Sergei GLEBOV: Professor Sahlins, thank you for your interest in the general questions we sought to discuss in the framework of our thematic issue “The Letter of the Law: the Institutionalization of Belonging to Polity” and for your willingness to share your thoughts with our readers. Let me begin by asserting that the narrative of Modernity is essentially a narrative of the nation: the “revolutionary nation” as the political body and the “eternal nation” as the physical body of the society, united by a common language, culture and memory. All contradictions and ruptures of Modernity are mysteriously brought together when viewed through the national perspective: the inevitable monological form of narrative finds its ultimate subject in the singularity and homogeneity of society as embodied by the nation. The revolutionizing effect of forging the common narrative of the nation (parallel to the forging of national identity itself) is well known, not least thanks to your seminal studies.
What remains understudied yet is the functioning of societies that have not fully experienced the integrating potential of “nationalization.” Old regime polities, as well as contiguous empires of the nineteenth century (Russian but also Habsburg) did not overcome local particularities (in both a regional and social sense) as rival sources of group identification, parallel to the pan-imperial narratives of unity and loyalty. In the twentieth century, the Soviet Union represented a modern post-revolutionary polity, yet it never managed to become a “proper nation state” because a federalist model was employed to accommodate deep cultural, economic and ethnic inequalities concealed by the umbrella of political loyalty to the regime. Today we witness attempts to reshape Europe as a supranational community. What is important in all those different cases is that the internal heterogeneity of society goes far beyond a “normal diversity,” to the extent that it includes the co-existence of different political subjects holding different degrees of sovereignty, competing principles of social identification, and narratives of memory. We believe that your interest and experience in studying the beginning of the synthesis of “national narrative” provide you with a unique perspective on the world before and beyond the nation.
To begin our conversation, let me ask you how accurate is the very perception of the national ideal as a monologue (even if established as a result of disputes and conflicts)? A decade ago, James Lehning^[1] challenged the perceived wisdom of Eugene Weber’s model of forging a nation through institutional standardization, suggesting instead a more complicated vision of national unity as a result of negotiations of mutual projections by social actors. What is your attitude to Lehning’s model, and does it change the perception of the nation as a normative monologue?

Peter SAHLINS: It is worth beginning with Eugen Weber’s model of the transformation of Peasants into Frenchmen,^[2] since, I’d like to suggest, Lehning’s attempt to revise Weber’s formulation is still very much framed by the same kind of oppositions that he purports to disrupt – between the traditional and modern on the one hand, and peasants and Frenchmen on the other. What this suggests, to me at least, is the deep-seated nature of the paradigm of “cultural modernization” and the difficulty, even within the framework of cultural history, of disrupting it or dislodging it in some significant way. In my earlier work, on boundaries,^[3] and also in my work on peasant rebellion in the nineteenth century,^[4] I was very critical of Weber’s model and by extension of Lehning’s attempt to reformulate it, largely because of the implicit model of collective identity it contained. Specifically, I would suggest that such models still imagine identity to be constructed as a series of expanding concentric circles, in which identity and loyalty decrease in correlation with geographic distance from a specific social ego at its center, such that a peasant’s attachments, in this schema, would be primarily to his or her family, and would be diluted in their extension to a kin network, then to a neighborhood, then to the village community itself, then perhaps to a valley, a region, and only distantly and weakly to the nation as a whole. Implied here is also a paradigm of nation-building that assumes that when nations are built from distant centers, they reverse the vectors of loyalty and identification, effacing the embedded concentric circles, such that a direct and unmediated identification between the peasant and the nation, in this case France, is achieved. This is what you’ve called the nation as normative monologue. To my mind, deploying this model is not always the most useful way of making sense of the regularities in the historical record because to do so presupposes, including in Lehning’s reformulation, an original position occupied by peasants as outside of the discursive, institutional or political community called the nation. My own work included an effort to re-imagine the peasantry as part of France, to write the history of the peasantry, however marginalized and peripheralized with respect to a distant political center, as nonetheless engaged, or at least articulated within the same historical processes. In doing so I tried to rethink the model itself, abandoning the metaphor of circles for the notion of segments, which I borrowed from a certain anthropology, and which was well known, at least among the structural functionalists, through the work of Evans-Pritchard.^[5] In this “segmentary model,” identity is conceived in all of its possible iterations as an oppositional and contingent and relational quality, capable of collapsing lesser distinctions into more inclusive ones. In my work, this meant that peasants might express their identities in village communities at the same time that they could consider themselves Frenchmen or Spaniards, and this occurred in an historical context that we might consider precocious, since the institutional mechanisms outlined by Weber that link peasants and the nation – roads and railroads, schools and military service – did not yet exist. Still, through segmentary oppositions, peasants could identify themselves as part of France, but only in opposition to an Other. In the Pyrenean borderland in the Pyrenees, the Other was Spain, even if Spain had just as ephemeral an institutional existence in the pre-modern world, at least in terms of the homogeneous creation of national institutions. Nevertheless, discursively, the Spanish nation or Spain as a nation was an entity, which became strategically deployed within peasant society in order to state a set of claims about local and national identity. Key here were the ways in which the national as a category became articulated with the local, in such a manner that neither effaced or erased the other: a localizing of the national and a nationalizing of the local. So I was most interested in my early work in critiquing the expectation that nation-building involves the complete effacement of other kinds of identities and other kinds of differences. Not that this wasn’t, in fact, the political project, since it really was a goal of statesmen and politicians (and educators and army officers) who “built nations,” but it was never a successful project, and not even in the most precocious and developed of the nation-states, England or France, or to a certain extent Spain, did the effort ever come to approximate the lived experience of peasants and others. All the more important, I think, turning to imperial and post-imperial histories further east on the continent, to emphasize the extent to which national-building agendas, agendas of nationalization, with their integrating, homogenizing efforts, were never nearly as successful as nation-builders imagined them to be.

