Kenneth Pomeranz

President of the Association, 2013Kenneth Pomeranz. Photo by Cliff Moore

This presidential address was delivered at the 128th annual meeting of the American Historical Association, held in Washington, DC, in 2014. You can also watch a video of the address.

Histories for a Less National Age

A year as AHA President has offered good opportunities for reflecting on a topic that has long concerned me: how our discipline is responding and might respond to the changing geography of our own era. There are at least two distinct phenomena involved, though they are interrelated, and often get lumped together under headings like “globalization.” Moreover, we are responding to these changes very differently in our research and in our teaching. Later, I will offer suggestions for narrowing that gap.

The first development involves the increasing importance of flows between what has loosely been referred to as “the West” and parts of what is even more loosely called “the rest”: movements of people, goods, pollutants, ideas, and so on in both directions. One result has been a greater awareness of past flows across these same spaces, and of the degree to which Westerners have often ignored dynamism—and thus the existence of “real history”—elsewhere in the world. This has led to calls for devoting more of our research and teaching to what Luke Clossey and Nicholas Guyatt called “wider world history” in an article for the May 2013 issue of Perspectives, which was followed by an online forum.1 Clossey and Guyatt argued that “the West” is still hugely overrepresented in most U.S. history departments—though less so than it used to be, and less so than in Canadian and UK history departments. Other participants in the forum suggested that the rebalancing of attention that Clossey and Guyatt seek had already gone a bit further than they indicated, but the contributors generally agreed that this process was far from complete, and they pointed to possible opportunities for speeding it along—including some that might avoid a zero-sum game with European and U.S. histories. (This allowed everyone to politely avoid discussing whether there was an ideal balance, and if so, what might need to be reduced in order to achieve it.)

The second development involves an increased questioning of the units most often emphasized in historical analysis, whether of the West or of other places. Chief among these is the challenge to “methodological nationalism”—the still-powerful nineteenth-century assumption that nations were the logical containers of meaningful history, serving as the most important point of reference even for research and teaching conducted on different scales, so that studies of, say, Pittsburgh, Omaha, and Atlanta ultimately mattered as they helped us build a better story of “the United States,” or as they might be juxtaposed to Stuttgart and Essen for the purpose of comparing the U.S. and Germany.

Of course, methodological nationalism never had quite the hegemony that we sometimes imagine, particularly for studies of earlier periods. And it has faced many challenges in recent decades, especially from histories which assert that class, ethnicity, gender, sexuality, and so on were often more meaningful historical units than nations. As many of those new histories pointed out, nations often proclaimed a certain homogeneity that really applied only to small parts of their population, and actively suppressed (sometimes with the aid of historians) the narratives and identities of other groups.

What has happened more recently, though, is that these challenges from below to methodological nationalism are now accompanied by challenges from history on larger scales. There are now more credentialed historians who say they teach and/or write “world history” than at any point since the nineteenth-century professionalization of our discipline. The new histories from “above” the nation seem harder to reabsorb inside a national frame than most challenges from below—a gendered history of Britain's World War I mobilization is, after all, still about British history even as it disaggregates that unit. And while teaching units larger than the nation are not new, previous examples were rarely used to challenge methodological nationalism, and indeed often reinforced it. Both Western Civ and civilizational area studies units—mostly utilized in introductory teaching—fit comfortably within a system where serious research and advanced teaching were generally reserved for national units, and they raised some of the same intellectual issues (e.g., of false claims of internal homogeneity or of a shared cultural essence). Consequently, they face many of the same criticisms leveled against methodological nationalism, plus others, and do not have the self-evident claim on our attention that nations get by having armies. Thus their status is considerably shakier still than that of national history. They do not currently seem likely beneficiaries of moves to de-privilege the nation, except perhaps in some radically altered form.

As I hope to show later, world history can potentially do something very different for us from what these other supranational units have done. At any rate, the new challenge from “above” represented by world or global history has been accompanied by other challenges that question both national and civilizational frameworks by devising units that cross borders without claiming to encompass the polities and cultural areas they slice through. This includes research and occasionally teaching that focuses on diasporic groups, transnational professional or intellectual networks, and other spatially dispersed groups that might nonetheless share some sense of identity; studies of commodity chains; and studies of interactive spaces such as the Atlantic littoral or the Silk Road, which are interesting in part because their dense networks of interaction neither relied on nor produced much movement toward cultural homogeneity or shared identity.

The push for broader inclusion and the push for different units can reinforce each other. Those interested in large-level syntheses clearly need monographic research on many still-neglected areas; and studies of, say, Indian Ocean networks are likely to stimulate writers of East African national histories in many of the same ways that studies of Atlantic networks, southwestern borderlands, and the Great Lakes “middle ground” have energized U.S. national history. But the people emphasizing more inclusion and those emphasizing new frameworks often talk past each other or work at cross-purposes—partly because we have thus far reacted to these challenges rather differently in our teaching and in our research.

While reframing represents a response to both new research findings and our changing experience of the present—the preferred stimuli for historiographic change—it also represents responses to potentially far less welcome trends. These include powerful political pressures to focus education more narrowly on what will supposedly yield immediate economic returns, rather than on nurturing knowledge and habits of thought relevant to the reflective citizen or individual (or even to the later phases of a person's career). And there may also be cultural reasons to worry about a general privatization of concerns that can easily marginalize history. Even twenty years ago, Roy Rosenzweig and David Thelen's big survey of Americans' uses of the past pointed to a probable decline in the number of people who felt that major public events had influenced them, and a strong tendency, even among those who did feel that public events mattered, to put their importance in very personal terms—that a war had created pressures to grow up quickly, for instance—rather than in terms of a stake in the outcome of public contention.2 De-emphasizing national units need not mean abandoning history with direct political relevance, but it runs that risk—though the state is not, in fact, going to wither away any time soon.

In such an environment, moreover, making “wider world” history seem relevant can be particularly difficult, despite rising awareness of the economic importance of the world beyond U.S. borders. On the one hand, one could imagine Rosenzweig and Thelen's most basic finding—that people find the past interesting and meaningful when they can place themselves or their loved ones in relationship to it—reinforcing various other arguments for doing “world” courses, since everyone can locate themselves within that unit, rather than multiplying national or area-focused courses on parts of the wider world. Moreover, many advocates of world history, myself included, have argued at times that the world is precisely the unit with which we should encourage people to identify: if national history helped strengthen the sense of national citizenship, the argument goes, what could be more appropriate for a world with so many pressing and inherently global challenges and such enormous inequalities and barriers to solidarity than to provide histories of all humanity? But dangers lurk here, too, even beyond the obvious ones of superficiality, and of financially pressed deans deciding that a department with some members who can teach “world history” does not need an army of Africanists, South Asianists, East Asianists, Europeanists, Latin Americanists, and so on.3

In contrast to area studies (which for our purposes here includes American studies) and national history, which both emphasize cultural particularity, and to transnational but non-global units like the Silk Road or the Mediterranean, most global history narratives thus far have been strongly materialistic, and focused on general processes. They thus have the potential to unwittingly reinforce a popular view of “globalization,” which acknowledges the growing economic importance of the wider world, but claims that the world is becoming a sufficiently seamless marketplace that one need not understand particular histories or cultures to function in it. As just one indicator of how common this view has become outside the academy, consider the terms used for “wider world” societies. Texts from 1989 later scanned into Google Books used the phrase “developing countries” roughly forty-three times as often as they used “emerging markets”; by 2008 (the last year for which data is available), that difference had dropped to a bit over 2:1. In the daily press, the reversal has gone further: the New York Times used “developing country” more than twice as often as “emerging market” from 1990 through 2000; for 2000–2013, it has been more than 2:1 the other way, and for the last twelve months more than 3:1.4

