Historians’ engagement with the social sciences, which was promising and multi-dimensional into the 1970s, shriveled soon thereafter. The factors that seemed to justify this turning away appear today to be much less compelling. Times have changed, and both historians and social scientists may now be ready to re-engage with an interrupted project to intensify interdisciplinary dialogue, improve the methods of historical explanation, and construct a more historical body of social theory.

In the mid-1970s, a broad range of scholars, spanning the ideological spectrum, was actively engaged in what has been called the New History. Some of the participants in this broad movement come under discussion below, but for now it is sufficient to mention the French Annales School, the American New Economic History, the British Marxist Social History, and Historical Sociology in several forms, including modernization theorists, comparative revolutions theorists, and world-systems theorists. The adherents to these different historical orientations disagreed with each other about most substantive matters. Their single point of agreement was a shared feindbegrif—the rejection of narrative history. It inspired Burke to introduce the final volume of the New Cambridge Modern History, which appeared in 1979, with the following declaration: “In the twentieth century we have seen a break with traditional narrative history, which, like the break with the traditional novel or with representational art or with classical music, is one of the important cultural discontinuities of our time.”1

Obviously, no single influence can claim full credit for this historiographical discontinuity, but the most profound factor must be the spread of an uncomfortable feeling that the narrative form greatly restricts the types of possible historical questions and feasible modes of explanation. Narrative history has attached to it, like a ball and chain, the discrete, short-term historical event—l’histoire événementielle. If, by the time of Burke’s writing, several decades of Annales School preaching had done anything, it had undermined the notion that this concept should be the foundation stone of historical explanation.

In the opinion of Bob Dylan, “You don’t need a weatherman to know which way the wind blows.” But, history in those years did have a weatherman, and in the very year in which Burke associated the New History with the future and the avant-garde, the weatherman spoke. In “The Revival of Narrative: Reflections on a New Old History,” Stone (then Professor of History at Princeton, Director of the Shelby Cullum Davis Center for Historical Studies, and member of the editorial board of Past Present) sensed that the vast majority of historians, certainly in Britain and the United States, had no heart for the new scholarship. They once had felt an obligation to pay allegiance to it, but they were now withdrawing that allegiance.2

Stone identified two broad reasons for this turning away. First, the political sentiments that had led many historians to pay at least lip service to material and collective historical analysis were now eroding. Stone pointed to their disenchantment with economic determinism and an associated decline in ideological attachments, especially to Marxism. Such historians were now searching for another basis on which to advance their political commitments. Second, Stone felt that social-scientific historians, quantitative historians especially, had not fulfilled their expansive claims—had “failed to deliver the goods”—and had alienated readers with their often-inaccessible techniques and methodologies. Historians, he claimed, were disillusioned and perhaps also rather fearful.

Stone foresaw historians refocusing their research increasingly on individual agency and turning away from analytical in favor of descriptive modes of explanation. “If I am right,” he concluded, “the movement to narrative…marks the end of an era: the end of the attempt to produce a coherent and scientific explanation of change in the past.” This statement, a veritable confession, was stunning at the time. Stone himself had been a prominent practitioner of a sort of social-scientific history. He had tried to produce coherent explanations of the English Civil War and the emergence of modern family life that relied on social-scientific analysis. His prediction of where history as a discipline was headed appeared to reject his own life’s work, yet he did not seem to regret it at all. Indeed, he seemed to take a good deal of pleasure in prophesying the collapse of the temple—and his prophecy came to pass. Indeed, Stone’s prescience was not limited to “Anglo-Saxon” scholarship; the shift toward micro-history and a narrowing of the venerable Annaliste concept of mentalité was also underway in France. By the time of the linguistic or cultural turn of the early 1980s, the project to integrate history more closely with the social sciences had become bizarre to most historians, who could only utter the world “science” in an ironic tone of voice.3

The Problem of Narrative History

“Narrative” refers to the organization of material in a chronological sequential order to form a single coherent story (albeit with possible sub-plots). One might express surprise that history could take any other form, since chronology is so closely associated with it, and since historians have always told stories. Indeed, the ancients regarded history as a branch of rhetoric rather than philosophy. Moreover, it would be an exaggeration to claim that even social-scientific historians ever completely abandoned narrative. But they ceased to rely on it to give meaning to their work: They questioned its epistemological validity and its cognitive value, and they argued that it prevented historians from asking the right questions. Narrative relies on description more than analysis, focusing on man, or individual actors, more than circumstances, and emphasizing particular more than collective experience.

The defenders of narrative usually see it as an accurate reflection of how people live and experience life. Thus, Ricoeur claimed that narrative is the literary genre that provides the best analogy for the actuality of life. Life and narrative both take place chronologically. Likewise, Berlin defined historical explanation as, mostly, “the arrangement of the discovered facts in patterns which satisfy us because they accord with life as we know it and can imagine it.” “Knowing” is the key word in this quotation: It is not explanation, in the sense of demonstrating causation through induction or deduction; it denotes immediate, unmediated, knowledge (verstehen). It allowed Collingwood, the influential philosopher of history, to assert, “When the historian knows what happened, he already knows why it happened.” In this same spirit, White, a very different sort of philosopher of history, argued that the function of history is to produce stories that will disclose (not explain or analyze, but reveal) the condition of the present time.4

These theorists of narrative make claims that associate the writing of history with the writing of fiction. They view it as the exercise of creative imagination, via observation, empathy, intuition, and unmediated understanding. “The past is a foreign country,” it became fashionable to say, implying that one could never “explain” it in anything like a scientific way. To think otherwise was naïve.5

The critics of narrative history objected to its very “naturalness”—the seductive power that carries readers along, offering the intellectual satisfaction of “things fall[ing] into place.” We might concede these points in narrative’s favor and yet object that good stories may well be untrue. Chronology is not causation; description is not explanation—understood in the sense not simply of elucidation or making clear but of determining cause.6

Surely, one might counter, a carefully constructed narrative may be essential to explain events even in the causal sense? It may be essential or necessary, but it will rarely be sufficient. After all, how do we decide which narrative to use of the many that might plausibly be constructed? The discipline in the traditional discipline of history is the obligation to seek concurrence (or consilience) among all available evidence in the construction of a narrative. Yet, even on this view, the criteria for concurrence must reside outside the framework of the narrative itself.7