SG: I wonder if I can interject a question at this point regarding something that you mentioned in your answer, namely, your reliance upon and indebtedness to anthropological models. Could you elaborate on how important anthropology has been to your intellectual project, and, furthermore, what, in general, is your perception of the relationship between anthropology and history? Are we indebted to anthropologists and if so, what kind of an intellectual debt do we owe to them? Is it methodology, analytical concepts, or a conceptualization of the language of social sciences?

PS: There are, obviously, two histories here that come together. There is the history of the disciplines, but also a personal trajectory. My own exposure to anthropology took shape as a contingent and accidental development, namely my birth and education in a family which lived all over the world, and in which anthropology and culture was the stuff of the dinner table conversations. I was never trained in anthropology but I grew up in a world in which the concepts and key categories of anthropological knowledge, at least of a certain moment, were part of an everyday language, so generally speaking my “training” as an anthropologist comes from home. More generally, I think that history as a discipline has developed and flourished during the last century through a process of cannibalizing, if you will, collateral disciplines in the social sciences and humanities. One can point to different decades in the twentieth century in which different auxiliary disciplines, from economics to sociology to anthropology to literary theory to geography, have been not just helpful but necessary for history as a discipline in its continuous self-re-invention. The anthropological moment of historical inquiry has in some sense passed, meaning that the heyday of this borrowing can be seen in the works of E. P. Thompson or Natalie Zemon Davis or any of the so called “new historians” of the Anglo-Saxon world, whose research agendas came out of an interest in social history and “history from below,” beginning in the late 1960s. This was an especially creative and fertile time for the marriage of anthropology and history, a moment during which a single collateral discipline, in this case anthropology, really allowed history to pose new questions about the collective logics of behavior, or invent new objects of inquiry (such as kinship, ritual, or other symbolic practices). The marriage also provided historians with a vocabulary with which to investigate and to answer queries that, at least in their most successful iterations, were never efforts to import wholesale the methods of anthropological inquiry onto history, which of course wouldn’t work in that the field is not the archive, and historians will always be bound to a great extent and constrained by this silence of their informants… Rather, historians imported not the research methods of anthropology but its vocabulary, its questions, and certain of its intellectual concerns and agendas… All this is not to say that this moment has definitely receded into the mists of time, but there is an enduring legacy to be found in the ever widening set of legitimate historical subjects, and there is still fruitful cross-fertilization that can occur at this point in time, particularly around the much studied question of identity. At the same time, it should be emphasized, that history as a discipline – its central paradigms and informing principles – has already learned its lessons from anthropology and has moved on to other disciplines, from which it takes equally in measure to think of new sets of problems and ways of interpreting them. Similarly, when history turned to literary theory in the 1980s, what was at stake was less a wholesale importation of methods – even if there are historians who would argue that history is a text and should be read in the same way as a literary creation – but most practicing historians still work in archives and now understand that they are working with texts in an important, literary sense, and that all of the aporia and explicit meanings of a text that need to be studied as part of the way of making these texts speak to a particular intellectual problem that’s been posed. So I would not be a historian who continuously waves the flag of anthropology feeling that this is in any way a definitive solution or even the first steps down a particular path, but, rather, one of the many tools in the rather capacious toolbox of the historian that can be used to make sense of a changing and evolving set of problems that we will continue to invent and give our best to answer.