Some other, still larger-scale, world history narratives—those of Ian Morris and David Christian, for instance—build a coherent story around the interrelated growth across millennia of human population, technology, knowledge-sharing networks, energy consumption, and environmental impact, leading to a present that could usher in either sustainable universal prosperity or ecological catastrophe.5 This is a truly global story, bringing in places and epochs that historians have usually avoided; it builds potential bridges to other disciplines, argues for the necessity of a common human citizenship-cum-stewardship of the planet, tempers the triumphalism of many “globalization” narratives, and provides a historical context for some of today's greatest issues. And it develops precisely the mutually reinforcing “triple helix” of themes—population growth, increased technical capacity, and increasingly dense long-distance connections—that Jerry Bentley (author of Traditions and Encounters, the best-selling world history textbook, and founding editor of the Journal of World History, which was for years the only scholarly journal explicitly devoted to world history) picked out as most suitable for analysis on a global scale.6 But it has its own problems, some of which I have discussed at greater length elsewhere.7 Among others, it tends to reduce history to the history of technology, and contemporary issues to the search for a technological fix for our fossil fuel addiction; meanwhile, questions of politics, of who exactly has created and benefited from economic growth, environmental damage, and increased cross-cultural contact, all tend to be elided by a narrative in which “humanity” becomes the actor that we follow through time.8

Narratives on very large spatial scales also tend to use large temporal scales, in part because they highlight themes and processes that touch many places but may do so decades or even centuries apart. They can thereby become disconnected from the time scales of human lives—and thus from both the humanistic project of recovering people's experiences and the social science project of reconstructing decision mechanisms and causation.9 Since working on such scales requires more frequent explanatory recourse to motives and forces that do not change much, it can also thwart our efforts to show that certain things which contemporary society and/or disciplines that take their behavioral assumptions from it tend to ascribe to timeless human nature—including the privileged status of “growth” itself—are in fact historical products. In short, it is much better at making the strange familiar than at making the familiar strange, or at least contingent; and history needs to do both. My point here is not, however, that we should therefore avoid such very large scales, but rather that we need to clarify their relationship to other scales. Different historical scales do not nest neatly within each other, like Russian dolls; they are focusing devices that always obscure some patterns to make others stand out, with consequences that can be moral as well as intellectual.

This problem exists whether the scales in question are small, medium, or large—and we are hardly the first generation to notice the issue. Fifty years ago, amidst a wave of decolonization that seemed to represent the near-universal triumph of the nation as an organizing principle for human societies, David Potter sounded two cautionary notes for historians. First, he emphasized that the apparent primacy of national identities could only be a relative and historical matter: the sense of belonging to a nation had no necessary priority over religious, class, or other solidarities, and much of its apparent ascendancy at the moment reflected a widespread but non-obvious belief that nations, unlike, for example, classes and sects, had the right (within some unspecified limits) to wield force. Second, because identifying some group as a nation licensed it, at least in some eyes, to do things that would be unconscionable if other groups did them, the stories historians told about the emergence and further histories of nations carried large implications, whether they liked it or not. Thus, as historians “made use of nationalism” (as a concept)—and quite legitimately focused much of their attention on it, since it was one of the most important phenomena of modern times—they also needed to remain wary of how nationalism made use of historians.10 In a similar spirit, we need to think today both about how we use the global and about how various versions of the global may use us.

Consequently, despite my own commitment to the emerging field of world history, what I offer here is not intended as an example of what Gordon Wright, categorizing the presidential addresses before his, called the “manifestoes.”11 Instead, I see it as a set of questions about how we are coping with the changing geography of our world, containing as much uneasiness as celebration or exhortation. It thus follows more in the spirit of Joseph Miller's 1999 address. Miller asked what the processes that eventually created an “African history” cognizable by the rest of the discipline revealed about the discipline; he simultaneously asked how the discipline as a whole might profit from observing what had been necessary to meet professional standards of history while studying places nobody had in mind when those canons were created.12 While world history is not as firmly established in the discipline as is African history, it seems to me sufficiently accepted to ask similar questions about what has been adopted, changed, and jettisoned along its path to inclusion. Its greatest value, I will argue, is not that it is necessarily the best scale on which to do history, but that it helps force into the open the necessity of being self-conscious about all scales.

Let us look, then, at the rather different ways we have been responding in research and in teaching, and at where the opportunities and pitfalls may be. If we consider trends in historical research, two things become evident very quickly. First, research on transnational topics is booming, particularly for the early modern and modern periods. (Research on earlier periods was never as fully “nationalized” to begin with.) It takes many forms, but is generally concerned with mapping networks of one sort or another that cross national, and often area studies, lines. This work is getting plenty of recognition from the discipline at large; indeed, it seems to be riding a wave of enthusiasm. Fewer than 40 percent of the books we honored with prizes last year framed their subjects within a single polity, even though many prizes were originally endowed in national terms (e.g., the best book on English history).13 The books that have been selected for “featured reviews” in the AHR since that section was reinstituted in 2007 have a roughly similar profile: only 36 percent of the 135 books in question were on topics clearly contained within a single modern nation. Another 12 percent were explicitly framed with reference to an area studies region such as Latin America or South Asia, making such books exactly as numerous in the featured reviews section as those that were either explicitly global or framed by non-traditional regions such as the Indian Ocean. Fully 36 percent crossed national lines in one way or another—by following networks of trade, migration, scientific collaboration, and the like, or by comparisons of some sort—but without specifying some larger geographically defined object: their subject matter might all come from various parts of Europe, for instance, but without any attempt to make claims about Europe as a bounded whole. These numbers are strikingly different from those for our recent research as a whole: more than 70 percent of all the books reviewed in the February, April, and June 2013 issues of the AHR, for instance, would have fallen firmly within the “single nation” category, or roughly twice the percentage as in these two groups of books singled out for notice. (Data for five prestigious presses was in between, but much closer to the distribution for all books reviewed than to that for books receiving featured reviews.)14 They also represent a striking shift away from national units (and “the West”) when compared to the books selected for featured reviews when the AHR previously had that section, in 1993–1996.15

Very little of this research, however, is actually at the global level. Indeed, except for environmental history and histories of very recent times, there is not much research that takes the whole world as its focus. This is not particularly surprising. Aside from the very general trends already mentioned, probably only a few aspects of history are best approached on a truly worldwide scale. Most world history instead focuses on mapping specific, limited sets of connections—of trade, disease transmission, intellectual influence, violence, or whatever—which usually stop well short of spanning the whole world, or else makes comparisons that interrogate claims about the universality or regional specificity of particular processes by looking at a few specific, bounded sites. And while exploring these specific connections or comparisons can be considered to be doing “world history,” there is no reason that it has to be, any more than it must be called doing the history of “the Atlantic,” “Eurasia,” “Africa,” or any other old or new space that might be drawn by enclosing the dots that such work connects.

Meanwhile, though this kind of spatial reframing apparently meets our approval as research, it is hard to find in our curricula. This spring and summer, I organized a survey of the courses offered by a representative sample of 218 AHA member departments at colleges and universities.16 At the introductory level, 46 percent of all classes last year were on national histories or histories firmly contained within national borders; U.S. history alone made up 35 percent of introductory classes. Another 20 percent were either in Western Civ or, more often, “area studies” culture areas, such as European or East Asian history surveys. Of the 34 percent in all other categories put together, fully half were world history surveys. Histories of non-traditional regions (e.g., the Atlantic world, the Indian Ocean, the Silk Road), courses that were based on comparisons or connections that crossed national boundaries but did not cover the whole world (e.g., the history of Buddhism, the age of revolutions, Christian missions, Gandhi and Mao, the Irish diaspora), and international relations all put together—in other words, the sorts of big but sub-global units that are much in vogue in our research—came to only 10 percent altogether.