Megill, who offers a relatively even-handed view of the pros and cons of narrative, notes, “A characteristic of secular, modernist academic culture is its commitment to metaphors of verticality: surface reality vs. deeper, hidden reality.” Indeed, things directly observable are not actual reality at all. The task of inquiry—scientific inquiry—is to get down to what is fundamental and usually hidden from direct view. Thus, explanation is deeper than description, and the normal practices of narrative historians are easily dismissed as epistemologically naïve. Such had been the criticism of traditional history as written in the 1950s, 1960s, and 1970s, and it bit hard. On many fronts, and not only in economic history, it supported the development of a New History that would gain explanatory power through alliance with the social sciences and cease to rely on narrative structure alone to impart meaning. Such was the rationale behind the founding of this journal in 1970. When, in 1972, the Social Science History Association was founded in the United States (followed by a biennial European Social Science History Conference), it organized into at least a dozen distinct thematic networks. Historians with a broad range of research interests were eager to embrace the social sciences.8

Social-science history failed in achieving all of its aspirations. Stone refers to disappointment in its achievements as one of the catalysts of the revival of narrative. But he may have been disingenuous on this point. What he and many historians most disliked about the New History—why his prediction was so laced with schadenfreude—were its technical inaccessibility and the scale of its ambition. Historical research in the new mode required team efforts, large databases, computers (which then meant punch cards, main-frame computers, and code writing), as well as recourse to methods and theories that are not intuitively available to the uninitiated. In short, it had to assume the organizational form of a “science.” For Stone, it was a big step too far. He held fast to the conviction that even the most ambitious historians should communicate their findings to an educated lay reader directly, without mediation.

This conviction, it appears, is deeply ingrained in historians. Ankersmit, an influential theorist of historical epistemology, places a heavy emphasis on the distinction between historical research and historical writing. The first is necessary, but neither philosophically interesting nor challenging. It is a relatively straightforward affair: The historian is simply telling it as it was. The writing of history, however, is everything. Indeed, historians appear (should appear) to their readers only as writers. Their role as researchers is (should be) sheltered from direct view. To be sure, this historical writing must respect the findings of research, but historians have broad latitude in arranging and “telling” historical stories. “The creation of an historical account,” Ankersmit concluded, “is a question of Ahnen, intuition, and such like.”9

From this perspective, social science might be acceptable as an “add-on,” a way to orient thinking. Historians might borrow its vocabulary or its conceptual categories, but only opportunistically and for limited purposes. Indeed, this limited appropriation remains as far as Burke is willing to go in History and Social Theory. In his view, empiricists and theorists are not two groups with overlapping interests that might collaborate in a joint enterprise, but two ends of a spectrum. Thus, Burke pronounces as obvious in 2005 what Stone insisted upon in 1979: History is and should always remain an artisanal craft. A dose of social science might add to the beauty of the artisan’s product, but an embrace of social science would corrupt its purpose.10

The responsibility for this state of affairs—the limited transformative effect of the New History–cannot all be laid at the feet of historians. It takes two to tango; the theories offered by the social sciences had, and have, significant deficiencies for historical explanation. Furthermore, just when historians were turning to the social sciences, the social sciences exhibited little serious interest in history. Most social-science disciplines were consolidating the structural theories that were prevailing at the time, anxiously burnishing their own scientific credentials. Like a suitor spurned, historians turned to a new love interest that denied the beauty and especially the power of the old one. Stone pronounced this turn to be a return, but historians should know that no return can ever be a full reversion to the status quo ante.

Postmodern Narratives

The new historical narrative was different from traditional narrative given that many returners found themselves strangely aligned with postmodernists who harbored, in Lyotard’s words, an “incredulity toward master narratives of all types.” In this same vein, Foucault regarded the very continuity of narrative as suspect, since it unavoidably privileged the central subject of a narrative. On the face of it, these new philosophical currents raised a formidable barrier to any contemplation of a narrative revival. Yet the desire to turn away from the New History overrode the intellectual incoherence of the new alliance. Historians persuaded themselves that a shift in scale and in theme could allow them to evade the evils of earlier narrative history.11

Ironically, historians made this move by embracing a new historiographical initiative that intended neither a rejection of theory nor a revival of narrative. Italian microstoria, as developed by Ginzburg and Levi, among others, deployed an intensive focus on micro-level sources and on “exceptional normal” individuals (rather than “ordinary people”) with the aim of testing the validity of such macro-scale explanatory paradigms as Marxism, modernization theory, and functionalism. These micro-histories favored synchronic relationships rather than narration to uncover the complexities of situations that encouraged theoretical reflections of broad significance. Micro-history was critical, to be sure, but it was engaged with social-scientific theory.12

These brilliant works had an immediate international impact. But as micro-history migrated abroad, it quickly came to breathe a different spirit. In Trivellato’s words, “In the Anglophone world, microhistory became first of all a tool to shed light on marginal figures who entice everyone’s curiosity and mobilize readers’ empathy, sometimes to free scholars from evidentiary standards perceived to be too confining, and always to render academic writing accessible to a broader readership.” A micro-history shorn of its theoretical engagement and focused on archival virtuosity and empathetic storytelling became the preferred vehicle to return to narrative, presumably because these new narratives were no longer “master” narratives. What Walt Disney did to Pinocchio, America’s returning historians did to microstoria. Micro-history became part, ultimately a minor part, of a larger “return” to a different sort of narrative than Stone could have suspected—a petite narrative, immersed in a sauce of thick description and shaped by a postmodern philosophical posture that elevated subjective experience and the identity of the historical subject.13

By no means should we dismiss postmodern scholarship altogether. Although its moment has past, it did much to sharpen the epistemological consciousness of historians. Especially relevant to the theme herein was its critique of the social-scientific epistemologies with which the New History had hoped to engage. This critique, by itself, might have advanced the agenda of those aiming for a more historical social science, since there was certainly much to criticize in the prevailing models of the social sciences, especially their claims to universality.14

