers us another view in his sense of waxing and waning centralizing tendencies that draw chains of societies in their wake. Unfortunately, however, the new macrohistory network that discusses such processes on a global scale seems divorced from the rest of the association. Few people with interest in particular regions attend their sessions.

We might expect to be gun-shy of universals now that we see the damage that has been done by colonialism and the modernist project. Yet I would argue that we have a great need to rise above the particularism and divisiveness of our times, politically as well as intellectually. We can accomplish this goal, at least partially, by turning the categories of analysis created by the colonial and modernist projects of the past into objects of historical study. If, through this approach, we can envision a new set of dimensions, we can at least hope that they will not be as harmful as those of the past.