At the advanced undergraduate and graduate levels, we find far more courses that have no geographic referent at all, because they focus on meta-topics: courses on theory and/or method, courses on historiographical classics, research or honors seminars that mix students from many fields, and so on jump to 19 percent of all offerings. But if you put those aside and consider only courses that have a time/place topic, the share of national histories and of traditional regions at the graduate level is almost the same as for introductory undergrad courses. There is some shift within the set of transregional/transnational courses offered at the upper-division level away from “world” and toward the other categories I described—indeed, world classes drop to a mere 4 percent at the advanced undergraduate level, though they make a partial comeback at the graduate level. Most importantly, though, the preponderance of traditional units overall remains almost unchanged.

Thus it appears that the kinds of stories we find it interesting to explore and to tell each other are much less “national” and “conventionally regional” than those that we tell our students, especially our beginning students. The contrast becomes particularly striking when we put it side by side with practices in disciplines relatively close to ours. In a parallel survey, only 6.6 percent of all courses in anthropology were billed as courses about a particular nation; 11.5 percent specified an area studies region; 10.8 percent at least implied that their unit of analysis was the whole world. More than 70 percent either focused on theory and method (at least in the course title) or saw no need to specify any geographic/cultural container. When it comes to research, anthropology is as committed as we are to contextualized “thick description”—it is an anthropologist's term, after all—and probably not coincidentally, it belongs to the dwindling group of disciplines that emphasize book-length scholarship almost as strongly as history does.17 But its members are much less likely to identify with “areas”: the AHA has about 20 percent more members than the AAA, but there are 160 percent more historians who belong to the Association for Asian Studies, and more than five times as many historians in the Middle East Studies Association.18 Sociologists are even less likely than anthropologists to define themselves or their courses using the units that we generally use.19 In political science, where state institutions naturally loom large, the number of courses focused on national and traditional area studies units was much larger than in sociology and anthropology, but still much less than in history—and even further away from the figures for history if one focused on the most prestigious programs.20 Such surveys are imperfect, but the contrast is stark. We are clearly much more committed than our colleagues to bounded, conventional units—or at least to taking them as a point of departure—and this shows up especially strongly in our teaching.21

In part, the differences reflect different theoretical concerns, which inform both research agendas and teaching. For a historian, it is fascinating to read the debates anthropologists have been having over the last two decades about the desirability of doing “multi-sited ethnography”—work that follows migrants, products, ideas, or other flows across space—as opposed to more traditional work anchored in one community, and to note similarities to and differences from our discussions about transnational/world history. In some ways, the concerns are very familiar—the time commitment involved in gaining a deep understanding of even one social setting, questions about the possibility of teamwork in research that aims at a qualitative, personal understanding, and so on.22 But it is also striking how centrally concerned the anthropological debate has been with how to avoid framing either the single or the multi-part field site, or any larger entity for which they might stand, as a “bounded whole,” and with calling attention to the objects of analysis as analytical constructions rather than real “communities” or “cultures.”23 Historians, by contrast, have generally been happy to acknowledge that no social unit is completely self-sufficient, while only occasionally calling each other's attention to that fact, and to frame what we offer to students with the most recognizable of these units—which are also the units that students are most prone to mistakenly consider natural.

Why, then, do we continue to rely on national (and other traditional) units more than anthropology, sociology, and even political science do? And why in teaching much more than in research?

In part, maybe it is because we study the past, and no current trends can change the importance of state and nation-building over centuries gone by. Perhaps, too, the fact that we must bridge the gap between present and past (and often between “our society” and the past of “another” one) means that to add a further level of strangeness by framing our stories through unfamiliar spatial units seems likely to be a bridge too far for some undergraduates. But there are also larger theoretical issues, and some very practical ones.

Historians have never gone nearly as far toward defining our discipline methodologically as other disciplines have. As Timothy Mitchell has pointed out—in an essay that, oddly, leaves out history—the high tide of area studies in the United States, from the late 1950s to the early 1980s, coincided with periods in which the social science disciplines invented as their objects of study supposedly coherent and distinct entities called “the economy,” “the political system,” “the social system,” and “culture.”24 (Remarkable as it may seem to us today, a Google Ngram confirms Mitchell's claim that “the economy” was not a very common term before the 1950s, and did not really take off until the 1970s; “social system” also boomed in the 1950s, and “political system” in the 1960s, before both declined again in the 1970s.) These systems belonged to nations, or perhaps to area studies regions. National or area specialists, particularly historians, then looked at how these systems interacted within a national body, much as a doctor might consider the interaction of the circulatory, digestive, nervous, and endocrine systems. (And like general practitioners, we often relied on these specialists to provide models that let us fill in what probably happened in their areas of specialty when our sources could not tell us.) But faith in the usefulness of these constructions crumbled in the 1970s and 1980s, as confirmed by the decline in the use of these terms, and the increasing reluctance of many anthropologists and sociologists to even use the terms “culture” and “society.” In other words, the people who had previously seen themselves as mapping the subsystems that each nation supposedly had lost faith in the existence of those systems as integral units at roughly the same time that methodological nationalism came under attack in various fields. With the objects that had defined them no longer reliable, and having largely ceded the particular (naturalized as the national) to others, economics was restructured as the home of a distinctive type of reasoning—which it now seeks to apply everywhere, leading to what the rest of us often see as disciplinary imperialism—while anthropology was reconfigured as the home of a method (participant observation) and lots of reflexivity. Sociology and political science tried to make equivalent moves, but failed to reach agreement on what their shared method and assumptions were, and fragmented into very loosely related subfields.25

History, on the other hand, experienced these epistemological crises very differently, because we occupied two different positions in the system of knowledge that Mitchell describes. We were specialists in the interrelationship (at the level of a nation or a civilization) of the “systems” in which different disciplines were invested, but that meant we had no stake in the integrity or self-sufficiency of any one of them. On the contrary, our stock-in-trade was to insist that you could not understand a society's intellectual life without exploring its social structure, or its economy without its politics, and so on. While many people were increasingly skeptical about specific ways of making these connections—notably Marxism and modernization theory—one could still assume that such links existed. Similarly, new histories of race, class, gender, ethnicity, and sexuality emphasized the value of deconstructing national narratives, but not necessarily in order to leave that deconstruction permanent; most such studies worked within a single nation, and their authors often explicitly aimed at creating a more inclusive national history, rather than showing that there was no such thing as a coherent society with a coherent history.

Historians did, of course, have an investment in the coherence of history itself—the idea that various sets of changes over time, whether serial or parallel, formed intelligible patterns; but that did not require a commitment to finding that coherence on any particular spatial level on any particular occasion. Though we most often found coherence on the level of the nation, it might also be on the level of an ethnic group (as in African American history), a civilization, or elsewhere. The coherent history we relied on might even be of an entity that was not our explicit topic, as when some national histories—especially non-Western ones, but even some European national histories, such as Germany's—were described in terms of their nonconformity with a stylized Western path that was “offstage” and thus unexamined. All of these objects have indeed now come in for strong questioning, but mostly in relative, not absolute, terms: Who, after all, would say that neither nations nor classes nor ethnic groups have any reality? Thus, our object has not been as badly shaken as in other disciplines; we have therefore felt less need to define ourselves by a method—which we almost certainly could not do without the discipline fragmenting very badly. Being less methodologically defined, we have found it easier to continue treating our topics as more or less bounded fields of which we illuminate certain illustrative parts (even if we are more inclined than we used to be to question the representativeness of those parts), rather than as a set of dots connected by the very visible, somewhat arbitrary, hand of the scholar/teacher. And those boundaries often make practical sense even when they are theoretically shaky: we can all agree that the U.S.-Mexico border is much less of a firm line in reality than it is on a map or in Westphalian theories of the state, but it nonetheless matters powerfully to people's lives.