But another feature of postmodernism raised a formidable barrier to engagement in such a debate at the same time that it powerfully encouraged historical fragmentation. The lack of a real spur to empirical refinement weakened the already fragile belief that historical knowledge is cumulative and self-corrective. A discipline without such a sense offers its practitioners no strong reason to pay attention to each other, let alone to be attentive to the existing stock of empirical findings. At best, small “communities of discourse” survived in this process of intellectual involution, and the resulting fragmentation is now a universal lamentation of the historical profession. We have hundreds of histories, but no history.15

Yet another contribution of the postmodern inflection of the return to narrative was its “subaltern” preference—its reliance on the authority of empirical experience and its dedication to the “restoration of lost voices,” as part of the construction of historical identities for neglected, suppressed, and despised populations. This orientation became the most visible characteristic of American historical scholarship in the decades after 1980, and its appeal has grown as the public influence of historians has declined. But a history that can communicate only to small communities of common discourse has little ability to reach beyond them. In the absence of credible methods to identify the agency of the voiceless, one must always wonder if the historian’s voice is the only one that is actually uncovered.16

Social Science History or Historical Social Science?

The first sustained encounter of history with social science ended badly. To be sure, certain historians and social scientists (such as the present author) persisted in their noble dream, but the vast majority turned away. In a sense, the turning occurred on both sides. History became more humanistic, and most social sciences strove to become more “scientific.”

Was this a bad thing? Many view this outcome as inevitable and, all things considered, desirable—the ineluctable natural division of labor between the deductive and the inductive disciplines, between a nomothetic (predictive, law-giving) economics and an idiographic (unique, descriptive) history. Yet, the gap is not simply a reflection of eternal academic structures. It is now larger than before, and it may well be the source of missed opportunities for both history and social science. Anyone contemplating a future of fruitful interaction must consider how the gap that now separates these disciplines should be closed. Two roads stand before us—making history more social-scientific and developing a historical social science. In retrospect, they were evident already in the heyday of the New History.

The New Economic History and the Annales School

The approach taken first, certainly in the United States, was to make history more social-scientific by appropriating both methods and theories primarily from economics but also from sociology and political science. This strategy reflects, to some degree, a Cold War–inspired desire, widely shared among American historians, for a more objective historical scholarship with standards sufficiently rigorous to withstand ideological manipulation. Benson, one of the most outspoken of the historians who embraced it, predicted in 1966 that by 1984 (a date presumably not chosen at random), his view of historical human behavior as amenable to scientific examination would be fully accepted. This exuberance was not uncommon, but also not uncontested. Already in 1963, Bridenbaugh could warn against the “bitch goddess Quantification” (and get away with it).17

The search for a new rigor in historical research was by no means confined to economic history, but it was pushed further and more irrevocably in this area. “There,” as Lamoreaux observed, “a small group of economists launched a veritable revolution, seizing control of the discipline’s organizations and using them to build a coherent body of scholarship based on the application of economic theory and econometric techniques to the study of the past.” The New Economic History (which, as it aged, began calling itself “cliometrics”) confronted traditional history with a direct challenge to its practices and norms. The articulation of explicitly stated, testable hypotheses needed to be made central to scholarly inquiry in history, as it was already in economics. Testable historical hypotheses would often require the articulation of counterfactual propositions. While challenging the discipline of history, this standpoint uncritically embraced economics. Indeed, economic history ran the danger of becoming a branch of applied economics, deploying neoclassical theory to study the same market phenomena and the same necessarily short-term processes that preoccupied most of their mainstream colleagues.18

This academic adventure was dramatic: Journals were transformed; reputations turned up-side-down; and audacious claims were made about long-festering academic debates. Yet, however dramatic, this New Economic History did not have much influence on the direction of either discipline. Most historians turned away at the first opportunity, and, as already noted, economists, not directly challenged, were merely flattered.19

In Europe during the same period, another response to the call of interdisciplinarity was unfolding. Rather than making history social-scientific, social science should be made historical. To those seized with this vision, the first step was to renovate history as a discipline to make it suitable to take its rightful place as the “queen of the social sciences.” Such was the stated objective of the French Annalistes. Braudel, the most prominent among them during the postwar era, argued that for history to become the unifying center of all the social sciences, it had to overcome its continued allegiance to a “pernicious humanism” (what he later describes as a “militant anthropocentrism”). Furnished with a “clear awareness of the plurality of social time,” and capable of incorporating the dialectics of both time and space, history—and history alone—could reconstitute the global nature of human phenomena.20

Braudel’s was a double critique—(1) of history’s excessive fondness for “mere” events (those “crests of foam that the tides of history carry on their strong backs”) and (2) of “imperialistic” social sciences, fragmented and shallow. His vision was of a reformed history that could open a dialogue with the social sciences. The intention was not so much to refashion history in the image of the social sciences but to make social science historical. History was to become a modern discipline not just by incorporating existing, ahistorical social theory—however much Braudel had fallen under the influence of French structuralism—but by deploying a new concept of duration and periodization, and a radically broadened (less anthropocentric) field of vision with respect to context and agency.21

This vision was much discussed among historians, but it failed to attract the sustained attention of social scientists, with the possible exception of anthropologists, whose discipline had serious problems of its own. It was a strong vision that did not become anything more; it is now visible only to those old enough to know where to look.

Loosening Structures and Rediscovering Events

Forty years have passed since the two paths toward historical engagement with the social sciences were both rejected. Stereotypical characterizations of the intellectual naiveté and political errors of both camps have had time to harden. But the problem with which scholars grappled back then did not disappear, and important new initiatives emerged to address it. Unlike in the immediate postwar decades, the attempts in the past two decades to restore history to social science and social theory to history have tended to come from social scientists. In many cases, social scientists see the opportunities that historians left on the table—data and case studies that these useful “drones” will not or cannot pursue themselves. But another factor is a more acute awareness of the weaknesses of ahistorical theory, especially the place of structuralism.