SG: Despite some efforts to undo the boundary between the modern and pre-modern forms of citizenship, historians still operate under the assumption that there occurred, at the time of the French revolution, a profound break with citizenship based on privilege (or private law). To what extent has your own work contributed to complicating that boundary? Has the story of the passage from a foreigner to a subject altered our perception of the roots of modern citizenship?

PS: It’s harder to imagine in a French institutional context and historiography, but there have been a lot of efforts in English to de-center the French Revolution itself as the origins of modernity, at least within the accepted narrative of the development of modern citizenship. My own work, and that of other historians of the eighteenth century, has helped us to de-mythologize, in some sense, the central place of the French Revolution in the discipline itself, especially as it’s been developed in France and in Europe. This is not to say that we, historians, who are deeply attracted to the mutations of the eighteenth century long before the French Revolution necessarily see the Old Regime as inevitably containing all of the elements of “modernity” that would come to maturity in the nineteenth and twentieth centuries. The problem, rather – and this a theme that recurs in your questions – lies in trying to imagine an Old Regime that is independent of its outcome, or of what we see retrospectively as some kind of inevitable outcome, meaning 1789 and all that. Many historians using different approaches and drawing on different methods, especially in intellectual and cultural history, are finding possibilities to talk about the ways in which the discursive contributions and transformations of the eighteenth century find their expression in the French Revolution but cannot be situated as a cause. We are far beyond the conservative reactions during the revolutionary upheaval itself that linked the rhetoric of Enlightenment and the revolutionary process: “it’s Rousseau’s fault,” “it’s Voltaire’s fault.” So my work, in that sense, like the work of Keith Baker, Roger Chartier, and younger scholars like Michael Kwass and others, is very much part of this effort to move back from the Revolution as some inevitable outcome of the Enlightenment or eighteenth century developments, and to think through the kinds of modernity that took shape in the eighteenth century, but that were not structurally determined to produce, inevitably, a revolutionary outcome.
All this to preface my comments on citizenship itself. What struck me and what got me started on the project that became my last book,^[6] Unnaturally French, was an initial surprise, and indeed astonishment, about the vocabulary that I discovered when reading juridical texts in the seventeenth and the eighteenth century. I never expected to find the word “citizen,” citoyen, and indeed, to find it recurring frequently, albeit within a relatively isolated linguistic domain, among lawyers. More, it was a bit of a revelation to read how the category was quite elaborated in jurisprudential terms. That was the starting point for re-thinking what it would mean to actually use a term we associate with certain characteristic features of modernity, in particular with political participation, equality, and cultural homogeneity. What do we mean when use such a term in which the referent had nothing to do with the modern world? One possible response might simply end the claim there and say “well, the word is used but it had nothing to do with the thing as we know it and as we practice it in the post-revolutionary world.” That was not my choice because I did believe that there was something intrinsically important about the way in which the word was used, which stood in some relation, but again, not easily predictable or inevitable one, to the development of the modern notions of belonging, attachment, and loyalty. And one of the ways of thinking about what that relationship might be between the modern and the pre-modern forms of citizenship was to explore, in some important sense, what the difference would result in not using the word. Some of my critics have said, well, in fact, all you’re talking about is “subjecthood” that happens to be called “citizenship,” or more accurately, “subjects” who happen to be called “citizens,” but there is no fundamental distinction. And that is actually a position that comes, among others, from Rousseau himself – at the moment of a great intellectual and cultural mutations of the eighteenth century and the Revolution – this idea that all inherited linguistic categories were wrong and that the world can be linguistically created anew. Rousseau himself was quite critical of what he called “the egregious error” of the sixteenth century jurisconsult Jean Bodin, who had relied heavily on the term “citizen” in his treatises. For Rousseau, the citizen was unimaginable before the time that he himself could think it up, in its modern iteration. In my own understanding, there is an important and subtle distinction between subject and citizen in the Old Regime, in both discursive terms and in their practical consequences. I think that the subtle distinction has to do with the fact that although all citizens were subjects, and although in fact most subjects were citizens, the distinction made a difference, that is, it had had practical consequences to name someone a citizen. True, as Bodin himself had seen, the category of “citizen” was heterogeneous and internally differentiated; citizens were by their nature unequal in obligations, in their privileges, in their liberties, in their franchises, to use all the terms of the Old Regime. But despite this heterogeneity, there was in fact an underlying unity, and one with real life practical consequences, that emerged not from efforts to create homogeneity out of difference, but that appeared by drawing distinctions between citizens and foreigners, that is, those outside the boundaries of citizenship. Foreigners suffered any number of legal disabilities, including, most importantly, the inability to deed and inherit property from natural-born Frenchmen and women (the gendering is important here because unlike modern citizenship in its initial iteration, Old Regime citizenship was a status to which women as well as men could both aspire and acquire). That oppositional nature of citizenship actually places the category of citizenship much closer to a notion that did not exist linguistically in the Old Regime, the category of “nationality.” This is a tricky and complicated arena, because in so many ways the notion of pre-modern citizenship leads not in some evolutionary sense towards modern citizenship, but rather towards a modern conception of nationality itself, at least in a juridical sense. So the confusion is doubled, because on the one hand the term “citizen” doesn’t seem to belong in the eighteenth century, and on the other hand, when it does appear in the Old Regime, it’s read better through the lens of nationality and nationality law, then it is through citizenship as political participation, as a conditional equality, or as corresponding to a certain cultural homogeneity. And so my own work in some sense has followed this slippage and contributed, hopefully, with some productivity and fruitfulness to the confusion by stressing the extent to which in examining the pre-modern citizen and pre-modern citizenship, what I am really after is the… well, the French title of my Annales article, “Nationality avant la lettre,”^[7] the history of nationality before the word itself had come into being.