So the good news is that we have been shaken up less than many other disciplines, and we can frame our teaching in terms of units that are quite legible and still make intuitive sense as at least heuristic wholes: no matter how much we or our colleagues may question whether there are “societies,” or worry about the ideological implications of imagining things called “nations” moving like monads through time, these units have an intuitive appeal. So does trying to figure out how the different kinds of human endeavor located within a particular set of those boundaries related to each other over a certain bounded time period—assuming that there must be something that ties together the philosophes, eighteenth-century French agrarian problems, rivalry with England, and so on, and that it is a reasonable project to get students to the point where they can discuss all these things (or explain why they don't have to) in explaining the revolution of 1789.

Moreover, these categories give us some valuable turf, which becomes all the more valuable to the extent that other disciplines are retreating from it, at least for pedagogy. There are merits—both for public service and for enrollments—in being the principal remaining department, other than perhaps literature, to which a student who just did a semester in Italy, or who is about to intern with a company that is heavily involved in Mexico, or who is starting to think that his/her future in-laws might be Vietnamese, can turn for nationally based courses. And at the risk of repeating myself—and the obvious—nations remain very important; they also become all the more historically interesting as they become less of a self-evident telos, and we instead need to explain how they became such important, often dangerous, foci of power and identification.

But why, when we look beyond the nation, does our teaching—as opposed to our research—jump straight to the level of the whole world?

The answer is probably overdetermined. On the one hand, there are pressures to fit our curriculum to those of our majors who are thinking of becoming history teachers—a hefty percentage on many campuses. Seventy-five percent of U.S. high school graduates now take a course labeled “world history,” but the expected content varies greatly from state to state; an offering that tries to give some introduction to everything thus has a certain logic to it, and can help assuage the fears of future teachers who are handed a set of standards including many topics they would never otherwise study.26 (By contrast, relatively few teachers have the chance to choose the overall structure of their classes, even if they wish they could; thus a world history field driven by “trickle-up” from high school standards will tend toward a more fact-based set of goals than most of us would probably like.) The same holds true when we think of world history as part of general education and preparedness for citizenship. If we seek to encourage global awareness, then a course that addresses the whole world seems logical. Even if what we seek is to interest students in some part or parts of the wider world, there is still some logic to remaining very general at the introductory level, and letting them choose which of the various places they've been introduced to they follow up on; that would produce something like our current curricula, where “world history” has become the second most common introductory survey, but upper-division courses overwhelmingly focus on nations or area studies regions. Moreover, the disjunction between the way we do introductory teaching and the way we frame our research fields is hardly new. As far as I can tell, even at its peak of popularity, Western Civ remained solely an introductory course, without any upper-division or graduate courses framed around it or faculty who considered it to be their specialty. Since world history has at least some upper-division classes, graduate programs, and self-proclaimed specialists to its name, we have perhaps slightly narrowed the gap between how we introduce our field and how we practice it. But the current moment gives us an opportunity to go further in rethinking the relationship between general education and the specialized production of new knowledge in our field—one that it would be a shame to miss, or to leave entirely to those who teach introductory courses.

There are several reasons to do this. If globalization has an implication for us as civic educators, it is not simply that “once upon a time, history made national citizens; now it must make world citizens.” Identities are multiple, and often mutually reinforcing rather than competitive; so, too, are the contexts for public action. As David Hollinger put it, in a slightly different context, we also have obligations to “a public smaller than the species.”27 Moreover, we might sometimes serve the whole species best by addressing that smaller public as Americans—or as “Americans in the world,” rather than demanding a choice between global and national attachments.

Meanwhile, the problems we face as a species also cry out for people to get as much practice as they can in selecting appropriate time scales for framing historical processes, and in detecting what might be missed if we focus exclusively on any one particular time scale. Environmental issues are the obvious example here—as many aspects of the climate change debate make especially clear—but any situation in which the short- and long-term effects of some important development are different will do, and history is, to a significant degree, about both tracing such interrelated processes and reconstructing how people experienced and responded to these different dynamics.28

And here, at least, mundane considerations may align nicely with theoretical ones. General education is changing, in ways that provide opportunities for history departments that respond creatively to today's new geographies. It is worth noting here that we cannot assume that general education requirements and our place within them are secure: pressures to provide undergraduate degrees more cheaply frequently translate into pressures to decrease general education requirements, and perhaps to focus them more narrowly on writing and basic numeracy. But on the other hand, a majority of college administrators told a survey conducted after the 2008 crash that general education was an increasingly important priority for them. Moreover, “global/world cultures” ranked near the top of the areas of knowledge they said that their requirements targeted; it was cited by almost twice as many of them as U.S. history, and by more than twice as many as languages.29 My point is not, of course, that we should simply provide the curriculum that others think we should provide, either for general education or otherwise, but these desires may represent opportunities for us to bolster our place in public culture (and the employment possibilities for our students) through offering appealing forms of “wider world history.” At the same time, there also seems to be a (perfectly reasonable) desire to rethink general education requirements so that they are less a matter of meeting a checklist of knowledge goals, as in a model defined by seeking “cultural literacy”; instead the focus is shifting toward using general ed to make sure that the students have certain skills that their majors may not emphasize, and that they can effectively move back and forth between the kinds of questions, evidence, and methods that their majors highlight and those highlighted by other fields.30

In short, we have important strengths to play to, and need not pander; but we are not putting our best foot forward if we present ourselves as simply providing an introduction to the history of one or more societies “out there,” or even to the history of all of us, without also advertising the ways in which we teach people to problematize and switch among spatial and temporal units of analysis. Bob Bain has reached a similar conclusion with respect to teacher training in particular: observing differences between teachers asked to design a U.S. history course and a world history course, he notes that problems in the latter case are caused much less by a lack of factual knowledge about certain areas than by confusion about what would be useful themes and a workable organizing framework.31

And as we consider this, the last major redesign of introductory history teaching—the creation of the Western Civilization course—should serve us as a cautionary tale. As Daniel Segal has pointed out, James Harvey Robinson and his students, who largely created Western Civ and the early textbooks for it, were very much committed to extending the period covered by “history” back as far as possible, at the same time that Robinson wanted history written and taught so as to explain “this morning's newspaper.” Equally convinced that all history was a seamless whole, and that from its beginnings the history (here opposed to “pre-history”) that mattered was essentially that of people attempting to solve fundamental intellectual, scientific, and cultural problems, Robinson et al. presented the story of “the West”—already detached from any real geography, as the course moved from Mesopotamia to the Mediterranean to northern Europe and across the Atlantic, with strong unifying themes compensating for the geographic incoherence—as a historical stream that any other place could join. Western Civ was thus presented as a world history in embryo, even if it had not so far included most humans—implicitly devaluing other histories.32 Something that calls itself “world history” and takes “humanity” as its subject, even as it must be highly selective in its examples, runs similar risks, even if some of our examples now come from the Indus or Yellow River valleys. And in this case, too, those risks can be amplified, rather than diminished, by a strong thematic unity—especially a thematic unity that treats some past people as having successfully faced problems of “sustainability” analogous to ones that seem to be reaching a decisive moment today.

Not that we can or should write large-scale history without big themes. Indeed, I would argue that we need more of them, and more in which—unlike population growth, greater energy use, and so on—historians are the obvious people to do the teaching. Very few other historical trends have as clear a directionality or can be as readily found in the very deep past as Bentley's “triple helix,” but many others are nonetheless relevant across vast spaces and time scales: the rise and decline of human retinues as status symbols, changing relationships between humans and territory (as property, as space to be ruled, and as places that define identity), monetization and commodification in their many forms, changing ideas of the sacred, and so on. The raggedness of these other large themes—the fact that they do not fit everywhere, or at all times—is a feature, not a bug. It helps call attention to the artificiality of our spatial and temporal units, what they reveal and conceal, and the need to use them provisionally. I propose these other themes not as replacements for the grand ecological and technological themes so central to current world history courses, but as supplements without which such courses can obscure, rather than teach, some of the central virtues of historical thinking in particular.