Structure is an important concept in all the social sciences, but it is a particularly stubborn obstacle to theoretical innovation in economics, given its position at the heart of neoclassical economic theory. Neoclassical economics relies heavily on the concept of equilibrium, which is the imagined state that would be attained when economic agents have accomplished all desired adjustments in endogenous variables to achieve their aims—that is, when all initial conditions and intervening events in the past (exogenous shocks) have been rendered irrelevant, their effects so eroded as to become inconsequential. Thus, the long run in economics is not historical; nor does it lead to historical insights. It is just the opposite; it refers to the state in which history has been erased, revealing pure, stable, unchanging structure. In this context, events are transient in their effects.22

Because they felt this tension acutely, economic historians gradually developed two influential responses to the ahistorical character of neoclassical economics. The first was to bring back institutions—“back” because the chief rival to classical and neoclassical economics before the triumph of the neoclassical synthesis (between the time of Marshall and Paul Samuelson) was the historical-institutional school, later to become known as the German historical school. This group held that economic behavior was not embedded in timeless structures but was contingent on institutions, forming complexes that exhibited an organic unity, or spirit, each of which possessed its own possibilities and, indeed, its own economics. But “back” is also the appropriate term because the task now was not only to acknowledge that history matters because of institutions, exogenously determined, but also to bring historical institutions into the realm of economic explanation—that is, to endogenize institutional change.23

Several major theorists have advanced this new institutional economics, but they share an approach to institutional change in which structure becomes something repeatedly reformulated through an internal dynamic. Historical sociologists and political scientists refer to this dynamic as “systematic constructivism”; it involves transactions between persons, groups, and other social sites that lead to accumulations of systematic knowledge that can alter existing structures. The historical challenge in this work is to distinguish what might be called “fateful” events from “mere” events. A historical economics, or historical social science, generally, would be able to identify sequences of events in the past that exert enduring effects upon later, or current, conditions that cast a long shadow forward in time.24

What characterizes the consequential, fateful events just mentioned? This question leads to the second major response to the problem of ahistoricism in economics. David and Arthur reasoned that fateful events are phenomena that do not adhere to the bounded probability distributions at the heart of neoclassical economics. They invoked nonergodic probability theory to characterize them, allowing probability outcomes that are highly sensitive to initial conditions, or initial moves. Such sequences do not gradually supersede early moves in a smoothing-out process (such as repeated flips of a coin) but take a particular direction that closes the door to alternatives, under the influence of these fateful moves. An economy’s later course—whether in institutions or technologies, or spatial concentrations of production—is dependent on the initial path; that is, it is path-dependent.25

As a theoretical concept, path dependence is daunting, if not forbidding, but what might be called the “spirit” of path dependence has stimulated numerous economists and other social scientists to create what might be called the “history matters” movement. The burgeoning literature associated with it is based more on methodology than theory, but it is dedicated to uncovering past events—often adventitious and sometimes fleeting—that have cast long historical shadows. The aim is to identify historical data sets that contain “natural experiments” (whereby a defined population received a treatment whereas others did not) and to use econometric techniques capable of divining the channels of causality that link this treatment to different outcomes in the subject population, which often endure for centuries. The fact that thus far, the “history matters” literature has been driven more by a set of techniques and by data availability than by a set of questions violates an earlier maxim, namely, that social-scientific history begins with questions formed as refutable hypotheses. But these works have succeeded in alerting scholars in a variety of subfields of economics and political science to the broad influence of path-dependent processes.26

Both institutional and path-dependence theories are attempts to loosen structural constraints without losing all coherence; to recognize the power of human agency while preserving the ability to distinguish important (fateful) acts from “mere” events; and to capture the irreversibility of time’s arrow, which can create long-persisting sub-optimalities that the passage of time does not erode.

Historical Sociology and Political Science

The other social sciences have also experienced a renewed interest in the role of history within their disciplines, emphasizing the past as context, as process, and as example—that is, as case study and analogy—much more than it has involved the development of a historicized body of theory. The structural sociology associated with both modernization theory and Marxist materialism inspired a form of historical sociology that Sewell has called teleological temporality. In works of this sort, history reveals the working out of the inherent logic of social development, as understood within a modernist or a Marxist teleology. Such work could be very powerful; Immanuel Wallerstein’s world-systems theory and Tilly’s account of state formation come to mind. But it did not so much apply theory to historical data as it used historical data to illustrate and animate theory.27

More common in both sociology and political science are comparative historical studies, which treat events and episodes—usually revolutions and other political turning points—as separate trials reminiscent of a laboratory experiment. This work has led to influential theoretical statements (as in the work of Skocpol, Goldstone, Ertman, and Pierson), but not to the development of a historical social science. It tends to treat history as a trove of information. The independence of the case studies is too readily assumed, and the comparative analysis suffers—to put it in social-scientific terms—from too few Ns and too many variables.28

Criticisms within these disciplines about their legacy of structural theories has generated a new interest in the interaction of determinate relationships and contingent events, as well as to a facile deployment of metaphors of feedback processes, configurations, crystallizations, and layerings. These terms all seek to suggest some classificatory order for historical phenomena. They are evocative, but they remain highly informal and intuitive.29

The mission in this article is not to judge specific studies but to note the absence of an engagement with the discipline of history in this historical turn. The situation in the recent past seems to be the reverse of the postwar era discussed above: Since the 1980s, historians in general have been reluctant or ill-equipped to engage with historical social scientists, whereas social scientists have been groping toward historical alternatives to what might be called their legacy theories—the modernization theory and structuralism that derive from Auguste Comte, Max Weber, Talcott Parsons, and Claude Lévi-Strauss, among others.

The Present Moment

There are now signs that historians, in significant numbers, are emerging from the comfortable refuge that their predecessors occupied nearly forty years ago. The historicist and culturalist outlook that they adopted at the time united them with others who were attempting to defend the humanistic domain against hostile, scientistic criticism. Historians could see themselves as fighting the good fight on behalf of cultural understanding, tolerance, and empathy in a hostile world. But they paid a price for this stance, and the price is now coming due.

