Explanation/Understanding
It is at the level of explanation/understanding that the autonomy of history in relation to memory is affirmed most forcefully on the epistemological plane. In truth, this new phase of the historiographical operation was already implied in the preceding one insofar as there is no document without some question, nor some question without an explanatory project. It is in relation to explanation that the document is proof. Nevertheless, what explanation/understanding adds that is new in relation to the documentary treatment of the historical fact has to do with the modes of interconnectedness of the documented facts. To explain, generally speaking, is to answer the question “Why?” through a variety of uses of the connector “because.”1 In this respect, to the degree that we need to hold open the range of such uses, to the same degree we must keep the historiographical operation in the neighborhood of approaches common to every scientific discipline, characterized by recourse in different forms to modeling procedures subject to verification tests. In this way, model and documentary proof go hand in hand. Modeling is the work of the scientific imagination, as has been emphasized by R. G. Collingwood, Max Weber, and Raymond Aron in dealing with singular causal implication.2 This use of the imagination carries our minds far beyond the sphere of private and public memory into the range of the possible. If the intellect, however, is to remain within the domain of history, and not slip over into that of fiction, this use of the imagination must submit itself to a specific discipline, namely, an appropriate dividing up of its objects of reference.
This dividing up is governed by two guiding principles. According to the first of these, the explanatory models in use by historians have as a common feature that they relate to human reality as a social fact. In this respect, social history is not one sector among others, but rather the point of view from which history takes its stand, that of the social sciences. By privileging along with one school of contemporary history, as I shall do below, the practical issues of the constitution of the social bond and problems of identity attached to it, we lessen the distance that was opened during the first half of the twentieth century between history and the phenomenology of action, but without abolishing it. The human interactions, and in general the kinds of interval, of inter-esse as Hannah Arendt liked to put it, that occur between agents and recipients of human action only lend themselves to the modeling processes by which history inscribes itself among the social sciences at the price of a methodological objectification that has the value of an epistemological break in relation to memory and ordinary narration. In this regard, it is helpful to keep history and the phenomenology of action distinct for the greater benefit of their ongoing dialogue.
The second guiding principle concerns history’s place within the field of the social sciences. It is through the emphasis that history places on change and on the differences or intervals affecting such changes that it distinguishes itself from the other social sciences and principally from sociology. This distinctive feature is common to every department of history: economic reality, social phenomena in the strict sense of the term, practices and representations. This common feature defines the referent of historical discourse within the common referent of the social sciences as a limit function. Changes and differences or intervals have a clear temporal connotation. This is why we can speak of a long time span, of the short run, of a point-like event. The discourse of history can thus once again move closer to the phenomenology of memory. To be sure. However, the vocabulary of the historian constructing his hierarchy of time spans, like that of Labrousse and Braudel, or breaking them up, as has been done since Labrousse and Braudel, is not that of phenomenology referring to the lived experience of temporal duration, as was the case in the first part of this work. These time spans are constructed. Even when history ingeniously mixes up their order of priority, particularly in the case of reactions against the rigidity of the architecture of spans piled on one another, it is always in terms of multiple spans that the historian models lived time. Even if memory is the test of the variable depth of time and orders its memories in relation to one another, outlining in this way something like a hierarchy among them, it does not spontaneously form the idea of multiple time spans. This is rather the prerogative of what Halbwachs calls “historical memory,” a concept we shall return to when the time comes. The historian’s handling of this plurality of time spans is commanded by a correlation among three factors: the specific nature of the change considered—economic, institutional, political, cultural, or whatever; the scale with which it is apprehended, described, and explained; and, finally, the temporal rhythm appropriate to that scale. This is why the privilege accorded economic phenomena by Labrousse and Braudel and, following them, the historians of the Annales school had as its corollary the choice of the macroeconomic scale and of the long time span as regards its temporal rhythm. This correlation is the most marked epistemological feature of history’s treatment of the temporal dimension of social action. It is reinforced by a supplementary correlation between the specific nature of the social phenomenon taken as referent and the type of privileged document. What the long time span structures on the temporal plane is the priority of series of repeatable facts, rather than singular events likely to be remembered in a distinctive way. In this sense, these facts are open to quantification and to being dealt with mathematically. With serial and quantitative history,3 we distance ourselves as much as possible from Bergson’s or Bachelard’s temporal duration. We are in a constructed time, made of structured and quantified durations. It is with regard to these audacious structuring operations, which marked the middle of the twentieth century, that the more recent history of practices and representations has elaborated a more qualitative treatment of durations and thus appears to have redirected history back in the direction of the phenomenology of action and the phenomenology of temporal duration united with it. But for all that this history does not deny the objectifying stance that it continues to share with the more notable efforts of the Annales school.
Having said this, as regards the referents of historical explanation, it remains to characterize in a more precise way the nature of the operations related to explanation. I have mentioned the eventual diversity of uses of “because” connected to the answers to the question “why?” Here is where we must insist on the variety of types of explanation in history.4 In this regard, we can say without injustice that there is no one privileged mode of explanation in history.5 This is a feature that history shares with the theory of action to the degree that the penultimate referent of historical discourse is those interactions capable of engendering the social bond. It is not surprising therefore that history unfolds the full range of modes of explanation likely to make human interactions intelligible. On the one side, the series of repeatable facts of quantitative history lend themselves to a causal analysis and to the establishing of regularities that draw the idea of a cause, in the sense of efficacy, toward that of lawfulness, toward the model of the “if . . . then” relation. On the other side, the behavior of social agents, responding to the pressure of social norms by diverse maneuvers of negotiation, justification, or denunciation, draw the idea of a cause toward the side of explanation in terms of “reasons for . . .”6 But these are limit cases. The great mass of historical works unfold in a middle region where disparate modes of explanation alternate and sometimes combine in an unpredictable way. I have titled this chapter “Explanation/Understanding” in order to make sense of this variety of historical explanations. In this regard, we can take the quarrel that arose at the beginning of the twentieth century around the terms “explanation” and “understanding,” taken as antagonistic to each other, as surpassed. Max Weber, in combining explanation and understanding from the start, was perspicacious in elaborating the leading concepts of his social theory.7 More recently G. H. von Wright, in Explanation and Understanding, constructed a mixed model of explanation for history that made causal (in the sense of law-like regularity) and teleological (in the sense of motivations capable of being rationalized) segments alternate.8 In this regard, the correlation mentioned above between the type of social fact taken as determining the scale of description and reading, and temporal rhythm can offer a good guide in the exploration of differentiated models of explanation in their relation to understanding. The reader may be surprised not to see the notion of interpretation appear in this context. Did it not stand alongside that of understanding in the great age of the quarrel between Verstehen and Erklären? Was not interpretation held by Dilthey to be a special form of understanding linked to writing and in general to the phenomenon of inscription? Far from objecting to the importance of the notion of interpretation, I propose to give it a much broader sphere of application than did Dilthey. For me, there is interpretation at all three levels of historical discourse: at the documentary level, at the level of explanation/understanding, and at the level of the literary representation of the past. In this sense, interpretation is a feature of the search for truth in history that runs across these three levels. Interpretation is a component of the very intending of truth in all the historiographical operations. I shall deal with it in part 3 of this work.