This explicit playing with scales and units is something we do all the time in our research, and many of us do it in the classroom as well; but it would probably help to do more of it, and to advertise it as one of the important operations we teach. Moreover, it would help not to have the structure of our curricula working against us, which it seems to me they do at the moment. To have introductory courses almost all take either a single nation, a “civilizational” region, or the whole world as a frame is to highlight the units that are most likely to seem “natural” to beginning students, and to faculty advisers from other disciplines, trying to help them figure out what can be learned from twelve different courses that fill an “international studies” or “humanities” requirement; we thus fail to signal that we teach the skills involved in choosing and maneuvering among different scales of inquiry. To have the “world” unit be a common one for surveys but then disappear at the advanced undergraduate level suggests that when we get really serious, we invariably look at nations or civilizations, much as we did decades ago. (It also suggests that it's OK to use a “world” unit for general education that we don't use in teaching more committed and sophisticated students.) To have very few courses at any level organized around non-traditional regions like the Atlantic and Indian Ocean worlds, which are producing some of the research we ourselves are most excited about, seems a huge missed opportunity; so does a curriculum structure in which once students advance from very broad introductory courses to narrower ones, they never return to larger spatial and temporal scales, even though many of them would find this stimulating, and more and more of us do exactly that at some point in our research careers. (Our current practice more closely resembles the progression through increasing specialization of some natural science curricula; but if what we are teaching is different, why should it be similarly structured?)

Self-consciously highlighting questions of scale and of boundaries has much to offer us. Again, we know this from many trends within our research: the strong interest in internationalizing U.S. history, the scholars who have made “territoriality” an explicit subject of research, the aforementioned work on non-traditional regions, on various sorts of networks, on commodities, and so on.33 When we also highlight the importance of temporal scale, we help clarify why historical narratives sometimes complement, but also sometimes contradict, the macro-narratives of other disciplines, and thus form an important part of the intellectual equipment that people need. Certain economic narratives, which assume an eventual return to “equilibrium” without specifying how long that will take, are particularly obvious cases requiring a historical narrative as complement—or as correction, if what might happen in the interim makes pressures toward the original equilibrium moot.34 But there are many others, including many that historians create when we frame large processes as the response of a single reified actor (nation, religion, civilization, or other) to a clearly defined “challenge”; and it is hard for even the best historians to avoid doing that when we create our large narratives.35

The point is not that we therefore give up on big pictures, but that we emphasize, as part of our place in the academic division of labor, that we provide ways to think about the interrelationship of different scales: temporal scales, spatial scales, and also levels of abstraction. We all know the perils of formulations such as “Germany wanted” or “middle-class voters feared,” as well as process-centered analogues like “urbanization required,” but we also know that we cannot do completely without them. It would seem to follow that we should have explicit discussions with our students about when such simplifications are acceptable and when they are not—conversations that we have with each other every time somebody says that his or her goal is to add “nuance” or “complexity,” and somebody else insists that this additional nuance meet the “So what?” test. Negotiating all these kinds of scales is an essential part of our research lives, which are thus centrally concerned with the organization of new and old knowledge, not just “the production of new knowledge”: an increasingly common description of what research should achieve, but a highly misleading one, and not just for history. Emphasizing reorganization of knowledge as central to our research can help us frame what we do (beyond digging individual facts out of archives) in ways that should be familiar to people all across the campus. Scientists, after all, understand that one can study a forest, a tree, or cells in a tree without any one of those levels making the others superfluous, and the whole field of complexity theory focuses on patterns that emerge on one scale and resist reduction to the dynamics of any more fundamental scale.36 Likewise, placing the movement among various scales that is essential to that effort front and center as a skill that we teach could help clarify what we offer to students beyond historical facts and the “communication” and “critical thinking” skills that many disciplines can claim.37

World history courses tend to place these issues in particularly sharp relief, as they require particularly rapid zooming in and out—if, that is, we treat that zooming as an opportunity to show our students how historians work, rather than as an embarrassing necessity that we hope the beginners in those classes won't notice. (Unfortunately, textbooks generally seem to take the latter approach, very rarely calling attention to shifts of scale.) To cite a personal example, when I used to teach a nineteenth- and twentieth-century world history survey at the University of California, Irvine (mostly to science majors), the first three-week unit featured a kaleidoscope of different kinds of analytic units: a lecture on rapid urbanization that went from London to Chicago to Buenos Aires to Shanghai; one on agrarian crisis in which growers of staples such as rice and wheat were juxtaposed to those who went through the boom and bust of industrial crops like rubber and palm oil, and temperate zones to tropics; and one on fossil fuels and energy use that went to huge time scales. This set the late-nineteenth- and early-twentieth-century scene, and was followed by a unit on world wars, revolution, and political polarization—a unit with much shorter time scales, more sustained stories about particular places, and shorter, more observable chains of influence linking one event to another. Only after teaching the course a few times did I realize what could be gained by making these shifts of approach part of the subject matter—but when I did, at least some students seemed to find the whole course considerably more meaningful. Scattering courses at various scales throughout our curricula, and teaching more courses based on non-civilizational regions, seem like other important ways to bring our teaching more in line with our research, inviting our students over the course of their undergraduate careers to think about issues that we often reserve for each other. It can also make it more evident how a set of history courses on different times and places can nonetheless be cumulative, even if not as obviously cumulative as a math major might be.

Last but not least, this kind of self-consciousness—in part a matter of being open with students about the quandary that challenges to both methodological nationalism and “civilizational” essentialism have left us with—frees us to continue teaching courses about the readily cognizable units that many allied disciplines have largely abandoned without falsely naturalizing them. It helps us bring into the classroom a number of the questions that we ourselves have found particularly exciting in recent years, about the making, unmaking, and contradiction of territorial institutions and identities at various levels. It links those theoretical concerns to the practical questions of the student who is not headed for an academic career: “What things will be different if I go work in India, and why? Is it worth understanding those differences, or are they bound to become ever less important over time, as some of my other courses suggest?” And it gives us ways to help our audiences, both inside and outside the academy, frame better questions about what it means—in terms of specific everyday practices and the extent and limits of specific people's networks—to say that we live in an increasingly global world.

None of us, I suspect, really doubts either that nations are historical artifacts or that they remain important. But there may be less unanimity about area studies “civilizations”—which never had the institutional power of nation-states (except, to a very limited extent, in today's EU) or attracted us all that much as research (as opposed to pedagogical) units. So let me close by suggesting that at this level, too, an effective response to so-called “globalization” is not to simply drop units once we see that they are not bounded wholes, but to highlight how they are made and remade, and what they are and are not good for—both as analytical units for us and as often-naturalized units used to mobilize resources for “real world” projects.

For this purpose, let me turn to East Asia, since it is the region I know best—although I suspect that a similar argument could be made about some others, too. Let me also emphasize that everything I am about to say could be presented to even beginning undergraduates as part of a survey that takes apart commonsense notions of why East Asia is a unit and builds in their place a historically based understanding of how regional particularities can emerge, disappear, reappear, and matter to daily life (albeit unevenly) across the region at specific moments, including ours.

Comparing 1980, when I started graduate school, to today, it is striking how much we overlooked in treating East Asia as a single region back then. It is even more striking that, despite thirty years of globalization that supposedly transcends regions (and critiques of Orientalism, which some believe showed that area studies had always been a bad idea), East Asia probably makes more sense as a teaching and research unit today.38 To an extent unimaginable during the Cold War, the region is now crisscrossed by dense networks connecting all its constituent countries: not only in trade and investment, where the flows are now significantly larger than those between these countries and either the U.S. or Europe, but in exchanges of popular culture, movement of students (where they also now eclipse movements between East Asia and the West, at least by some measures), and so on.39

It is not just that the region has become more densely interconnected than it could be when politics made the Taiwan Straits, the Sea of Japan, and the Bohai Gulf (separating South Korea from China) impenetrable barriers. It is also that it is a different kind of region from the kind we were once encouraged to imagine—one that is clearly a historical product, rather than a more or less transhistorical fact. The old story was that of a region characterized by a shared high culture radiating from two centers: a Chinese one, later supplemented or even replaced by a Japanese one; and by a socioeconomic basis, irrigated rice. But it was never clear how far down the social scale or across the region “Confucianism” stretched in practice, though it clearly stretched a long way as a vague talisman of cultural sophistication; and it is completely clear that even most of China was not growing irrigated rice through most of history. There were, to be sure, many significant shared references, but they never integrated the region the way that the current webs of transnational ties do.