Sewell, an enthusiastic social-scientific historian who made the cultural turn just as Stone was announcing the revival of narrative history, recently reflected about the changes to his discipline throughout his academic career: “I worry that the triumph of cultural history over social history has perhaps been too easy.… [It meant] the abandonment, without argument, of fundamental social-historical notions of social structure.… The result is a form of history that is disarmed in the face of certain important questions posed to us by the history of our own era—the changing structure of world capitalism.” In other words, historians and other social scientists who rejected structure for a postmodern sensibility—for social constructions founded in semiotics, language, and signs—achieved only an illusory liberation from uncongenial fact on the ground.30

In this postmodern epistemology, the world is a highly malleable place, shaped and reshaped by semiotic practices. But a contrite Sewell now observes that the world is not infinitely malleable—it comes back to bite you—and that historians like himself, whose political engagement had initially led them to abandon structuralism, now find themselves disarmed, unable to explain the changing material world. Sewell expresses, in short, a partial disillusionment with postmodern ideas. The insight that the social world is constituted by the interpretive practices of those who comprise it once seemed liberating—from the burden of history. But today that outlook appears more like disarmament—a voluntary shedding of the tools needed to understand consequential historical change.

The incipient return is not driven only by disillusionment and fear but also by new interests in large-scale historical questions for which the cultural turn is neither suited nor sufficient. This article cannot shoulder the task of surveying the new frontiers of historical research in any detail. It must suffice to note that global history is today the unavoidable mantra of historical research. Although economic historians have long studied globalization as manifested in long-distance trade, industrial competition, multi-national firms, etc., today the global, or transnational, perspective is the preferred vantage point from which to study many political, social, and cultural questions. This type of history requires some structure—a framework to distinguish the stable from the moving parts of the multiple contexts involved in a global historical investigation.31

Another example is the new interest in large-scale comparative history. The cultural turn succeeded in destabilizing the Eurocentric theories that had restricted both historians and social scientists to a strikingly path-dependent course of research. Once it broke that mold and proclaimed a sort of multicultural victory, there emerged pressing new questions derived from current events (the rise of Chinese and other non-Western economies) for which cultural studies had no answer. What has come to be known as the Great Divergence literature has pressed the case for the value of theory-informed, systematic, comparative historical study.32

At the same time, the technological changes of the past twenty years are “flattening the world” (Friedman’s new global competitiveness) and reducing international inequality while they greatly increase intra-country income inequality. Capitalism today is not the same beast that historians thought they knew in 1979. Sewell’s point was that historians, after the cultural turn, were left without tools to study the further evolution of capitalism. They knew only how to critique the high-modernist, “Fordist” economic life that is now a relic of the nostalgic past.33

Sewell’s self-critical analysis of history’s path for the last thirty years led him to call for something akin to an embrace of path dependence and the new institutionalism (in a manner different from that of the economic historians discussed above, of whose existence he was seemingly unaware). Historical events, he argued, are best seen as “happenings” that transform structures. “To understand and explain an event…is to specify what structural change it brings about and to determine how the structural change was effected.” In making his distinction between mere events and fateful events, he urges historians to break with narration, to suspend time, temporarily, in order to analyze and explain: “There can be no adequate diachronic narrative of an event without a synchronic understanding of the structures that the event transforms.”34

Back in 1979 Stone claimed to “detect an undercurrent sucking historians back to a form of narrative history.” But now one can detect an undercurrent sucking historians back from narrative history, and from micro- and subjectivist histories, to a concern with coherent, causal explanation of societal change. This return must include a more direct and fruitful engagement with social scientists, who for some time now have been addressing the rigidities and ahistorical character of their own theoretical equipment.35

Those abandoning narrative will have to do more than meet those abandoning structure at mid-field. The way forward is not primarily a question of compromise and reconciliation. Rather, the task before both historians and social scientists is to reformulate their understandings of both structures and events. Structure, in the sense of society as a reified totality, needs to give way to an understanding of the multiplicity of structures in all their intersections and interactions. Although mental or cultural structures (mentalities), as well as material structures, ordinarily shape people’s practices, situations arise in which peoples’ practices constitute, reproduce, and alter structures. Events, as historians comprehend and recount them, must yield to a distinction between mere events and fateful events, sequences leading to path-dependent outcomes and structure-modifying acts. This perspective requires a sustained attention to formal explanation, to causality.

This prescription conforms broadly to the vision of temporal layers developed by Braudel and the Annales School. Braudel spoke of the structures—enduring features—changing glacially, if at all, that constrained human action, and of events—mere events—fleeting and powerless to change anything of importance, numerous as they were. But between these two extremes of historical time he located something called conjuncture, marking how events and structures could come together in fateful ways. Braudel barely outlined what happened in this middle range of historical time, when combinations and sequences of action reset the course of events; less visionary, down-to-earth historians often made this mysterious category of time the butt of their jokes. At that point, historians had few allies among the social scientists. Today, a good many historical social scientists (a critical mass?) are searching for a historicized understanding of the realm in which agency and structure confront each other, and, if the signs are indeed propitious, a new era, in which historians will again seek to offer coherent explanations of change in the past, beckons.36



For the shared rejection of narrative history, see Peer Vries, Vertellers op drift: Een verhandeling over de nieuwe verhalende geschiedenis (Hilversum, 1990), 14. Peter Burke, “Introduction,” New Cambridge Modern History (New York, 1979), XIII, 1.


Bob Dylan, “Subterranean Homesick Blues,” Columbia Records, 1965; Lawrence Stone, “The Revival of Narrative: Reflections on a New Old History,” Past Present, 85 (1979), 3–24, repr. in Stone, The Past and the Present (Boston, 1981), 74–96.


Stone, Past and the Present, 91. For Stone as practitioner of social-scientific history, see idem, “History and the Social Sciences in the Twentieth Century,” in Charles Delzell (ed.), The Future of History (Nashville, 1977), 3–43; for the turn in France, Pierre Nora, “Le retour de l’ événement,” in Jacques Le Goff and idem (eds.), Faire de l’histoire (Paris, 1974), 3 v., which formed something of a Francophone pendant to Stone’s article; for the multistranded flight from Braudelian structuralism, Burke, The French Historical Revolution: The Annales School 1929–89 (Stanford, 1990), 65–93. Burke judged the Annales School to be so fragmented and diverse (in 1990) that it no longer existed as a “movement” (106–107).