A final lexical and semantic comment at the threshold of this chapter: The reader may be more surprised by my silence about the narrative dimension of historical discourse than my silence about the theme of interpretation within the explanation/understanding framework. I have deliberately put off its consideration, leaving it to the framework of the third operation of the historiographical operation, the literary representation of the past, to which I shall accord an importance equal to that of the other two operations. This is not to say that I take back anything learned from the discussion in the three volumes of my Time and Narrative. But, in reclassifying narrativity in the way we are going to discuss, I want to leave to the end one misunderstanding suggested by the upholders of the narrativist school and taken for granted by its detractors, the misunderstanding that the configuring act that characterizes emplotment would as such constitute an alternative in principle to causal explanation.9 Louis O. Mink’s convincing argument, which I continue to respect, seems to me compromised by the imposition of this unfortunate disjunction. The cognitive function of narrativity seems to me, taking everything into account, better recognized if it is linked to the phase of historical discourse representative of the past. Our problem will be to understand how the configuring act of emplotment gets articulated through the modes of explanation/understanding placed in service of the representation of the past. To the extent that representation is not a copy, a passive mimesis, narrativity will suffer no diminutio capitis from being associated with the properly literary moment of the historiographical operation.
This chapter is constructed on the basis of one particular working hypothesis. I propose to examine the kind of intelligibility proper to explanation/understanding in terms of a class of objects of the historiographical operation, namely, representations. This chapter therefore pairs a method and an object. The reason for this is as follows. The notion of representation and its rich polysemy runs through this whole work. It first brought to light the perplexities of the phenomenology of memory starting from the Greek problem of the eikōn. And it will reappear in the following chapter in terms of the historiographical operation itself in the form of the written representation of the past (the writing of history in the narrow sense of this term). The notion of representation also figures two times in the epistemological portion of this work: as the privileged object of explanation/understanding, and as the historiographical operation. At end of this chapter I shall propose a confrontation between these two uses that are here made of the notion of representation.
In this chapter, the object-representation thus plays the role of privileged referent, alongside the economic, the social, the political. This referent is picked out from the much vaster field of social change, taken as the overall object of historical discourse. This is the point of this chapter.
But before reaching that stage of the discussion, three steps must be taken.
In the first section, I propose a quick review of the important moments in French historiography during the first two-thirds of the twentieth century, up to the period qualified as that of a crisis by commentators, whether historians or others. Within this chronological framework, which is essentially structured by the great adventure of the French Annales school and dominated by the overarching figure of Fernand Braudel, I shall bring to the fore the questions of method and of the promotion of a privileged object, for a long time now known by the name mentalités, the term having first been introduced in sociology by Lucien Lévy-Bruhl with the phrase “primitive mentality”10 (section 1: “Promoting the History of Mentalities”).
We shall follow this double inquiry to the point where the crisis in method is matched by a crisis in the history of mentalities, which has continued to suffer from its debatable origin in the sociology of “primitive mentality.”
I shall interrupt this double inquiry to consider three authors—Foucault, Certeau, Elias—whom I shall present as “advocates of rigor” from whom I shall seek help in characterizing in a new the way the history of mentalities as a new approach to the total phenomenon and at the same time to a new object of historiography. By way of these considerations, the reader will become accustomed to associating the notion of mentalities with that of representation, as a way of preparing the moment when at the end the latter will be substituted for the former, thanks to its conjunction with the notions of action and of agent (section 2: “Some Advocates of Rigor: Michel Foucault, Michel de Certeau, Norbert Elias”).
This substitution will be prepared through a long intermediary section devoted to the notion of scale. Although one does not see the same things in microhistory, this variety of history illustrated by the Italian microstorie will provide the occasion for varying the approach to mentalities and representations as a function of an “interplay of scales.” Just as macrohistory is attentive to the weight of structural constraints exercised over the long time span, to a similar degree microhistory is attentive to the initiative and capacity for negotiation of historical agents in situations marked by uncertainty.
The step will then be taken from the idea of mentalities to that of representations in the wake of this notion of variations in scale and within the framework of a new overall approach to the history of societies, the one proposed by Bernard Lepetit in Les Formes de l’expérience. There the accent will be found to be on social practices and the representations integrated into these practices, the representations figuring as the symbolic component in the structuring of the social bond and the identities that are at stake within it. We shall pay particular attention to the connection between the operation of such representations and the different sorts of scales applicable to social phenomena: a scale of efficacy and coercion; one of standing within public esteem; and one of time spans embedded within one another (section 3: “Variations in Scale”).
I shall end with a critical note in which we shall draw further upon the polysemy of the term “representation” to justify the split between represented object [représentation-objet] and the operation of representing [représentation-opération], to be taken up in the next chapter. The name Louis Marin will appear for the first time in the closing pages of this chapter, where the adventures of explanation/understanding will have been parsed in terms of the history of mentalities become the history of representations (section 4: “From the Idea of Mentality to That of Representation”).
§
PROMOTING THE HISTORY OF MENTALITIES
I have chosen from the immense literature having to do with explanation in history what concerns the emergence, then the consolidation and renewal of what in turn or in an alternative manner has been called cultural history, history of mentalities, and finally history of representations. I shall explain below why, upon reflection, I adopted this latter name. In the present section, I propose to comment upon the choice of this trajectory, not being already able to justify it. The notion of mentality represents, in fact, one that is particularly vulnerable to criticism owing to its lack of clarity and distinctness, or, if one is charitable, its overdetermination. The reasons why it imposed itself on historians are thus all the more worthy of interest.
As for what concerns me, these reasons run as follows.
First, staying as close as possible to professional historians, what interested me was the progressive promotion of one of those new “objects” that recent history has made a fuss about, to the point of becoming what I shall call a pertinent object, in other words, an object of immediate reference for all the discourse that relates to it. With this promotion comes a redistribution of values of importance,11 of degrees of pertinence, that affect the ranking of economic, social, and political phenomena within the scale of importance and finally the scale adopted by the historical gaze in terms of micro- and macrohistory. This displacement on the plane of objects of reference, of immediate pertinence, goes with a displacement on the plane of methods and of modes of explanation. The concepts of singularity (of individuals or events), repeatability, serial ordering, are particularly affected; even more so are those of a collective constraint and correlatively of reception, passive or otherwise, on the part of social agents. This is why we shall see such new notions as appropriation and negotiation appear at the end of our comments.
Taking a step back in relation to the historian’s work, I wanted to verify the thesis that history, as one of the social sciences, does not disregard its discipline of distantiation in relation to lived experience, to collective memory, once it declares itself to have moved away from what is called, most often wrongly, positivism or more equitably “historicizing history” to characterize the age of Seignobos and Langlois at the beginning of the twentieth century. We might think that with this “new object” history moves closer—whether it knows it or not—to phenomenology, in particular to the phenomenology of action or, as I like to put it, to that of the acting and suffering human being. Despite this shrinking of distance, the history of mentalities and of representations nonetheless remains situated on the other side of the epistemological break that separates it from the outcome of the phenomenology practiced in the first part of this work devoted to memory, and especially to collective memory insofar as memory constitutes one of the powers of that being I call the capable human being. The most recent developments in the history of representations do approach this phenomenology in that the objective posture of this history allows for notions akin to that of “can”—can do, can say, can recount, can impute the origin of one’s actions to oneself. As a result, the dialogue between the history of representations and the hermeneutics of acting will turn out to be sharper edged owing to the fact that the invisible threshold of historical knowledge will not actually have been crossed.