East Asia may be somewhat atypical in the degree to which it is becoming more of a region in an era that is supposedly making regions irrelevant, but it is certainly not completely unique. And when we look at the making of today's East Asia, we do not find one uninterrupted story, either of ongoing integration or the maintenance of a shared ancient heritage. Instead we find multiple, layered, region-making processes, with the crucial links often emanating from various margins rather than the supposed centers of power and high culture.

Most recently, we see the effects of several decades in which there were particularly strong buildups of technological and financial resources in some of the region's smaller and less fully sovereign polities (Hong Kong, Taiwan, South Korea), which were thereby well positioned to play important roles in the post-Mao remaking of China. Beneath those networks we find deeper layers of regionally circulating religious, philosophical, medical, and other texts, and strong diasporic networks (e.g., of Chinese merchants); the latter, ironically, were kept largely regional partly by migration restrictions and racialized nation-building projects that impinged on East Asia from outside.

We also find the spread, both across space and across the social scale, of the old cultural traditions we once took to be the deepest, bedrock layer of regional formation; but when we look closely, we often see those traditions as at least partly retrospective creations inflected by political choices. Consider, for instance, Buddhism. When the Ming fell in 1644, what we now call “Tibetan Buddhism” was not recognized within China proper as one branch of a religion that also included “Chinese Buddhism”: Chinese practices were “the teachings of Buddha,” and Tibetan/Mongol ones were “the teachings of lamas,” unfit for civilized people. Later, the Qing imperial household (invaders from Manchuria) began patronizing Tibetan clerics and temples for political reasons (and perhaps for personal enlightenment) and in the process brought many Inner Asian monks to Beijing. Thus, in the long view, they prepared the ground for the integration of these “Buddhisms,” which would in turn help the Chinese/Inner Asian empire they created survive as a national state once they were gone—though that was obviously not their intention. Nor was it a task they completed: it was twentieth-century clergy, supposedly secular governments, and lay activists (many of them influenced by the Christian model of what a world religion should look like, and seeking alternatives to Japanese versions of a shared “Asian” identity) who produced what now appears as the longstanding “background condition” of a “Buddhism” shared across vast spaces and big differences in daily practice.40 So this cultural region is not a found object, but rather something always being re-created through interactions—and so, perhaps, not categorically different from non-civilizational regions like the Indian Ocean or Atlantic World after all.

One would hope that a student who had been through an assortment of courses that were self-conscious about spatial units in this way would have—in addition to a lot of content knowledge, and general skills of reading, writing, and critical thinking—some sense of the tools with which to approach questions about how social space changes over time, how that can matter to the people within a region, and how the spatial units that we use to frame questions matter with respect to what processes we do and don't see. These are important tools for students to hone; they provide leverage on vital intellectual questions about structure, agency, and culture, and about the possibilities for empathetic understanding of experiences remote from our own. But they are also highly practical skills that give students leverage on questions they might well confront in non-academic careers: How do we decide whether the “Pacific Rim” is a meaningful unit (and for what purposes) or just a name for some shipping lanes and the fond hopes of certain chambers of commerce? Does it matter whether one launches a given project from New York, Los Angeles, Singapore, or Shanghai? Are the particular networks or commonalities on which one bases such decisions robust enough to bounce back quickly from a crisis on the Korean Peninsula or over the Diaoyu/Senkaku Islands? Skills for thinking about such issues historically should matter to planners for everything from corporate supply chains to NGOs to government agencies, universities, and even K-12 systems guessing about future demand for particular languages. This is not, of course, the only way history can matter, but it is one that should find ready takers, while actually moving our pedagogy closer to what we often, all too revealingly, call “our own work.”

I said that this would not be a manifesto, and I will not end by saying “Let us go do X.” But I hope that I have given you food for thought. Our discipline has a rather unusual relationship to time, space, and place, and we live in a moment when many people wonder which pasts, if any, one must engage with in order to navigate something called “today's world.” The older units with which we are identified can seem to have perfectly obvious meanings; this is both an asset and a problem for us. We are already rethinking those units, in many interesting ways, but this rethinking is not as evident as it might be in the face we present to the wider world, either in our course listings or in the explanations we give of what we as a field do. Learning to choose and maneuver among a huge range of spatial and temporal scales, and explain what is revealed and obscured as one does so, has long been a central part of our craft, and one that people who must locate themselves and their options in both time and space will always have need of—maybe more than ever in a world that tells them that they cannot be insulated from the wider world while offering up increasingly decontextualized “information” and some questionable universals as dominant kinds of knowledge. So while challenges to methodological nationalism threaten formerly secure franchises from which we have gained a great deal, they also highlight the need for things that we are good at, and have gone quite far in discussing amongst ourselves. We can, I think, help both ourselves and others by doing our reframing for a less national age more self-consciously, and more openly.

Kenneth Pomeranz served as president of the American Historical Association in 2013. He is University Professor in History and the College at the University of Chicago, and previously taught at the University of California, Irvine. He is the author of The Making of a Hinterland: State, Society and Economy in Inland North China, 1853–1937 (University of California Press, 1937), and The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton University Press, 2000), among other works. His current projects include, among others, a book called Why Is China So Big? and a co-authored world history textbook.


A draft version of this address was presented as a paper at the East Asia Transregional Histories Workshop at the University of Chicago. I thank the organizers, the participants, and the two discussants—Dipesh Chakrabarty and Paul Cheney—for their comments on that occasion. Earlier drafts benefited from comments by Maureen Graves, Robert Moeller, Daniel Segal, and Julia Thomas. The surveys I discuss were carried out by Brendan Mackie, Patricia Kahle, and Pranav Jain, and the design benefited from data supplied by Robert Townsend and Elizabeth Townsend.


  • 1Luke Clossey and Nicholas Guyatt, “It's a Small World after All: The Wider World in Historians' Peripheral Vision,” Perspectives on History 51, no. 5 (May 2013),; “AHA Roundtable: ‘It's a Small World after All,’” Perspectives on History Online, Summer 2013,, with contributions by Mary Elizabeth Berry, Anne Gerritsen, Teofilo Ruiz, Kenneth Mills, Kenneth Pomeranz, and Luke Clossey and Nicholas Guyatt.

  • 2Roy Rosenzweig and David Thelen, The Presence of the Past: Popular Uses of History in American Life (New York, 1998), 38, 50, 55–56, 116–118, 128, 133–134.

  • 3Those of us who work in big departments should remember in this connection that more than half of college history courses are taught in departments with fewer than twenty full-time faculty or the equivalent thereof (e.g., eighteen full-timers and four half-timers). Data compiled by Robert Townsend from the AHA's 2012 Directory of History Departments and Organizations; e-mail message from Liz Townsend, July 5, 2013.

  • 4Google Ngram,; New York Times online archive search.

  • 5Ian Morris, Why the West Rules—for Now: The Patterns of History, and What They Reveal about the Future (New York, 2010); David Christian, Maps of Time: An Introduction to Big History (Berkeley, Calif., 2004).