The English word science has assumed a much more restricted meaning than, say, the German Wissenschaft. It is effectively restricted to the nomothetic disciplines, based on deduction and aspiring to lawmaking. The following statement of Braudel, probably written by 1979, is now met with incredulity: “Is it not the secret aim and underlying motive of history to seek to explain the present? And today, now that it is in touch with various social sciences, is history not also becoming a science of a kind, imperfect and approximate as they are” (Braudel, Civilization and Capitalism: The Perspective of the World [New York, 1982], 619–620). Even the term data currently makes many historians uncomfortable. According to Stephan Mihm, “For scholars trained in cultural history the phrase economic history implies a faith in quantification and ‘data’ that strikes them as naïve” (Mihm [with Sven Beckert et al.], “Interchange: The History of Capitalism,” Journal of American History, CI [2014], 512).


Paul Ricoeur, Time and Narrative (Chicago, 1984–1988), 3 v.; Isaiah Berlin, Historical Inevitability (London, 1954); Robin George Collingwood (ed. W. J. van der Dussen), The Idea of History, with Lectures 1926–1928 (New York, 1994), 214. In the same spirit, Jack Hexter pronounced history to be “a unique and separate domain of human knowing” (Hexter, “The Rhetoric of History,” History and Theory, VI [1967], 3–13). Hayden White, “The Question of Narrative in Contemporary Historical Theory,” in idem, The Content of Forms: Narrative Discourse and Historical Representation (Baltimore, 1987), 53.


David Lowenthal, The Past Is a Foreign Country (New York, 1985).


François Furet, “De l’ histoire récit à l’histoire problème,” in idem, L’ Atelier de l’ histoire (Paris, 1982), 73–90, argues that narrative relies on a logical fallacy—post hoc, ergo propter hoc, the confusion of temporal sequence with causal explanation.


Consilience (evidence from independent sources that converges to form strong conclusions) has also been invoked as the path whereby the sciences—and humanities—will be unified. See Edmund O. Wilson, Consilience: The Unity of Knowledge (New York, 1998). Allan Megill, Historical Knowledge, Historical Error: A Contemporary Guide to Practice (Chicago, 2007), 63–77. As Maurice Mandelbaum put it, “A story is either already known, or made up. Historians cannot make it up. Thus, they are engaged in inquiry into what is not yet known, and this inquiry cannot be essentially narrative” (Mandelbaum, “A Note on History as Narrative,” History and Theory, VI [1967], 413–419).


Megill, Historical Knowledge, 78–106. The expectations and disappointments of historians’ attempt to embrace the social sciences are recounted in Dorothy Ross, “The New and Newer Histories: Social Theory and Historiography in an American Key,” in Anthony Molho and Gordon Wood (eds.), Imagined Histories: American Historians Interpret the Past (Princeton, 1998), 92–93.


Frank Ankersmit, Narrative Logic: A Semantic Analysis of the Historian’s Language (Boston, 1983), 111; Vries, Vertellers op drift, 54–55.


Burke, History and Social Theory (Ithaca, 2005); Charles Tilly, “Three Visions of History and Theory,” History and Theory, XLVI (2007), 300. Similarly, William Sewell sees the relationship of history and social science as a one-way street. Historians use social theory to orient their thinking, or borrow its vocabulary, but they do not commonly intervene actively in social-theoretical debates (Sewell, Logics of History: Social Theory and Social Transformation [Chicago, 2005], 1–21).


Jean-François Lyotard, The Postmodern Condition: A Report on Knowledge (Minneapolis, 1984; orig. pub. 1979): “The narrative function is losing its functions, its great heroes, its great dangers, its great ages, its great goals. It is being dispersed in clouds of narrative language elements” (xxiv). Foucault addressed this issue in Michel Foucault (trans. A. M. Sheridan Smith), The Archaeology of Knowledge and the Discourse on Language (New York, 1971; orig. pub. in French, 1969), 12. For an accessible discussion, see Gary Gutting (ed.), The Cambridge Companion to Foucault (New York, 2007; orig. pub. 2003), 36. John Brewer offers a cogent discussion of this subject: “[A]s any theorist of narrative will tell you, there is no formal difference between grand narratives and micro-narratives. No story is innocent; all narratives involve plotting. They necessarily involve choice, inclusion and exclusion” (Brewer, “Microhistory and the Historiography of Everyday Life,” Ludwig-Maximilians Universität Working Paper no. 5 [Munich, 2010]).


Carlo Ginzburg (trans. John and Anne Tedeschi), The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller (Baltimore, 1980; orig. pub. as Il formaggio e i vermi: Il cosmo di un mugnaio del ’500 [Turin, 1976]); Giovanni Levi (trans. Lydia G. Cochrane), Inheriting Power: The Story of an Exorcist (Chicago, 1988; orig. pub. in Italian, 1985); Ginzburg, “Microhistory: Two or Three Things That I Know About It,” Critical Inquiry, XX (1993), 10–35; Levi, “On Microhistory,” in Burke (ed.), New Perspectives on Historical Writing (University Park, 1991), 93–113; Francesca Trivellato, “Is There a Future for Italian Microhistory in the Age of Global History?” California Italian Studies, II (2011), available at http://escholarship.org/uc/item/0z94n9hq. The purpose of theory was now not to facilitate generalization and comparison but to uncover the limits of existing theory. As Jacques Revel, an early enthusiast, wondered, “Why make things simple when we can make them complex?” (Revel, “L’histoire au ras du sol,” in Monique Aymar’s French translation of Levi’s Il formaggio e i vermi—Le pouvoir au village: Historie d’un exorciste dans le piedmont du XVIIe siècle [Paris, 1989], xxiv)—quoted in Trivellato, “Is There a Future?”).


Trivellato, “Is There a Future?” For an even more disparaging assessment of “Anglophone” micro-history, see Jo Guldi and David Armitage, History Manifesto (New York, 2014), 44–45.


For the “demise” of postmodernism, see Victoria E. Bonnell and Lynn Hunt, Beyond the Cultural Turn (Berkeley, 1999); Ernst Breisach, On the Future of History: The Postmodern Challenge and Its Aftermath (Chicago, 2003); Gabrielle Spiegel (ed.), Practicing History: New Directions in Historical Writing after the Linguistic Turn (New York, 2005).