SG: Your complex vision of nationality avant la lettre in Ancien Régime France suggests a definition of the citizen as someone not subject to the limitations imposed upon a foreigner; albeit, of course, “citizens” were not equal before the law or to each other. Then to what extent does “pre-modern” or pre-revolutionary citizenship in France compare to, or help our understanding of the phenomenon of subjecthood in composite imperial states? To put it simply, does such citizenship equal subjecthood, given that it was defined by obligations, privileges, and rights particular to one’s social position?

PS: It is always a struggle, and frequently a productive one, to try to make intellectual linkages between the pre-modern and the modern that are not over-determined. Some sites and regions of the modern world would seem to lend themselves more easily than others to certain kinds of comparisons. So, for example, in thinking about the modern experiences of empire and nation among the sprawling polities of the nineteenth century empires, that is, Russian but also Habsburg, an obvious point of reference in the pre-modern world might be Spain and the Spanish Habsburg monarchy of the early modern period. The Spanish Habsburg empire, like most states in the pre-modern period, has been usefully identified by John Elliot among others as a “composite” monarchy, that is, one that is literally composed of distinctive polities. It is thus quite similar in structure to the nineteenth century Habsburg Empire as described by Benedict Anderson, who points out in Imagined Communities how the emperor himself holds literally dozen of separate titles corresponding to the component polities of the empire, from “king of Bohemia” to “duke of Carinthia” to “Margrave of Istria” and so forth. Now, in the pre-modern world, what is important, and this is in some sense a pertinent observation for the 19th century empires as well, is that each of these polities that together “compose” the empire is constituted not simply of “subjects” of a particular jurisdiction but also by a legal framework of rights and disabilities that comes to approximate a modern notion of nationality. So, for example, in the early modern Spanish empire, each of the composite polities, that is, the Kingdom of Aragon, or of Castile, or of Naples, had its own institutions, its own legal framework of privileges and prerogatives and rights and obligations, but also of disabilities and exclusions – political, professional, legal – for those who were not members of that particular group. In other words, I think we can speak, with some caution and many caveats, of the existence of nationality law in the pre-modern Spanish empire in the same way that we can make this observation for the more politically homogeneous monarchies in the same period, such as France. Now, what’s interesting about Spain is that the movement of political modernization, including the reforms of the Bourbon monarchy at the beginning of the eighteenth century, produced an institutional and legal framework of belonging that transcended the more particularistic and heterogeneous framework of an early modern polity, and was thus more “modern” in the sense that a broader, “national” set of institutions and laws came to displace, however slowly and incompletely, the distinct privileges and franchises of the composite polities of the empire. The Bourbon reforms of the early eighteenth century in Spain were already anticipated and reflected in the so called “Indies Laws,” in these laws of the Habsburgs, in which it is possible to tease out the notion of a “Spanish” identity, and indeed, a “Spanish” nationality that transcended and displaced the different legal frames of belonging within the composite monarchy. Projecting this model onto the contiguous empires of the nineteenth century, it’s altogether possible to imagine how the movement toward a legal idea of nationality as both a project of state building from above but also a project of resistance to imperial structures was not dissimilar to processes found in the pre-modern world. I’m not saying that pre-modern states attempted the kind of nation-building that later empires undertook (without much success), but I am suggesting that the conditions under which it became possible to develop a claim that was rhetorical but also institutional about the separateness and identity of a component part in a monarchy was not entirely dissimilar from the structures and processes of group membership as these took shape in the early modern world.