  • 6Jerry H. Bentley, “World History and Grand Narrative,” in Benedikt Stuchtey and Eckhardt Fuchs, eds., Writing World History, 1800–2000 (London, 2003), 47–65, here 63–65. There is also considerable resemblance to the strategy employed in J. R. McNeill and William H. McNeill, The Human Web: A Bird's-Eye View of World History (New York, 2003). Interestingly, Bentley's own research contributions to world history placed more emphasis on a narrative treatment of cross-cultural encounters than on demography, technology, the environment, or a structural view of the growth of enduring long-distance networks. See Jerry Bentley, Old World Encounters: Cross-Cultural Contacts and Exchanges in Pre-Modern Times (New York, 1993).

  • 7Kenneth Pomeranz, “Teleology, Discontinuity and World History: Periodization and Some Creation Myths of Modernity,” Asian Review of World Histories 1, no. 2 (July 2013): 189–226, especially 197–201, 206–209, 213–223. I develop some of these themes in greater detail in “Environmental History and World History: Parallels, Intersections, and Tensions,” in Prasenjit Duara, Viren Murthy, and Andrew Sartori, eds., A Companion to Global Historical Thought (Chichester, UK, forthcoming, 2014), 351–368; and in “How Big Should Historians Think? A Review Essay on Why the West Rules—for Now: The Patterns of History, and What They Reveal about the Future by Ian Morris,” Cliodynamics: The Journal of Theoretical and Mathematical History 2, no. 2 (2011): 304–329.

  • 8For other problems with treating humanity as a whole as the subject of history, see Dipesh Chakrabarty, “The Climate of History: Four Theses,” Critical Inquiry 35, no. 2 (Winter 2009): 197–222.

  • 9See, for instance, the discussion of the relationship between temporal scale and agency in Sebouh David Aslanian, Joyce E. Chaplin, Ann McGrath, and Kristin Mann, “AHR Conversation: How Size Matters—The Question of Scale in History,” American Historical Review 118, no. 5 (December 2013): 1431–1472, especially 1444–1449, 1453–1455; and Pomeranz, “Teleology, Discontinuity and World History,” 191–195, 215–220.

  • 10David M. Potter, “The Historian's Use of Nationalism and Vice Versa,” American Historical Review 67, no. 4 (July 1962): 924–950.

  • 11Gordon Wright, “History as a Moral Science,” American Historical Review 81, no. 1 (February 1976): 1–11, here 2.

  • 12Joseph C. Miller, “History and Africa/Africa and History,” American Historical Review 104, no. 1 (February 1999): 1–32.

  • 13Those that did included books on the Roman and Japanese Empires—single polities but hardly single nations. Without them, the number drops below 30 percent.

  • 14At the University of California, Chicago, Columbia, Harvard, and Yale—picked partly for the ease with which such data could be obtained from their websites—60 percent of history titles deal with national or subnational topics.

  • 15In that period, the topics of at least 57 percent of the books selected were contained within a single modern nation, while another 16 percent were about Europe or Europe and North America. By contrast, only 24 percent could be said, even by a generous definition, to cross national lines without claiming to represent a larger “civilizational” unit, and none constructed regions that crossed area studies lines, such as the Atlantic world. Moreover, only 4 percent of these books dealt with any part of Asia, and none focused on sub-Saharan Africa.

  • 16By representative, I mean that they were distributed among schools granting associate's, bachelor's, master's, and doctoral degrees in the same proportion as our 733 member departments. This does, unfortunately, mean that community colleges, whose departments rarely join the AHA, are grossly underweighted, but I suspect that including more of them would only reinforce my findings.

  • 17See Leigh Estabrook with Bijan Warner, “The Book as the Gold Standard for Tenure and Promotion in the Humanistic Disciplines,” Andrew W. Mellon Foundation Research Report (Champaign, Ill., 2003), based on a 2003 study of Committee on Institutional Cooperation universities (not counting the University of Chicago) funded by the Mellon Foundation, Anthropology is also the only discipline that gets more International Dissertation Research Fellowships (from the Social Science Research Council) and Fulbright-Hays funding—the largest sources of funds for dissertation research abroad—than history does. Rina Agarwala and Emmanuel Teitelbaum, “Trends in Funding for Dissertation Field Research: Why Do Political Science and Sociology Students Win So Few Awards?,” PS : Political Science & Politics 43, no. 2 (April 2010): 283–293, here 284–285.

  • 18Membership data from AAA and ASA websites; AAS data by field provided courtesy of Irene Dolozor, e-mail of August 19, 2013. MESA membership data courtesy of Sara L. Palmer, e-mail of September 26, 2013. The Latin American Studies Association also provided membership data (e-mail of Milagros Pereyra-Rojas, September 26, 2013), but it is harder to interpret. In part this is because the association has a very large share of members based outside the U.S.; in part it is because its membership categories have changed dramatically. In 1974 (the first year for which data is available), members were recorded in only six disciplines, but by 2012 it was thirty, leading to large declines in almost all of the originally listed categories.

  • 19In sociology, 11.8 percent of classes were focused on a single nation, 1.8 percent on an area studies unit, and 8.5 percent on the world, with almost 80 percent not geographically defined. (The ASA is the same size as we are, while one-seventh as many sociologists are AAS members, and barely one-tenth as many belong to MESA.) There were a small number of schools with a combined sociology/anthropology department; results for those schools were not materially different.

  • 20The overall figures were as follows: national units, 30.0 percent; traditional regions, 9.4 percent; unconventional regions, 2.8 percent; world, 14.2 percent; theory/method, 26.4 percent; non-geographic, 16.3 percent; other/unclassifiable, 0.6 percent. For the ten most prestigious research universities in the sample (Cornell, Emory, Georgetown, Harvard, Johns Hopkins, Michigan–Ann Arbor, Northwestern, the University of Pennsylvania, Princeton, and Rutgers–New Brunswick), the numbers were as follows: national, 24.0 percent; traditional regions, 12.9 percent; unconventional regions, 2.9 percent; world, 15.6 percent; theory/method, 27.8 percent; non-geographic, 16.9 percent; other/unclassifiable, 0.2 percent. At my home institution (not included in the sample), only 12.4 percent of political science courses explicitly targeted a nation, 9.4 percent a region, and 6.5 percent the world.

  • 21As one further example, consider the “sections” within professional organizations. The AAA has forty, of which at most seven are associated with a “place” of any sort. The APSA has forty-four sections, of which only three (Canada, Europe, and Africa) are based on any kind of geographic or cultural unit. The ASA has fifty-two, of which two might be said to refer to a place, but only very loosely: Latino/a (which presumably covers people of that heritage wherever they are) and “Asia and Asian America.” The AHA has no sections, but a large number of our affiliated societies have a national or regional basis, while many others have a temporal one (absent in these other fields).

  • 22See, for instance, George E. Marcus, “Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography,” Annual Review of Anthropology 24 (1995): 95–117; Mark-Anthony Falzon, “Introduction,” in Falzon, ed., Multi-Sited Ethnography: Theory, Praxis, and Locality in Contemporary Research (Burlington, Vt., 2009), 1–23; Matei Candea, “Arbitrary Locations: In Defense of the Bounded Field-Site,” ibid., 25–45; Joanna Cook, James Laidlaw, and Jonathan Mair, “What If There Is No Elephant? Towards a Conception of an Un-Sited Field,” ibid., 47–72; Cindy Horst, “Expanding Sites: The Question of ‘Depth’ Explored,” ibid., 119–133; Karen Isaksen Leonard, “Changing Places: The Advantages of Multi-Sited Ethnography,” ibid., 165–180; and George E. Marcus, “Multi-Sited Ethnography: Notes and Queries,” ibid., 181–196.

  • 23There does, however, seem to be less concern among anthropologists about the specifically linguistic aspects of these challenges. Perhaps this is because a few languages are so widespread in today's world (especially among those who cross borders frequently), and perhaps also because living informants who move do not lose the capacity to talk to ethnographers in their old languages, while historians, who must often track migrants through what those around them record, need to read the languages of each place those people move through.