However, there is a more subtle reason for my interest in the history of mentalities or of representations, a reason that will grow to the point of affecting the end of this investigation. Anticipating the last section of this chapter, I confess that this reason definitively imposed itself at the moment when, for reasons we shall speak of, the notion of representation was preferred to that of mentality. A case then came to the fore no longer of confusion or of indistinctness, but rather of overdetermination. As it turns out—and it will be necessary to show that this is not the result of some semantic contingency, of a regrettable homonymy resulting from a poverty or parsimony of vocabulary—the word “representation” figures in this work in three different contexts. It first designates the great enigma of memory, in relation to the Greek problem of the eikōn and its embarrassing doublet of phantasma or phantasia. We have said and repeated that the mnemonic phenomenon consists in the presence to the mind of an absent thing that, furthermore, no longer is but once was. If it is simply evoked as a presence, and in this sense as pathos, or if it is actively sought out in the operation of recalling that concludes in the experience of recognition, what is remembered is a representation, a re-presentation.
This category of representation appears a second time in the framework of the theory of history as the third phase of the historiographical operation, once the historian’s labor, begun in the archives, ends in the publication of a book or an article to be read. The writing of history becomes literary writing. An embarrassing question then invades the intellectual space thereby opened: how does the historical operation preserve, even crown at this stage, the ambition for truth by which history distinguishes itself from memory and eventually confronts the latter’s avowal of trustworthiness. More precisely: how does history, in its literary writing, succeed in distinguishing itself from fiction? To pose this question is to ask how history remains or rather becomes a representation of the past, something fiction is not, at least in intention, even if it may be so as a kind of added value. In this way, in its last stage historiography repeats the enigma raised by memory in its first one. Historiography repeats and enriches it through all the conquests I have placed globally under the aegis of the myth from the Phaedrus under the sign of writing. The question will then be to know whether historical representation of the past will have resolved, or simply transposed, the aporias linked to its mnemonic representation. It is in relation to these two major occurrences that it will be necessary to situate the use of the term “representation” by historians, at least as regards its conceptual aspect. Between the mnemonic representation from the beginning of our discourse and the literary representation situated at the end of the trajectory of the historiographical operation, representation presents itself as an object, a referent, of the historian’s discourse. Can it be that the object represented by historians bears the mark of the initial enigma of mnemonic representation and anticipates the final enigma of the historical representation of the past?
I shall limit myself in the remainder of this section to a brief summary of the leading moments of the history of mentalities since the founding of the French Annales school up to the period qualified as one of crisis by observers, be they historians or others. We shall deliberately interrupt this rapid overview and consider three major undertakings that, if they cannot be confined within the strict limits of the history of mentalities and of representations, have addressed to the human sciences a demand for rigor concerning which we will have to ask whether subsequent history has given an answer, and, more generally, if a history of representations is capable of doing so.
It is the first generation of the Annales school, that of the founders, Lucien Febvre and Marc Bloch, that is worth questioning first of all, not only because the foundation of the journal of the same name in 1929 marks an important date, but because the notion of mentality was clothed in the founders’ works with an importance that would not be equaled in the next generation, during the transitional period marked by Ernest Labrousse and more so by Fernand Braudel. This feature is all the more noteworthy in that Annales d’histoire économique et sociale—the journal’s initial title—was marked by a shift in interest away from politics toward economics and by a strong rejection of history in the fashion of Seignobos and Langlois, improperly called positivist, at the risk of confusing it with the Comtean heritage, and less unjustly called historicizing in virtue of its dependence on the German school of Leopold von Ranke. Singularity, whether of the event or of individuals, was set aside along with chronology marked out by narration and politics as the privileged site of intelligibility. One set out in search of regularity, fixity, permanence, on the model of geography, brought to a high point by Vidal de La Blache, and on that of Claude Bernard’s work in experimental medicine. To the supposed passivity of the historian confronted with a collection of facts, one opposed the active intervention of the historian facing the document in an archive.12 When Lucien Febvre borrowed from Lévy-Bruhl the concept of mentality, it was to give a particular history, having to do with historical biography, the background of what he called “mental tools.”13 In so generalizing the concept of mentality beyond what was still called “primitive mentality,” he killed two birds with one stone. He enlarged the sphere of historical inquiry beyond economics and especially beyond politics, and he gave the reply of a history anchored in the social to the history of ideas practiced by philosophers and by most historians of science. In this sense, the history of mentalities for a long time plowed a furrow between economic history and the dehistoricized history of ideas.14
In 1929, Febvre had already published his Luther (1928) to which he would add his Rabelais and his Amour sacré, amour profane: Autour de l’«Heptameron.»15 Beneath their biographical appearance, these three books pose a problem that will spring up again in another form when history comes to question its capacity to represent the past, that is, the problem of the limits of representation. Confronted with the problem of unbelief in the sixteenth century, Febvre established in a convincing manner that the believable available to the period (not Febvre’s phrasing), its “mental tools,” did not permit professing or even forming an openly atheistic vision of the world. What the history of mentalities proposed to demonstrate, leaving indeterminate the question what one was meant to think of by means of “mental tools,” was what a person of the time could and could not think about the world. Was the collective so undifferentiated as the notion of mental tools seems to imply? Here the historian gave credit to the psychology of a C. Blondel and to Lévy-Bruhl’s and Durkheim’s sociology.
In The Royal Touch: Sacred Monarchy and Scrofula in England and France (1924) and Feudal Society (1939), Marc Bloch ran into a comparable problem: Howcould rumor, the false news of the capacity of kings to cure scrofula, spread and impose itself unless with the help of a quasi-religious devotion as regards royalty?16 What had to be presumed, even while guarding against anachronism, was the force of a specific mental structure, the “feudal mentality.” In contrast to the history of ideas, not rooted in any social ground, history had to make a place for a deliberately historical treatment of “ways of feeling and thinking.” What was important were the collective, symbolic practices, the unperceived mental representations, of different social groups, to the point that Febvre could worry about the effacing of the individual in Bloch’s approach to the problem.
Between society and the individual, the interplay that Norbert Elias calls civilization was not measured by the same yardstick by the two founders of this school. The imprint of Durkheim was deeper on Bloch, while attention to the aspirations toward individuality among Renaissance figures influenced Febvre.17 But what united them was, on the one hand, the assurance that the facts of civilizations stand out against the background of social history, and, on the other, the attention paid to the relations of interdependence among the spheres of activity of a society, attention that freed them from getting caught up in the impasse of the relations between an infrastructure and a superstructure, as in Marxist approaches. Above all there was the confidence in the federative power of history in regard to the neighboring social sciences: sociology, ethnology, psychology, literary studies, linguistics. “The average man according to the Annales,” as François Dosse names him, a social human being, is not eternal man, but rather a historically dated, anthropocentric, humanistic figure inherited from the Enlightenment, the same one Michel Foucault will lambaste.18 But whatever objections one may oppose to this worldview, which stems from the inseparable interpretation of truth in history, we can legitimately inquire, at this stage of our own discourse, what are the internal articulations of these evolving mental structures, and above all how the social pressure they exercise on social agents is received or undergone. The sociological or psychological determinism of the Annales school during the period of its dominance will only really be called into question when history reflecting upon itself will have made problematic the dialectic between the top and bottom of societies at the point of the exercising of power.