SG: Well, it’s an interesting perspective. I am curious, though, about your use of the term nationality, especially when it is applied to pre-modern and composite polities. Am I right in assuming that your reference is really about an anthropological perspective of the sense of belonging to the polity first of all?

PS: No, not quite. I’m pulling the term much more into its juridical framework, away from the more anthropological sense of homogeneous cultural identity, which really accrues somewhat later and becomes ever more important as the nineteenth century progresses. And it’s certainly true that the term “nationality” appears in European languages with both meanings: the first literary uses of the term, say in Germaine de Stahl’s Corinne and Italy in 1807, refer to an idea of belonging that’s founded on some deep cultural, we might today say, ethnic sense of collectivity; but the term “nationality” also makes a near simultaneous appearance in administrative discourse and then a little bit later in legal terms to describe a juridical notion of belonging empty of a particular cultural content, that is, which doesn’t depend on a shared similarity of custom, language, culture, or even historical experience… In short, it doesn’t depend on the notion of shared culture, but on a legal framework that identifies the formal rules of inclusion and exclusion.

SG: To what extent did the revolutionary nation exist in a latent form within the ancien régime society, or did it emerge completely from scratch under the impact of revolutionary experience and to be further developed during the struggle with the remnants of the Old Regime and new challenges of the moment? What part of the legacy of the ancien régime was inherited by the revolutionary nation, either positively or by negation of the old world? Could we trace elements of the revolutionary discourse (in a literal sense, as rhetorical devices and tropes) in the pre-revolutionary cultural milieu, beyond usual references to the radical representatives of the Enlightenment? Is it possible to imagine the French Enlightenment not resulting in the revolutionary outburst of the type that actually did take place in 1789?

PS: The hardest assignment I ever faced as a professor was teaching the eighteenth century, for in this period all of the difficult and hoary questions are brought to the surface — history as outcome, history as origin, history as condition for the possibility of other histories, and history as a framework for being able to account for modernity. In brief, it’s nearly impossible, I’ve found, to teach the eighteenth century without knowing and anticipating the outcome, that is, without teaching the outcome. It’s very hard to treat the ancien régime as a period in and of itself, without teaching that the ancien régime, at least in France, comes to an explosive end in a rather abrupt and unexpected way. This presents enormous obstacles for trying to make sense of the ancien régime on its own terms starting with the nomenclature that we use to describe the eighteenth century. In calling something the “Old Regime,” we’re obviously implicitly positioning a new regime that succeeds it. In French, ancien régime is perhaps better translated as “past regime” as opposed to “old regime”, and the ancien régime was invented by the revolutionaries themselves at the beginning of the revolution in order to describe something that was quite new. Indeed, one could argue that there were certain critical moments — such as the decrees that came out of the night of August 4th, 1789 — in which the ancien régime was literally invented in order to be dismantled, and laws proposed at these moments become a kind of systematic inventory of the institutions of the Old Regime, described for the first time in their institutionalized, reified form as “old” precisely in order to be dismantled. As historians, as teachers, we’re confronted with this very deterministic and teleological reading of the eighteenth century that requires us to at once make sense of the fact that there is a revolutionary rupture at the end, and to avoid the argument that revolution was in constant preparation in the course of the eighteenth century. So, to return to your question, yes, many historians have tried to rethink in many different kinds of ways the relation of eighteenth century developments, especially intellectual ones, and the French Revolution. Some have located a revolutionary discourse as a set of rhetorical devices or tropes in a pre-revolutionary set of contexts. Consider the work of Keith Baker, who isolates three different discursive strands that will appear in the Revolution but whose intellectual origins he locates at different points in the eighteenth century. This is not to say that their utterance or their iteration during the course of the eighteenth century was a cause of the French revolution; indeed, most intellectual and cultural history these days stresses conditions, not causes, trying to disengage Enlightenment and Revolution at least from an over-determined relationship. But this can go too far as well. The reductio ad absurdum of trying to disengage the Enlightenment from the French Revolution in the work of Roger Chartier, for example, self-consciously and almost perversely states that “the Enlightenment did not cause the French Revolution but the French revolution caused the Enlightenment,” in the sense that it gave it coherence and identity and the historical role that the Enlightenment or Enlightenment thought would not have had had there not been a rupture with the Old Regime.