  • 24Timothy Mitchell, “Deterritorialization and the Crisis of Social Science,” in Ali Mirsepassi, Amrita Basu, and Frederick Weaver, eds., Localizing Knowledge in a Globalizing World: Recasting the Area Studies Debate (Syracuse, N.Y., 2003), 148–170, here 154–163.

  • 25In the case of economics, one can tell two rather different stories, which lead to the same conclusion. Mitchell's story, focusing on the 1970s, points to declining faith in Keynesian models that claimed to explain whole economies as more or less predictable systems: a tale in which the recognition of failure is tied to a loss of unifying faith and fragmentation of the field into people pursuing many different topics and united only by the faith in maximization under constraints and models rendered in mathematical form. Alternatively, one could focus on some of the new models (such as real business cycle theory) that treat old topics of economic debate as solved problems no longer needing discussion, thus freeing economists to go use their (vindicated) tools elsewhere. While these stories are fundamentally different, and may describe different parts of the profession, the effect is the same.

  • 26Robert B. Bain, “Challenges of Teaching and Learning World History,” in Douglas Northrop, ed., A Companion to World History (Malden, Mass., 2012), 111–127, here 111. In fact, as Bain notes, many “world history” courses are a series of strung-together area studies units, while others are essentially a Western Civ narrative, with brief excursions into other areas (usually in the distant past) added on without having much effect on the main story.

  • 27David A. Hollinger, “The Historian's Use of the United States and Vice Versa,” in Thomas Bender, ed., Rethinking American History in a Global Age (Berkeley, Calif., 2002), 381–395, here 384.

  • 28I discuss several examples, concerning both spatial and temporal scales, in “Teleology, Discontinuity and World History”; see especially 193–195, 206–209, 215–223. An example I have found particularly effective for teaching purposes and general audiences is the effects of mechanized transportation on the use of horses. Most people take for granted the long-term outcome, namely that horse-drawn transport has nearly disappeared from most of the world. When people learn that the initial effect was quite the opposite—because building railways caused a huge increase in the quantity of people and goods being transported, and they all had to get to and from the station, the number of horses in major urban centers increased much faster than the human population during the latter half of the nineteenth century and into the early twentieth—they are often surprised, but still note that “in the long run” the eclipse of horse-drawn transit was inevitable. That may even be true, but nobody saw it coming until quite late: even in 1900, almost fifty years after the invention of the internal combustion engine, experts were still nervously projecting the sanitary and other problems that would result from several decades more of growth in the urban equine population, rather than foreseeing that the automobile would render these worries moot. For a brief account, see Eric Morris, “From Horse Power to Horsepower,” Access 30 (Spring 2007): 1–9,

  • 29Hart Research Associates, “Trends and Emerging Practices in General Education: Based on a Survey among Members of the Association of American Colleges and Universities” (May 2009),, 4–5. Interestingly, a Hart survey of business executives also ranked “global issues” among the top five areas needing more emphasis in higher education. There is something odd about the very limited support for language requirements in this context, but it presumably reflects a feeling that English has become enough of a lingua franca that one can function in today's world without other languages. The argument against this seems to me part of the general argument about the continued relevance of cultural particularities that recurs throughout this address.

  • 30American Association of Colleges and Universities conference overview, “General Education and Assessment: A Sea Change in Learning,” February 28–March 2, 2013,; see also the Lumina Foundation's description of the Degree Qualifications Profile at

  • 31Bain, “Challenges of Teaching and Learning World History,” 113–117, 123.

  • 32Daniel A. Segal, “‘Western Civ’ and the Staging of History in American Higher Education,” American Historical Review 105, no. 3 (June 2000): 770–805, here 776, 778–780, 785; see also Adam McKeown, “What Are the Units of World History?,” in Northrop, A Companion to World History, 79–93, here 81.

  • 33See, for instance, Bender, Rethinking American History in a Global Age; Thomas Bender, A Nation among Nations: America's Place in World History (New York, 2006); Carl Guarneri, America in the World: United States History in Global Context (New York, 2007). On “territoriality,” see, e.g., Charles Maier, “Consigning the Twentieth Century to History: Alternative Narratives for the Modern Era,” American Historical Review 105, no. 3 (June 2000): 807–831.

  • 34Donald N. McCloskey, “The Economics of Choice: Neoclassical Supply and Demand,” in Thomas G. Rawski et al., Economics and the Historian (Berkeley, Calif., 1996), 122–158, here 128 fn. 1, points to the fact that positing an equilibrium-seeking process does not tell us how quickly or slowly that dynamic will work, but without noting that it can make economic analysis alone misleading.

  • 35I give some examples from a work I greatly admire—Alfred Crosby's Ecological Imperialism—in “Teleology, Discontinuity and World History,” 206–208, 217–219.

  • 36I found the forest/tree/cell analogy, used for slightly different purposes, in William H. McNeill, “A Defense of World History: The Prothero Lecture,” Transactions of the Royal Historical Society, 5th ser., 32 (1982): 75–89. For an introduction to complexity theory, see Melanie Mitchell, Complexity: A Guided Tour (Oxford, 2009).

  • 37There is a small but interesting psychological literature suggesting that beginning students have a strong tendency to explain everything in terms of individual agency: e.g., explaining Columbus's voyages in terms of his personal ambition or that of Ferdinand and Isabella, rather than exploring the position of Castile in existing political and commercial competition. When they are urged away from that, their first recourse is often to treat political units as individual agents (e.g., “Britain felt … ”). It would not be a bad first description of what we teach, methodologically speaking, to say that we try to make students able to consider the play of structure, culture, and agency in more sophisticated ways than that—which involves, among other things, getting them to think about what units are “real,” in the sense that they have either agency or an inherent logic that limits the possibilities available at a given time. (It also probably involves making them less eager to resort to mono-causal explanations of any sort.) For a discussion of the psychology of history learning that emphasizes this movement from individual causation to analysis of structures, see Ola Halldén, “Conceptual Change and the Learning of History,” International Journal of Educational Research 27, no. 3 (1997): 201–210.

  • 38It is convenient for these purposes that Edward Said's much-cited Orientalism was published in 1979—though Said himself was by no means calling for a total abandonment of area studies.

  • 39For comparisons of economic flows within East Asia and between East Asia and the West, see, e.g., figure 6, “Intra-regional Trade of Major Regions (1988–2007),” in Douglas H. Brooks and Changchun Hua, “Asian Trade and Global Linkages,” ADB Institute Working Paper no. 122 (December 2008),, 10; United States Census Bureau, “Trade in Goods with Korea, South,”; European Commission, Directorate-General for Trade, “European Union, Trade in Goods with South Korea,”, 2; Scott Snyder and See-Won Byun, “China-ROK Trade Disputes and Implications for Managing Security Relations,” Korean Economic Institute Academic Paper Series 5, no. 8 (September 2010), See also “S. Korea, China, to Hold New Round of FTA Talks Next Week,” Xinhua, January 2, 2014,; and Aaron Back, Toko Sekiguchi, and Yuka Hayashi, “China, Japan, South Korea Agree to Trade Talks,” Wall Street Journal, May 13, 2012, The former gives a figure for China-South Korea trade which is more than double the figure for the U.S. and South Korea, and more than triple that for the EU and South Korea; the latter specifically notes ways in which the proposed China/South Korea free trade area deal would serve as an alternative to the U.S.-proposed Trans-Pacific Partnership. For figures relevant to movements of students see, e.g., China State Administration of Foreign Experts Affairs, “Away from Home but Never Alone,” April 12, 2007,; Eunkyung Seo and Heesu Lee, “China Beats U.S. for Korean Students Seeing Career Ticket,” Bloomberg News, September 4, 2013, See also Hélène Le Bail, “The New Chinese Immigration to Japan: Between Mobility and Integration,” China Perspectives 61 (September–October 2005): 2–15.

  • 40Gray Tuttle, Tibetan Buddhists in the Making of Modern China (New York, 2005).