Following World War I, the Annales school (and its journal now subtitled Économies, sociétés, civilisations) was best known for its preference for taking the economy as its privileged referent. The use of quantification applied to repeatable facts, to series, treated statistically with the help of the computer, went along with this initial preference. The humanism of the first generation almost seemed repressed by the reverence for social and economic forces. And Claude Lévi-Strauss’s structuralism operated both as an encouragement and a competitor.19 It was then necessary to oppose to the invariants of the dominant sociology those structures that remained historical, that is, changing. This was exemplified by the long time span, placed by Braudel at the base of a pyramid of time spans following a schema that recalled Ernest Labrousse’s triad: structure, conjuncture, event. The time thereby given the place of honor was conjoined with the space of the geographers, whose own permanence helped to slow down the flows of time. The horror Braudel felt for the event is too well known to require emphasis here.20 What remains problematic is the relationship between these temporalities which accumulate and stack up more than they are dialectically related to one another, following an empirical pluralism deliberately removed from any abstract speculation, unlike George Gurvitch’s careful reconstruction of the multiplicity of social times. This conceptual weakness of the Braudelian model was only really taken up when the question of the variation in the scales considered by the historian was taken into account. In this regard, the reference to total history, inherited from the founders and forcefully reiterated by their successors, only allows for a prudent recommendation, that of professing interdependence there where others, the Marxists at their head, thought to discern linear, horizontal or vertical dependencies among the components of the social bond. These relations of interdependence could be problematized for themselves only once the preference for the long time span was clearly assigned to a choice, which up to then had remained unmotivated, for macrohistory, on the model of economic relations.
This coalition between the long time span and macrohistory governed the contribution of the second generation of the Annales school to the history of mentalities. Here another triad than that of hierarchical time spans has to be taken into account: that of the economic, the social, and the cultural. But the third stage of this three stage rocket, to use the apt phrasing of Pierre Chaunu, the advocate of serial and quantitative history, obeys the rules of the method of correlation governing the choice for the long time span no less than do the first two stages. The same primacy accorded to repeatable, serial, quantifiable facts holds for the mental as for the economic and the social. And it is the same fatalism inspired by the spectacle of the inexorable pressure of economic forces, and confirmed by that of the permanence of geographical inhabited spaces, that leads to a vision of a humanity overwhelmed by greater forces than its own, as can be seen in Braudel’s other great work, Civilization and Capitalism, 15th–18th Century (1979).21 Are we so far from Max Weber’s iron cage? Did not the focus on economics hinder the unfolding of this third stage, as seems suggested by Braudel’s reticence regarding Weber’s thesis about the Protestant ethic and the spirit of capitalism? Was not the dream of federating history with regard to the neighboring social sciences realized solely thanks to an anthropology intimidated by structuralism, despite its vow to historicize it? At the very least, up to his retirement and to the time of his death, Braudel continued forcefully to oppose the demand for a total history to its threatened dispersion.
In the review the journal Annales made of its first fifty years in 1979, the editors recall that the community gathered around it had wanted to propose “more a program than a theory,” but recognized that the multiplicity of objects submitted to an ever more specialized, more technical research risked making reappear “the temptation of a cumulative history, where the acquired results count more than the questions posed.”22 Jacques Revel confronts this risk in his article “Histoire et science sociale, les paradigms des Annales,” which follows that of A. Burguière.23 What, he asks, is “the unity of an intellectual movement that has endured for a half century”? “What is there in common between the highly unified program of the first years and the apparent bursting apart of more recent orientations?” Revel prefers to speak of particular paradigms that succeed rather than eliminate one another. The refusal of abstraction, the plea for the concrete against the schematic, makes the formulation of these paradigms difficult. The first thing to impose itself is the relative economic and social dominance of the first years of the journal, without the social ever becoming “the object of a systematic, articulated conceptualization.” “It is rather the place for an always open inventory of correspondences, of relations that ground the interdependence of phenomena.” It is easier to see the ambition to organize the social sciences, including sociology and psychology, around history, and the resistance to “the sometimes terrorist antihistoricism” fomented by the reading of Claude Lévi-Strauss’s Tristes Tropiques (1955) and Structural Anthropology (1958), than the conceptual structure that undergirds this ambition and this resistance.24 This is why the interplay of continuities and even more of discontinuities is so difficult to outline. We do not know exactly what “constellation of knowledge has come undone before our eyes over the past twenty years.” Is humanity by itself, if we may put it this way, the federating theme “of a particular ordering of scientific discourse,” such that we may assign to the effacing of this transitory object the subsequent fragmentation of the field of inquiry? Revel clearly has in mind discourse about the bursting apart of history, maybe even François Dosse’s talk of a “history in pieces.” He upholds the refusal and the conviction attached to the claim for a global or total history; the refusal of partitions, the conviction of coherence and convergence. But he cannot hide his unease: “It seems as though the program of global history offers only a neutral framework for the addition of particular histories whose ordering is not a problem.” Whence the question: “History burst apart or history under construction?” Revel does not answer.
And in this conceptual mishmash what becomes of the history of mentalities, which this summary inventory does not name (any more than it does the other main branches of the tree of history)?
Confronted with these questions and doubts, a few historians have known how to keep their focus on the question of intelligibility within the region of the history of mentalities, even if it means placing it under different patronage. This is the case with Robert Mandrou, all of whose work is placed under the heading of “historical psychology.”25 He is the one assigned in the Encyclopedia Universalis with the defense and illustration of the history of mentalities.26 Mandrou defines its object in the following way: “It takes as its objective the reconstitution of behaviors, expressions, and silences that express conceptions of the world and collective sensibilities; representations and images, myths and values, acknowledged or affecting groups or society as a whole, and that constitute the contents of collective psychology, provide the fundamental elements for this study.” (We can see here the equation between mentality for French-speaking authors and what Germans call Weltanschauung, which our concept of mentality is meant to translate.) As for method, the historical psychology that Mandrou himself practices is applied to narrowly defined operative concepts: worldviews, structures, conjunctures. On the one side, worldviews have their own kind of coherence; on the other, a certain structural continuity confers on them a noteworthy stability. Finally, short and long rhythms and fluctuations mark their encounters. In this way, Mandrou presents himself as a historian of the collective mind who gives the most credibility to the intelligibility of the history of mentalities, following a conceptuality that recalls that of Ernest Labrousse (structure, conjuncture, event)—and the least credit to a psychoanalytic rewriting of collective psychology, in contrast to Michel de Certeau.
It was also at the margins of the Annales school that Jean-Pierre Vernant in 1965 first published his major book Myth and Thought among the Greeks, which has been reprinted a number of times, and which he too subtitles a study in historical psychology, placing it under the patronage of the psychologist Ignace Meyerson (to whom the work is dedicated) and associating it with the work of another Hellenist, Louis Gernet.27 What is at issue are studies devoted to the inner history of Greek man, his mental organization, the changes that from the eighth to fourth century B.C. “affect the entire framework of thought and the whole gamut of psychological functions: the modes of symbolic expression, and the manipulation of signs, ideas of time and space, causality, memory, imagination, the organization of action, will, and personality” (xi). Twenty years later he will acknowledge his kinship with the structural analysis applied to different myths or mythic groupings by other scholars, including Marcel Détienne, with whom he published Cunning Intelligence in Greek Culture and Society.28 And the work he published with Pierre Vidal-Naquet, Myth and Tragedy in Ancient Greece, incontestably bears the same imprint.29 It is worth noting that Vernant does not break with the humanism of the first generation of the Annales school. What is important to him in the final analysis is the sinuous trail leading from myth to reason. As in Myth and Thought, it is a question of demonstrating “how, by way of the older tragedy of the fifth century, were outlined the first, still hesitant sketches of man as agent, as the master of and responsible for his acts, as the possessor of a will” (7). Vernant emphasizes: “from myth to reason: these were the two poles between which, in a panoramic view, the destiny of Greek thought seemed to play itself out by the end of this book” (7), without the foreignness of this form of thinking being overlooked, as the study on “the avatars of that particular, typically Greek form, crafty thinking, which is made up of cunning, cleverness, craftiness, deceit, and resourcefulness of all kinds,” the Greeks’ mētis, which “stems exclusively neither from myth, nor from reason.”