SG: Your work has played a very important role in applying anthropological methods to the study of group formation. In recent years, many scholars questioned not just the concept of identity (I have in mind the works by Frederick Cooper and Rogers Brubaker) but also the usefulness of operating with notions of groupness. For historians, however, the task becomes increasingly complex because in our research we are constantly in need of terms and concepts to describe “groupness” (e.g., nations, classes, ethnic, confessional and linguistic groups). How can we reconcile an understanding of groups as constructed and invented with the need to write readable and comprehensible histories taking into account how people described their own sense of belonging to a group?

PS: Right, well, this is a problem that periodically reappears throughout the social sciences, not just in history, and it’s often framed as the classic contrast between an emic and an etic approach. These are terms that come from the linguistic anthropology of Kenneth Pike in the 1950s,^[8] who tried to distinguish, without necessarily stating what the relationship was between the two, between approaches and categories of interpretation that emerged from the lived experience of the subjects who are the objects of study (the phonemic tools that the users of a language might have), and contrasting that with the phonetic grammar that is the etic, rules and categories imposed by non-users (and indeed incomprehensible to users themselves). It’s a classic opposition; never resolved, frequently invoked, and debated in particular as concerns the relation between the two perspectives. How do we reconcile, as you say, our use of these terms with more indigenous ways of formulating belonging and identification especially in the last couple of decades as we’ve come to understand the problem of identity as a construct or imagined category? I think that all of us who are sensitive to the slippage, and probably anthropologists more than others, would note that the last thing that an informant would be able to identify with is a claim that his sense or her sense of belonging was invented. And in fact, in terms of the patterns of collective behavior, so much of what we understand as group belonging is so deeply kinetic and real and indeed violent and destructive that the very description of identity as “invented” or “constructed” seems to do injustice to the very real nature of historical experience. I don’t think that there is a simple or universal solution, no magic bullet, and no knife with which to cut this Gordian knot, at least none that is consistently intelligible across disciplines and across cultures and across historical periods. I do think that in the historical work that I’m reading more recently, there’s a tendency to move away from the term identity because of the perception that it’s been overly contaminated by its use and abuse in other disciplines and especially in linguistic theory and cultural studies, in which the element of constructedness is really overplayed if not over-determined. We thus need to continuously expand our conceptual vocabulary, to develop a language that will allow us to think about collective belonging and affiliation in ways that communicate the sense in which these categories of belonging are simultaneously imposed as arbitrary, artificial constructions, but also lived and experienced in their vernacular iterations as something that is truly close to experience. Identity no longer does that, and the term I’m seeing more often now is identification, not used as much in an psychoanalytic sense, where it has a very precise meaning, but in a social historical sense, as a term which can in fact move between an emic and an etic perspective but also move between an official effort to classify and to categorize and to identify people and vernacular appropriations of these categories. The work of Gerard Noiriel, the historical sociologist in France, has been valuable in thinking about identification in its official framework, but what’s equally interesting is how the initially arbitrary attempts to classify, inventory and categorize people also become meaningful categories of collective belonging. Indeed, that’s the great paradox and the great mystery, how what initially appear as arbitrary, contingent and accidental, if politically powerful categories of belonging like the nation become real, and how people in history often begin with very strategic and instrumental stances when faced with such external identifications, but over time, these initially quite arbitrary distinctions turn into meaningful categories of collective action in that way. I think that it’s a very similar process to what happens with boundaries themselves, which I discuss in my earlier work. It was a problem that always somewhat baffled me, and I consider it one of the great mysteries of historical inquiry itself: how boundaries, initially completely arbitrary, were very much constructed, accidental divisions of territorial arrangements, became over time meaningful structures that informed and framed the way in which people collectively thought, acted, and behaved in the world.