According to Richard T. Vann, “Narrative structures…claim to represent real relations in historical actuality. But when we come to describe narrative structures we treat them as artifices, the product of different imaginations and sensibilities. We compare them purely on our preferences for the trope that informs them. Thus, there is no way to evaluate their truth-claims as narratives” (Vann, “Turning Linguistic: History and Theory and History and Theory, 1960–1975,” in Ankersmit and H. Kenner [eds.], A New Philosophy of History [London, 1995], 68). Peter Novick identified the 1980s as the moment of fragmentation, when it became clear to all that “there was no king in Israel” (Novick, That Noble Dream: The ‘Objectivity Question’ and the American Historical Profession [New York, 1988], 577–592). This theme was addressed in Naomi Lamoreaux, “Economic History and the Cliometric Revolution,” in Molho and Wood (eds.), Imagined Histories, 59–84. See also Megill, Historical Knowledge, 159–164, which defends the fragmentation of the discipline. Indeed, Megill pronounces himself “profoundly suspicious of attempts to overcome disciplinary fragmentation” (159).


Joan Scott, “The Evidence of Experience,” in Spiegel (ed.), Practicing History, 201.


Lee Benson, “Quantification, Scientific History, and Scholarly Innovation,” in Robert P. Swieringa (ed.), Quantification in American History: Theory and Research (New York, 1970), 25–29; Carl Bridenbaugh, “The Great Mutation,” American Historical Review, LXIII (1963), 315–331.


Lamoreaux, “Economic History,” 59. Not all observers interpret the same facts in the same way. Compare Lamoreaux with Jeremy Adelman and Levy’s account of the same story: “Once a mainstay of history departments, economic history was, with historians’ complicity, seized in the mid-20th century by economists who sucked the culture and chronology out of it and turned it into an obscure province of mathematical formulas” (Adelman and Levy, “The Fall and Rise of Economic History,” Chronicle of Higher Education, December 1, 2014, available at https://www.mail-archive.com/pen-l@lists.csuchico.edu/msg37633.html). Major statements of the New Economics History agenda include Douglass North, “The State of Economic History,” American Economic Review, LV (1965), 86–91; Robert Fogel, “The Reunification of Economic History with Economic Theory,” ibid., 92–98; Donald McCloskey, “Does the Past Have Useful Economics?” Journal of Economic Literature, XIV (1976), 434–461; Fogel and Geoffrey R. Elton, Which Road to the Past? (New Haven, 1983); Peter D. McClelland, Causal Explanation and Model Building in History, Economics, and the New Economic History (Ithaca, 1975).


That first opportunity proved to be the debate over Fogel and Stanley L. Engerman’s Time on the Cross: The Economics of American Negro Slavery (Boston, 1974), 2v. This comprehensive re-assessment of the economics of the slave labor in the Southern plantation system invited the critical scrutiny of a large contingent of New Economic Historians who wrote about one or another aspect of the same historical issues. True to the scientific spirit of this new history, they sought to replicate Fogel and Engerman’s quantitative findings, test them for statistical significance and robustness, and scrutinize the appropriateness of the hypotheses, which they could do because Fogel and Engerman, in true scientific spirit, made the relevant quantitative evidence available to their critics. For the resulting set of bracing critiques, see Paul David et al., Reckoning with Slavery (New York, 1976)—in many ways a noble monument to the new mode of historical scholarship, though most historians saw it as their hall pass to escape from the classroom. If these new historians could disagree so intensely, their critics persuaded themselves, their methods cannot be worthy of our serious consideration. For this episode in the flight of the historians, see Jan Willem Drukker, The Revolution That Bit Its Own Tail (Amsterdam, 2006), 140–147; Lamoreaux, “Economic History,” 72–75.

The flattery did not win over every economist. Robert Solow, for one, yearned for the more readable old economic history: “Apart from anything else, it is no fun reading the stuff anymore.… It gives back to the theorist the same routine gruel that the economic theorist gives to the historian. Why should I believe, when it is applied to thin eighteenth century data, something that carries no conviction when it is done with more ample twentieth century data?” (Solow, “Economic History and Economics,” American Economic Review, LXXV [1985], 330).


Braudel, “Histoire et sciences sociales: la longue durée,” Annales (economies, sociétés, civilisations), XIII (1958), 725–752; idem, “Pour une économie historique,” Revue économique, I (1950), 37–44. The summary statement is drawn from François Dosse, L’historie en miettes: des ‘Annales’ à la nouvelle histoire (Paris, 1987). “It [history] alone had access to what [Braudel] called the ‘ensemble of the ensembles,’” 89. Yet, the title of Dosse’s work translates to “history in crumbs.”


See Braudel’s inaugural address to the College de France in idem, Écrits sur l’ histoire (Paris, 1969), 11–40. Space does not allow a more extensive survey of social-science–inflected history in these years, but deserving of mention is the influential social-history movement in Germany, led by Hans-Ulrich Wehler, Geschichte als historische Sozialwissenschaft (Frankfurt am Main, 1973), and Jurgen Kocka, Sozialgeschichte: Begriff-Entwicklung-Probleme (Göttingen, 1977).


When Alfred Marshall, Principles of Economics (London, 1890), adopted the epigraph Natura non facet saltum (nature makes no leaps), he appropriated from evolutionary theory—as recently developed by Charles Darwin, Jean-Baptiste Chevalier de Lamarck, and Georges Louis Leclerc, Comte de Buffon—the doctrine of continuity, namely, that nature always has time enough to achieve its aims. John Maynard Keynes famously quipped, “In the long run we are all dead”; nature’s way with time is not that of human society.


The German historical school criticized the marginalist school for its misplaced scientism, which entered all sorts of normative claims into the theoretical canon of economics, based on little more than subjective judgments. According to these (German) institutionalists, marginalists claimed to describe “how the world works,” but they could only, at most, describe how the world ought to work under certain assumptions. The remedy was a more historically based economics, sensitive to the differences between societies and, especially, within societies over time. See Heath Pearson, “Was There Really a German School of Historical Economics?” History of Political Economy, XXXI (1999), 547–562; idem, Origins of Law and Economics: The Economists’ New Science of Law, 1830–1930 (New York, 1997).