However, the main tendency of the history of mentalities for the Annales school was to turn toward a less certain defense of its right to existence, beginning with the second generation, that of Labrousse and Braudel, and even more so at the time of the so-called “new history.” On the one hand, we have the spectacle of a loss of focus, which led to the talk of a burst-apart history, even of a history in little pieces; on the other, thanks to this same dispersion, there was a certain upturn. It was in this sense that the history of mentalities figured as a whole among the “new objects” of the “new history,” in volume three of the collection edited by Jacques Le Goff and Pierre Nora, Faire de l’histoire.30 Alongside “new problems” (volume 1) and “new approaches” (volume 2), the history of mentalities freed itself at the very moment when the project of total history was fading. For some, a passion for the long time span and quantitative studies coming from the older commitment to economic history remained, at the price of effacing the figure of the human being of the humanism that was still celebrated by Bloch and Febvre. In particular, the history of climate provided its measures and strategies to this “history without men.”31 This tenacious attachment to serial history by contrast makes the conceptual fuzziness of the notion of mentality reappear among those who accept the patronage of this special kind of history. In this regard, Jacques Le Goff’s presentation of this “new object” is more discouraging for the rigorous-minded than were the earlier summaries by Duby and Mandrou. The increasing importance of this topos, announcing its eventual disappearance, is greeted by a disturbing phrase from Marcel Proust: “I like ‘mentality.’ There are a lot of new words like that which people suddenly start using, but they never last.”32 That the expression refers to a scientific reality, that it contains a conceptual coherence, remains problematic. The critic would like to believe so; nevertheless, its very imprecision recommends it for speaking of what is “beyond history”—by which we can understand is meant economic and social history. The history of mentalities thus offers a “change of scenery . . . to those intoxicated by economic and social history and above all a vulgar Marxism” by transporting them to this “elsewhere,” to what were mentalities. And in this way one satisfies Michelet’s expectation of rendering a face to “the resuscitated living-dead.” At the same time, one links up again with Bloch and Febvre. One modulates things in terms of epochs, settings, in the manner of ethnologists and sociologists. If one speaks of archeology, it is not in the sense of Foucault, but in terms of the ordinary sense of stratigraphy. As for their operating, mentalities function automatically, without their bearers being aware of them. They are less well-formed, professed thoughts than commonplaces, more or less worn out heritages, worldviews inscribed in what one can risk calling the collective unconscious. If the history of mentalities could for a period of time merit its place among the “new objects,” it was owing to an enlarging of the documentary sphere whereby, on the one hand, every trace became the collective witness of an age, and, on the other, every document concerning behavior marked a gap in relation to the common mentality. This oscillation between the common and the marginal, thanks to the discordances denouncing the absence of the contemporaneousness of contemporaries, could seem to justify recourse to the category of mentalities despite its semantic fuzziness. But then it was not the history of mentalities, as such, that was to have been treated as a new object, but the themes thrown together in the third volume of Faire de l’histoire: from climate to festivals in passing through the book and the body,33 and those that are not named, the large-scale affects of private life,34 without forgetting young women and death.35
This inscribing of the notion of mentality among the “new objects” of history at the price of the expansion I have just spoken of was not tenable. The deep reason for the rejection inflicted upon it does not come down to the objection of semantic fuzziness. It has to do with a more serious confusion, namely, the uncertain simultaneous use of the notion as an object of inquiry, as a dimension of the social bond distinct from the economic and the political, and as a means of explanation. This confusion is to be attributed to the heritage of Lucien Lévy-Bruhl and his concept of “primitive mentality.” By primitive mentality one explained what were irrational beliefs from the perspective of scientific and logical rationality. One thought to get beyond this prejudice on the part of the observer, which Lévy-Bruhl himself had begun to criticize in his Notebooks published in 1949,36 by applying the notion of mentality to ways of thinking or sets of beliefs belonging to groups or whole societies that were sufficiently distinct so that it could be used both as a descriptive and as an explanatory feature. One thought that what counted as a distinctive feature was not the content of some actual discourse but an implicit note, an underlying system of belief. But, in dealing with the idea of mentality as both a descriptive feature and a principle of explanation, one did not really get out of the orbit of the concept of primitive mentality dating from sociology at the beginning of the twentieth century.
It was this impure mixture that Geoffrey E. R. Lloyd undertook to dissolve in unpitying fashion in a book that had a devastating effect, titled Demystifying Mentalities.37 Lloyd’s argument is simple and direct. The concept of mentality is useless and harmful. It is useless at the level of description, harmful at that of explanation. It had served Lévy-Bruhl for describing prelogical and mystical features, such as the idea of participation, assigned to “primitives.” It serves contemporary historians for describing and explaining divergent or dissonant modes of belief from an age in which today’s observer does not recognize his conception of the world. It is for a logical, coherent, scientific observer that such past beliefs, or even those of the present, seem enigmatic or paradoxical, if not frankly absurd. Everything prescientific and still unscientific falls under this description. It is a construction of the observer projected on the worldview of the actors in question.38 This is when the concept of mentality shifts from description to explanation and from being useless to being harmful, inasmuch as it dispenses with having to reconstruct the contexts and circumstances that surrounded the appearance of the “explicit categories we commonly use in our highly value-laden descriptions—science, myth, magic and the opposition between the literal and the metaphorical” (7). Next Lloyd’s work was devoted to a telling reconstruction of the contexts and circumstances for the appearance of the categories of a rational and scientific observer, principally in the age of classical Greece, but also in China. The conquest of the distinction between the prescientific (magic and myth) and the scientific is the object of close analysis, centered principally on the political conditions and the rhetorical resources for the public use of speech in polemical contexts. One will recognize here an attack on problems comparable to that of Jean-Pierre Vernant, Pierre Vidal-Naquet, and Marcel Détienne.39 The alleged unspoken and implict something that the concept of mentality is supposed to thematize in a global, indiscriminate manner dissolves into a complex network of gradual, circumstantial acquisitions.