SG: Thank you very much. Your answer suggests an interesting turn that can, it seems to me, add to the discussion a line of inquiry into the categories of practice and categories of analysis and the ways in which we can avoid the trap of applying categories derived from everyday life experience to our analytical purposes. For example, we had a very interesting exchange published in Ab Imperio that revealed difficulties which the French society in general, and French intellectuals in particular have in applying terms other than social, universally social, to the recent riots in France and one of the questions we tried to pose to our French audience was can we actually interpret these riots as the outcome of a new ethnic experience in the making, and the response of course was quite negative – No, you cannot do that, because this was a social experience, an experience derived from unemployment, class, and social identity.

PS: It’s very complicated in the case of the French riots of November 2005, in part because the social sciences in France are institutionalized within a political culture that is rather different from your experience or mine, in which certain terms like “ethnicity” are virtually taboo and are, in fact, legally proscribed, such that French social science cannot actually measure ethnicity because the French government in its census refuses to allow for that category to exist, given the Republic model of civic integration. I myself have spent many a long evenings arguing with French social scientists, whom I respect and whose work I think is wonderfully wise and important, but who I find to be blocked on this very issue of how to think ethnicity in France because they are so much the product of an institutional culture in which even the act of thinking ethnicity is so deeply challenging to the republican model of civic identity.

SG: Let me return for a moment to the historical conjuncture of the ancien régime and modern society. Did French society during the ancien régime have a vision – and means – to express itself in a self-descriptive narrative (whether in visual symbolism, in music, or in text)? How can historians write a narrative of the pre-national (non-national) polity? Should we see the project of “total history,” in its Braudelian version as such, as an attempt to write about multiple subjects and actors who are simultaneously incorporated into a variety of hierarchies without direct correlation of status among them?

PS: Well, it’s a good question, it’s almost an impossible question because the project of total history is almost by its nature, despite Braudel’s best effort, an impossible project. I don’t think that we historians ever really aspire to write about everything, and a narrative of the pre-national or non-national polity is certainly possible, but it would have to be framed by historical questions that both limit and delimit a subject matter. In the case of Old Regime France I might refer to the work of David Bell, who was interested in the development of nationalism in the eighteenth century. Bell argues, convincingly, that the discourse and literary efforts to express national sentiment in the eighteenth century can be found in a range of projects, from the efforts to create canonical list of important Frenchmen, to the political propaganda efforts during the Seven Years War, in which the nature of Frenchness is stated in opposition to the English. Such constructions of national identity and national sentiment are nonetheless rather distinct from the project of nationalism itself, which is much more a project born of the republican moment in the French revolution, when there is a more resigned acceptance or a belief that the nation doesn’t really exist but needs to be constructed, needs to be built, needs to be imposed. In Bell’s argument, the model for doing this was actually the Catholic Church. To a great extent the republican project took its cue, as well as its institutional resources from ecclesiastical domains, and in particular in the work of creating a homogenous Catholic culture in a heterogeneous setting such as France. So, that’s one possible way of thinking about or of writing a narrative of the movement from a pre-national to a national polity, from a society in which there are certain indigenous expressions of a literary and symbolic character of national sentiment; these, however, are still not societies we could characterize as being societies born of nationalism or societies that are modern in the same sense of post-revolutionary or nineteenth century societies. But I think that more generally the answer is that we can write the kinds of histories that best answer the kinds of questions that we ask. To quote Lucien Febvre, “to ask a question is the beginning and end of all history.” It depends what questions we ask of these pre-national polities and in part, the answer is already contained in naming the thing, being that the key word in that term “pre-national polity” is polity and it is the political framework which helps us to frame the unity of the narrative. This is not to say that in the case of France, for example, we’re going back to a king-centered narrative or a court-centered narrative of national history. Politics or polity means much more than simply the exercise of sovereignty by an individual empowered through the office of kingship. It means a set of institutions; it means a set of competing discourses and contested discourses, in which part of our job as historians is to analyze, not so much the individual utterances, but to sketch out the framework in which such utterances are meaningful.

SG: It might be a much more complex procedure in the case of empires because just suggesting a political imperial framework for a historical narrative is a highly controversial statement, precisely because we live, at least theoretically, in a post-imperial age, where national identities seem to be well-established (at least they have established themselves in schools and other institutions producing knowledge). Suggesting an imperial unity would appear as an attack on these national identities.