For an introduction to the concepts of the new institutional economics, see North and Robert P. Thomas, The Rise of the Western World: A New Economic History (New York, 1973); North, Institutions, Institutional Change, and Economic Performance (New York, 1990); idem and Barry Weingast, “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England,” Journal of Economic History, XLIX (1989), 803–832; Avner Greif, Institutions and the Path to the Modern Economy, Lessons from Medieval Trade (New York, 2006); for systematic constructivism, Tilly and Robert E. Goodin, The Oxford Handbook of Contextual Political Analysis (New York, 2006).


Paul David, “Path-Dependence: Putting the Past into the Future of Economics,” IMSSS Tech Report no. 533 (Stanford University, November 1988); W. Brian Arthur, Increasing Returns and Path Dependence in the Economy (Ann Arbor, 1994). Path dependence can be defined as a process in which the long-run character of a system depends critically on the history of the system, that is, on a specific sequence of events. Examples of its application include the celebrated case of the qwerty keyboard, standardization in technology more generally, regional economic development, market failure (or sub-optimality), and directionality.


For useful overviews of the substantial history-matters literature, see Nathan Nunn, “The Importance of History for Economic Development,” Annual Review of Economics, I (2009), 64–92; Abhijit V. Danerjee and Esther Duflo, “Under the Thumb of History? Political Institutions and the Scope for Action,” ibid., VI (2014), 951–971; Ran Abramitzky, “Economics and the Modern Economic Historian,” Journal of Economic History, LXXV (2015), 1240–1251. Particularly influential have been the many works of Daron Acemoglu and James Robinson, most notably Why Nations Fail: The Origins of Power, Prosperity, and Poverty (New York, 2012).


For an insightful review of the uses of historical analogy and example, see Barry Eichengreen, “Economic History and Economic Policy,” Journal of Economic History, LXXII (2012), 289–307.


Theda Skocpol, States and Social Relations (New York, 1979); Jack Goldstone, Revolution and Rebellion in the Early Modern World (Berkeley, 1991); Thomas Ertman, Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Europe (New York, 1997); Paul Pierson, Politics in Time: History, Institutions and Social Analysis (Princeton, 2004). Regarding history as a trove of information, consider Pierson’s assessment of the place of history in political science: “Of course, the social sciences have had a rich tradition of historical research. Scholarly communities devoted to extending such traditions flourish in parts of the social sciences. Indeed, some claim to witness a ‘historic turn’ in the human sciences as a whole. Yet in spite of this activity there has actually been surprisingly limited attentiveness to the specifically temporal dimensions of social processes. In contemporary social science, the past serves primarily as a source of empirical material rather than as the site for serious investigations of how politics happens over time” (Pierson, “The Study of Policy Development,” Journal of Policy History, XVII [2005], 34–51. See also idem, Politics in Time: History, Institutions, and Social Analysis [Princeton, 2004]).


Elizabeth S. Clemens, “Sociology as a Historical Science,” American Sociologist, XXXVII (2006), 30–40.


Sewell, Logics of History, 22–80 (emphases added). Another expression of regret, if not contrition, comes from Pierson, a political scientist: “The collapse of the modernization literature is a clear case of throwing out the baby with the bath water. The discipline [political science] jettisoned an undesirable functionalism and teleology.… Along with it went the discipline’s most sustained efforts to think about long-term processes of social and political change” (Pierson, “Big, Slow-Moving and Invisible,” in James Mahoney and Dietrich Rueschemeyer [eds.], Comparative Historical Analysis in the Social Sciences [New York, 2003], 199).

More recently, Guldi and Armitage’s Historical Manifesto lamented history’s retreat from the public arena, calling for a new dedication to “serious history” focused on large-scale and long-term themes of broad societal importance. However, a re-engagement with the social sciences does not figure in their vision for the future of the discipline. They are especially exercised by the malign influence of economists: “One consequence of the retreat of historians from the public sphere was that institutions were taken over by other scholars, whose views of the past were determined less by historical data and more by universal models. Notably, this meant the rising profile of economists…[whose] universalising models came to dominate conversations about the future” (11–12). By what means do they foresee historians regaining the capacity to speak cogently and usefully to the political nation? In this work, they have remarkably little to recommend. Apparently, their view is that traditional narrative methods will suffice if historians once again address questions on a sufficiently long time scale. They imagine “big data” to be the historian’s secret weapon but suppose that these data will speak effortlessly to narrative historians if only they assume the correct attitude.


For critical assessments of global historical research, see Maxine Berg (ed.), Writing the History of the Global: Challenges for the 21stCentury (New York, 2013). Even micro-history has acquired a global scope: Timothy Brook, Vermeer’s Hat: The Seventeenth Century and the Dawn of the Global World (New York, 2008).


Notable works about the Great Divergence include Roy Bin Wong, China Transformed: Historical Change and the Limits of European Experience (Ithaca, 1998); Kenneth Pomeranz, The Great Divergence: China, Europe and the Making of the Modern World Economy (Princeton, 2000); Bin Wong and Jean-Laurent Rosenthal, Before and Beyond Divergence: The Politics of Economic Change in China and Europe (Cambridge, Mass., 2011); Prasannan Parthasarathi, Why Europe Grew Rich and Asia Did Not: Global Economic Divergence, 1600–1850 (New York, 2011).


Thomas Friedman, The World is Flat: A Brief History of the Twenty-First Century (New York, 2005).


Sewell, Logics of History, 219.


The pull to causal explanation is strong, but many historians remain skeptical and even fearful. See the many caveats in the generally positive account of Adelman and Levy, “Fall and Rise of Economic History,” and the wide-ranging but also fear-riddled panel discussion in Beckert et al., “Interchange,” 503–536.


For a recent endorsement of Braudel’s concept(s) of temporality, see Trivellato, “Un nouveau combat pour l’historie au XXIe siècle?” Annales: Histoire, Sciences Sociales, LXX (2015), 333–343.