For all that, does Lloyd put an end to mentalities? Yes, assuredly, if it is a question of a lazy mode of explanation. But the answer has to be more circumspect if it is a question of a heuristic concept applied to what within a system of beliefs cannot be resolved into the contents of that discourse. Proof of this is the recourse Lloyd himself makes to the concept of a “style of inquiry” in his reconstruction of the Greek mode of rationality.40 So it is less a question of the “distinctive or striking peculiarities in patterns of discourse or reasoning, or again in the implicit beliefs that are inferred to underlie modes of behaviour [for the observer]” (4), than of what we might call the available belief structure of an era. To be sure, it is in relation to the observer that this belief structure is defined, but it is with regard to the actors that it is available. It was in this sense that Lucien Febvre could affirm that straightforward atheism was not a concept of belief available for a person of the sixteenth century. It is not the irrational, pre-scientific, prelogical character of a belief that is thereby pointed out, but its differential, distinctive character on the plane of what Lloyd calls precisely a “style of inquiry.” The notion of mentality is thus brought back to its status as a “new object” of historical discourse in the space left open by economics, the social, and the political. It is an explicandum, not a lazy principle of explanation. If we conclude that the heritage of the inadequate concept of “primitive mentality” remains the original sin of the concept of mentality, then it would be better indeed to give it up and prefer to it that of representation.
I propose a difficult conquest of the right to proceed to this semantic substitution, first by sending it to school with some strict teachers, then by proposing a detour through an intermediary concept, that of scale and of “changes in scale.”
SOME ADVOCATES OF RIGOR: MICHEL FOUCAULT, MICHEL DE CERTEAU, NORBERT ELIAS
I do not want to hand over the Labroussian and Braudelian models of the history of mentalities and representations to the criticism of a more recent historiography without having listened to three voices, two of whom come from outside historiography stricto sensu, but all of whom have raised to a previously unheard level the radicality of the discussion taking place throughout the human sciences. These are, on the one side, Michel Foucault’s plea for a science said to be without precedent, called the archeology of knowledge, and, on the other, Norbert Elias’s plea for a science of social formations that believes itself to be the enemy of history but that unfolds in an imperious fashion in a frankly historical way. Between them is Michel de Certeau, the inside outsider.
It is worth the effort of placing Foucault and Elias together in order to maintain the pressure of a demand for rigor directed against the discourse of professional historians rebelling against the favored model of the Annales school.
Foucault’s Archeology of Knowledge intervenes at the moment where the theory of the archive yields its place to that of archeology.41 He describes this turning point as an inversion in approach. Following the regressive analysis leading from discursive formations to bare statements comes the moment of turning back toward possible domains of application, without it for all that being a question of a repetition of the starting point.
It is first of all on the occasion of its confrontation with the history of ideas that archeology opens its way. It is against a discipline that has not been able to find its voice that it means to oppose its harsh schooling. Sometimes the history of ideas “recounts the by-ways and margins of history” (136)—alchemy and animal spirits, almanacs and other fluctuating languages; sometimes it is “the reconstitution of developments in the linear form of history” (137). Once again, denials abound: neither interpretation nor reconstruction of continuities, nor focusing on the meaning of works in a psychological, sociological, or anthropological manner. In short, archeology does not seek to reconstitute the past, to repeat what has been. But what does it want and what can it do? “It is nothing more than a rewriting: that is, in the preserved form of exteriority, a regulated transformation of what has already been written” (140). All right, but what does that mean? The descriptive capacity of archeology plays out on four fronts: novelty, contradiction, comparison, and transformation. On the first front, it arbitrates between the original, which is not the origin but a breaking point with the already said, and the regular, which is not the other of what is deviant but the piling up of the already said. The regularity of discursive practices takes its bearings from analogies that assure “enuciative homogeneity” and from hierarchies that structure these utterances and allow establishing derivation trees, as we see in linguistics with Propp and in natural history with Linnaeus.
On the second front, it credits the role of coherence in the history of ideas to the point of holding this as “a heuristic rule, a procedural obligation, almost a moral constraint of research.” Of course, this coherence is the result of research, not its presupposition, but it holds as an optimum: “the greatest possible number of contradictions resolved by the simplest means” (149). But it remains that contradictions are objects to be described for themselves, where we find gaps, dissension, defects in discourse. On the third front, archeology becomes interdiscursive, without falling into a confrontation between worldviews. In this regard, the competition among general grammar, natural history, and the analysis of wealth in The Order of Things demonstrated this comparison at work, apart from the ideas of expression, reflection, or influence.42 There is no hermeneutic of intentions and motivations, only a listing of specific forms of articulation. Archeology plays out its destiny on the fourth front, that of changes and transformations. Foucault is not taken in by either the quasi synchrony of immobile thoughts—an indicator of Eleaticism—nor by the linear succession of events—an indicator of historicism! What stands out is the theme of discontinuity, with ruptures, fault lines, gaps, sudden redistributions, which he opposes to “the practice of the historians of ideas” (170) who are overly concerned with continuities, transitions, anticipations, preliminary sketches. Here is the high point of archeology. If there is a paradox to it, it is not that it multiplies differences, but that it refuses to reduce them, in this way inverting the habitual values. “For the history of ideas, the appearance of difference indicates an error, or a trap; instead of examining it, the clever historian must try to reduce it. . . . Archeology, on the other hand, takes as the object of its description what is usually regarded as an obstacle: its aim is not to overcome differences but to analyze them, to say what exactly they consist of, to differentiate them” (171). In truth, it is the very idea of change, too marked by that of an active force, that must be renounced to the benefit of the idea of transformation, perfectly neutral in relation to the great metaphorics of flow. Should we reproach Foucault for having substituted for the ideology of the continuous one of the discontinuous? He readily returns the compliment.43 But it is the lesson I want to retain and the paradox that below I want to try to put to work.
The theme of archeology calls for the same perplexity in the face of an exercise that we can qualify as an intellectual asceticism. Under the sign of the two culminating ideas of the archive, as the register of discursive formations, and of archeology, as the description of interdiscursive transformations, Foucault has delimited a radically neutral terrain, or rather a costly neutralized one, that of statements without a speaker. Who can take up a position outside it? And how are we to continue to think about the formation and transformations not of discourses neutralized in this way, but of the relation between representations and practices? Moving from the archive to archeology, Foucault invited us to “reverse the procedure” and to “proceed to possible domains of application” (135). It is just this project that needs to be pursued following Foucault, in a field that cuts through the neutrality of the purged domain of statements. For a historiography that takes as the direct referent of its discourse the social bond, and as its rule of relevance the consideration of the relations between representations and social practices, the task is to leave behind the neutral zone of pure statements with an eye to reaching the relations between discursive formations, in the strict sense of the theory of statements, and the nondiscursive formations where language itself resists any reduction to a statement. Foucault, in truth, is not unaware of the problem posed by “institutions, political events, economic practices and processes” (162). Better, when he refers to these examples borrowed from the “non-discursive domain,” and does so within the framework of “comparative facts,” he takes the task of archeology to be “to define specific forms of articulation” (162). But can it do so without the kind of exit, of displacement I have spoken of?44 The notions of dependence and autonomy having ceased to function, the word “articulation” remains largely programmatic. But it needs to be made operational at the price of a displacement of Foucault’s displacement.
I do not want to leave Foucault without having again referred to the figure of Michel de Certeau, inasmuch as he offers a kind of counterpoint to the archeology of knowledge. There is also a “Certeau moment” on the plane of explanation/understanding. Essentially it corresponds to the second segment of the triad of “social place,” “scientific practices,” and “writing.”45 It is the high point designated by the term “practice” (69–86), to which must be joined the conclusion of L’Absent de l’histoire,46 without forgetting the pages of the same collection directly addressed to Michel Foucault: “The Dark Sun of Language” (115–32).