PS: Well, I think that’s right, but I do think that good history is really capacious in the sense that it tries to incorporate different subject positions and perspectives. There are certainly ways of framing an imperial history in a cross-cutting fashion that would allow for the exploration of the relation between these different voices such that a history of a nineteenth century empire or of imperial Russia takes on the opposing distinction of official and vernacular expressions of identity. That might be one way of trying to aggregate some of the vast diversity and heterogeneity of subject positions in a pre-national polity, to contrast efforts at the imperial or national level to create what you call normative nationality, efforts to impose homogeneity, efforts to create linguistic and cultural unity and so on; and vernacular efforts — not necessarily just responses but vernacular appropriations of these efforts too – that contrast and oppose official frameworks. This might be one way of trying to be broad and comprehensive in writing a history of empire, but without the aspirations of doing “total history.”

SG: Much of our conversation revolved around the language and models of social sciences. Let me then pose a question about the language and models of social sciences head on. Scholars operate with the language and models of social sciences born out of national state experience in the late nineteenth century aimed at describing the realities of nation states. How adequate are our attempts to apply those categories in describing and analyzing the “under-nationalized” societies of the ancien régime or European old regimes that survived until the Great War? Where do we look for alternatives or more suitable concepts and terms? If it was “impossible to express” the revolutionary experience, is it possible to express the pre-revolutionary world in its own terms, or at least not ascribe to it the post-revolutionary tropes of “conservatism,” “tolerance,” “moderation,” etc? In that connection, how useful are qualified terms such as “ancient,” “citizenship,” or “imperial citizenship,” which sometimes appear contradicting the universalizing modern meanings of the terms?

PS: This again is a very important and difficult question. It reminds me of the many different kinds of debates that we have in European history about the early modern period. The early modern period itself contains within it the conundrum that you mentioned because, in characterizing it as early modern we are trying to both assert the modernity and thus the tangibility of modern concepts into the slightly more recent past but also, at the same time, emphasize its “early” features and therefore its distinguishing qualities which make it not yet a modern society. There are debates that have long droned on that address this conundrum, and simultaneously the problem you raised earlier, about indigenous categories and those of social analysis. The historiographical debate that began half a century ago about the nature of social hierarchy in the sixteenth and seventeenth century is a good example: was early modern society organized by “classes” (in the Marxian sense) or by “orders” (in a early modern corporatist sense)? This was not simply a neutral intellectual debate, but one deeply engaged in politics, including the way in which it was mapped institutionally. Thus Marxists, including and especially the Soviet historians like Porchnev,^[9] insisted on the existence of class in the early modern period, against a monarchist and reactionary position of some French historians (like Roland Mousnier at the Sorbonne),^[10] who insisted that class was irrelevant and that we could only accept the “emic” categories of the early modern period, which were categories based on the concept of order, of a state and of the regulated hierarchy that could be found in juridical texts. So, the question is not new, and that debate is now quite old, but it will not be soon put to bed. I think the important thing is that we’re much more attentive now than we ever were about the relation between words and things, and that is what has been beneficial about the so-called linguistic or post-structural turn in historical studies. Our attention has been drawn, by literary critics especially, to the very categories that we use to describe the past, and historians have become more self-conscious and more selective in the ways in which we make sense of the past. This is a fairly abstract answer, but in practice it works out, I think, rather nicely for historians, who, as students of any given period, have a pretty good intuitive sense of when a word or concept is being used in a consistent and determined way that doesn’t seem to fully describe or doesn’t adequately describe the regular patterns of history or of historical behavior; at that point, then, the language can be modified, it can be jettisoned. Historians are, unlike many social scientists, basically inductive thinkers and deep empiricists, and if words don’t work, then maybe there are problems with words, and maybe there is intellectual work to be done in trying to explain why they don’t work or in modifying them in ways that allow them to be descriptive and interpretive of different historical periods. Do I believe that one can qualify such universal and apparently monolithic and homogenizing terms like citizenship by talking about its ancient or its imperial iterations or its royal iterations or the problem of the subject-citizen? Absolutely. Do I believe that there are ways of talking about nationality even though the word doesn’t yet exist, the obverse side of the same problem? Absolutely. Language is, in this sense, so much a tool of our understanding and shouldn’t be held as the prison of it.

SG: Thank you very much. This was our last question and we’ll end on this very optimistic note, at least for historians.