It is first of all as research that historiographical practice enters its critical phase, with the beginnings of the production of documents, which are set apart from actual human practice by a gesture of separation that recalls the collection of “rarities” in the form of archives according to Foucault (The Writing of History, 83). Certeau puts his own stamp on this inaugural operation by characterizing it as a redistribution of space that makes research a mode of “production of places.” But Foucault’s imprint is recognizable by the insistence on the notion of deviation that is expressly attached to this model. It is in relation to models that the differences taken as relevant deviate. Thus, in the history of representations characterizing the religious history practiced by Certeau, “sorcery, madness, festival, popular literature, the forgotten world of the peasant, Occitania, etc., all these zones of silence” (79) make for deviations. Each time, the relevant gesture is the means “of bringing forth differences relative to continuities or to elements from which analysis proceeds” (79). It is expressly to the totalizing claim of earlier history that this “research on the borderline” is opposed. But what models are in question? It is not a question of statements as in Foucault’s archeology, but of models drawn from other sciences: econometrics, urban studies, biology, as sciences of the homogeneous. Foucault would place these kinds of models among the “discursive formations” referred to right at the beginning of The Archeology of Knowledge. Nevertheless, this recourse to borrowed models suffices to justify the audacious extrapolation that makes Certeau say that the position of the particular in history is situated “at the limit of the thinkable,” a position that itself calls for a rhetoric of the exceptional whose outlines stem from the subsequent step of representation and literary writing, which we may take as Michel de Certeau’s major contribution to the problematic of the historiographical operation.
But before that we must say in what fashion L’Absent de l’histoire further expands the semantic space of the idea of a deviation by pairing it with the idea of the absent, which according to Certeau constitutes the distinctive mark of the past per se, as we shall see later in our section devoted to truth in history. History, in this sense, constitutes a vast “heterology” (173), a tracing of the “traces of the other.” But wasn’t it already the ambition of memory (which is named on the last page of this book) to produce the first discourse about the absent under the figure of the icon (180)? Whatever reservation we may have as regards the reduction of memory and history to one and the same celebration of absence, we can no longer oppose, in Foucault’s intransigent manner, the discontinuities linked to historical discourse and the presumed continuity of the discourse of memory. It may be here that Certeau begins to mark his own deviation in relation to Foucault. In the short, incisive essay titled “The Dark Sun of Language: Michel Foucault,” he sets out in search of his own difference. Bit by bit, he speaks haphazardly of his astonishment, his resistance, his second-degree assent, his ultimate reservations. It is true that he refers less to the archeology of knowledge than to the trilogy of works ending with The Order of Things. The alternating play of the order belonging to the “epistemological basis” of each episteme and the rupture that takes place between successive epistemes is welcomed, but leaves Certeau hungry for more. What “dark sun” is concealed behind this alternation? Is it not death, which is, however, named by Foucault himself? Yet he, Foucault, finally takes refuge in the “narrative” of this alternation between coherence and events. But it is beneath narrative that reason is truly “called into question by its history” (125). As a consequence, archeology does not escape the “equivocation” resulting from this unspoken something. It is in the wake of this suspicion that Certeau takes his distance from Foucault: “who is it that is to know what no one knows?” (161). In Foucault’s works, “who speaks and from where?” The question of May 1968 comes up. And a more cutting arrow is launched: “to speak of the death that founds all language is not yet to confront it, but may be to avoid the death that strikes this very discourse” (132). I fear that Certeau goes astray here, without being assured that he any better than Foucault escapes the question posed at the very heart of his work by the relation of historical discourse to death. For a reader who has both The Archeology of Knowledge and The Writing of History open before him must look on another side for the real gap between Foucault and Certeau, namely, on the side of the idea of production, and more explicitly the production of a place. The archeology of knowledge, we can say as does Certeau, does not speak of the place of its own production. Certeau therefore distances himself from Foucault by leaving behind the absolute neutrality of a discourse on discourse and by beginning to articulate this discourse in terms of other significant practices, the very task of a history of representations. In so doing, he puts off the difficulty posed by the question of the place of production until that inaugural moment when the gesture of doing history brings about a gap in relation to the practices through which human beings make history. This will be the moment of truth in history, where we shall encounter Certeau a final time. The actual reason for the deviation of Michel de Certeau in relation to Michel Foucault will have to be sought in how Certeau’s research is rooted in a philosophical anthropology in which the reference to psychology is fundamental and foundational. It was not an accident of compilation that brought together in The Writing of History the important article on “the historiographical operation,” which I have been distilling over the course of this work, and the two articles placed under the overarching title “Freudian Writings.” It is indeed a question of psychoanalysis and writing, more exactly, of the writing of psychoanalysis in its relation to the writing of history by historians. The first of these essays, “What Freud Makes of History,” was published in Annales in 1970. The question is what Freud as an analyst does with history. It is not when one undertakes to nibble at the obscure regions of history with supposed Freudian concepts, “such as the name of the father, the Oedipus complex, transference, in short when one makes use of psychoanalysis that one learns from it, but it is rather when one redoes the analyst’s work in the face of a case as singular as a pact concluded with the devil that the ‘legend’ (given to be read) becomes a ‘history.’”47 Here where the issue is Freud, the conclusion is that he is instructive, not because he makes something of the story told by others, and in the first place by historians, but when in his own way he does history. Beyond the fact that an important part of Certeau’s work results from this exchange among different ways of doing history, it is this very exchange that justifies the recourse to psychoanalysis in an epistemology of historical knowledge. The second essay is devoted to “The Writing of Moses and Monotheism,” the subtitle beneath the general heading “The Fiction of History.” What Freud makes understood in this controversial text is not some ethnographical truth, following the canons of that discipline, but the relationship of his construction, which he calls a “novel,” a “theoretical fiction,” with the fable, that is, the “legend” produced within a tradition. A writing, therefore, comparable under this heading to that of historians turns up in an incongruous manner within the territory of the historian. A historical novel takes its place alongside written histories. The lack of decisiveness of this literary genre between history and fiction, which we shall return to in the following chapter, adds to the difficulty and, in truth, constitutes it. What is important for the moment is the question of knowing in relation to what type of writing written history should situate itself. It is the search for this “place” of historical discourse among the ways of doing history that justifies taking psychoanalysis into account as an epistemology that, starting from within historical discourse, becomes external to it in regard to other ways of doing history. It is the very territory of the historian and his mode of explanation/understanding that finds itself expanded thereby. This carefully worked out opening is once again due to Certeau’s rigor.
The exemplary work of Norbert Elias proposes another kind of rigor for historians to think about than that practiced principally by Michel Foucault: not the rigor of a discourse on discursive operations apart from the field of practice, but the rigor of discourse on the conceptual apparatus put to work in an actual history bearing in a general way on the growth of political power from the end of the Middle Ages up to the eighteenth century. If this work can be criticized, it is not in terms of its conceptual coherence, but as regards its choice of the macrohistorical scale that remains unproblematic until it is confronted with a different choice, as we shall see in the following section. I must also add that Elias’s work is not defenseless in the confrontation with the reading I am going to undertake upon leaving behind the confused zone and semantic fuzziness that we have been considering.
I shall take as my guide part four of The Civilizing Process, titled “Towards a Theory of Civilizing Processes.”48 What Elias calls “civilizing processes” directly concerns my preoccupations relative to the establishing of a history of representations. It has to do with an ongoing process that, as Roger Chartier’s foreword to the French translation of The Court Society points out,49 is situated at the point of articulation between one noteworthy social formation, the central power of the state, apprehended in its monarchical phase during the Ancien Regime, and the modifications in sensibility and behavior we call civilization or, better, the civilizing process. In contrast to the future microhistory that installs itself straight away at the level of social agents, Elias’s sociology consists in a macrohistory comparable to that of the Annales school. And this is true in two ways: on the one hand, the civilizing process is correlated with large-scale phenomena at the level of the organization of society into the state, such as the monopolization of force and taxation and other such fees; on the other hand, this process is described as a series of progressively internalized constraints up to the point where they become a phenomenon of permanent self-constraint that Elias names habitus. The self is in fact what is at stake in civilization, what civilizes itself, under the institutional constraint. The descending course of an analysis from the top to the bottom of the social scale reveals itself to be particularly efficacious in the case of court society, where the social models unfold around a central core, the court, into coordinated and subordinated layers of society. One quite naturally thinks here of the relation between structure and conjuncture in Labrousse or the hierarchy of scales of different time spans in Braudel. In fact, things are more complicated than that, and the category of habitus will come to include all the features that distinguish a dynamic phenomenon of a historical order from a mechanical one in the physical order. It is worth noting that Elias does not speak of determinism—even if he does speak of constraint—but of the interdependence between the modifications affecting the political organization and those affecting human sensibility and behavior.
In this regard, Elias’s key concepts have to be carefully respected in their rigorously spelled out specificity: “formation” or “configuration,” designating the contours of the organizing phenomena, for example, the court society; “equilibrium of tensions,” designating the hidden springs of social dynamics, for example, in the gathering together of the warriors who preside over the court society and in the competition between aristocracy and bourgeois office holders that will contribute to the breaking up of that society;50 “evolution of formations,” designating the rule-governed transformations that simultaneously affect the distribution and the displacements of political power and the psychic economy that governs the distribution of motives, feelings, and representations. If Elias’s apparatus for describing and analyzing had to be designated by a single term, it would be interdependence, which leaves an opening to what in an approach more sensitive to the response of social agents would be called appropriation. Elias, to be sure, did not cross this threshold—and its important corollary, uncertainty—but the place for it is clearly designated. It is located on the trajectory from social to self-constraint that the “Outline of a Theory of Civilization” brings to the fore. The category of habitus, as a result, will become a problem. Elias only covers this trajectory in one direction, the return voyage remains to be done. But what is important in Elias’s eyes is first of all that the process is not rational in the sense of being willed and directed by individuals. Its rationalization is itself the effect of self-constraint. Next, the social differentiation, resulting from the increased pressure of competition, gives rise to an increasing differentiation, and hence to a more complete, more regular, more controlled articulation of behaviors and representations, something that is well summed up in the expression “psychic economy,” for which the term “habitus” constitutes an exact synonym. Certainly, it is a matter of a constraint, but of a self-constraint that includes reserves of reaction that express themselves on the level of the equilibrium of tensions. All the leading terms of Elias’s text—differentiation, stability, permanence, control, predictability—are capable of dialecticalization. All the described phenomena of self-constraint constitute formulas of dispersion for the drifting toward the extremes that the process of civilization undertakes to resolve. Thus habitus consists in a regulation sanctioned by the equilibrium between these extremes.51 The phenomenon of the diffusion of self-constraint is interesting in this regard. It provides the occasion for introducing, along with the concept of social layer (first with the pair warrior/courtier, then with aristocrat/bourgeois) that of a psychic layer, close to some instances from psychoanalytic theory (superego, ego, id), despite Elias’s mistrust regarding what he takes to be the antihistoricism of Freudian theory. This phenomenon of diffusion from (social and psychic) layer to layer also brings to light the phenomena of dispersion and recentering, thanks to the phenomenon of diminishing contrasts that makes us “civilized.”
The most noteworthy contribution of The Civilizing Process for a history of mentalities and representations has to be sought in Elias’s examination of two major modes of self-constraint, that of rationalization and that of shame. It is within the framework of the court, with its quarrels and plots, that Elias, encouraged by La Bruyère and Saint-Simon, situates one of the key moments of the conquest of reflection upon and the regularization of our emotions, of that knowledge of the human heart and the social setting that can be summed up by the term “rationalization.” In this regard, the trace of the heritage of the court can be followed up to Maupassant and Proust. Something is at issue here that is more than what the history of ideas calls reason. There is a close correlation between the social cohabitation of human beings and what a “historical psychology” (406) will take as a habitus of the psychic economy considered as an integrated whole. The history of ideas wants only to consider contents, “ideas,” “thoughts,” the sociology of knowledge focuses on ideologies, or even a superstructure, psychoanalysis on a conflict between competing drives detached from their social history. But rationalization consists in an internal relation within each human being that evolves in correlation with human interrelationships. The civilizing process is nothing other than this correlation among the changes affecting the psychic structures and those affecting social structures. And habitus lies at the crossroads of these two processes.52 A sense of shame is the second figure that the “habitus of the West” brought about. It consists in a regulation of fear in the face of the inner perils that, in a regime of civility, took the place of the external threat of violence. The fear of revealing one’s inferiority, which is at the heart of weakness before another’s superiority,53 constitutes a central theme in the conflict that constructs our psychic economy. Here once again, “we can only speak of shame in conjunction with its socio-genesis” (416). Much more could be said concerning the characterization of this sense of shame (Elias associates it with “embarrassment”). Essentially it has do with the process of internalization of fears that within the emotional order corresponds to rationalization within the intellectual one.
I have said enough to indicate the points where Elias’s analyses lend themselves to a dialecticalization of the described processes that he describes in a unilinear fashion from the top to the bottom of the social scale.54 Below, we shall examine in what way the theme of appropriation may balance that of constraint. Elias himself opens the way to a parallel dialecticalization in one passage where, after having emphasized the nonrational character (in the sense indicated earlier) of the formation of habits, he comments: “But it is by no means impossible that we can make out of it [civilization] something more ‘reasonable,’ something that functions in terms of our needs and purposes. For it is precisely in conjunction with the civilizing process that the blind dynamics of people intertwining in their deeds and aims gradually leads towards greater scope for planned intervention into both the social and individual structures—interventions based on a growing knowledge of the unplanned dynamics of these structures” (367).55
Diversity. . . . A town or a landscape from afar off is a town and a landscape, but as one approaches it becomes houses, trees, tiles, leaves, grass, ants, ants’ legs, and so on ad infinitum. All that is comprehended in the word “landscape.”
§ Pascal, Pensées56
In the preceding analyses the question of scale, and more precisely of the chosen scale adopted by the historian, was not posed. To be sure, the heuristic models proposed and used by Labrousse and Braudel and by a great part of the Annales school clearly stem from a macrohistorical approach,extended from the economic and geographical basis of history to the social and institutional level and to phenomena said to be of the “third type,” from which stem the forms of the most stable predominant mentalities. But this macrohistorical perspective was not deliberately chosen, hence not preferred to something that could be taken as an alternative. The sequence “structure, conjuncture, event” in Labrousse and the hierarchy of time spans in Braudel implicitly rest on an interplay of scales, but as the tripartite composition of Braudel’s The Mediterranean testifies (which remains the model for this genre), the preference accorded to a reading from the top to the bottom of the hierarchies of time spans was not thematized for itself, so that we can envisage changing scale and the very choice of a scale as a power open to the historian’s discretion, with all the liberties and constraints that result from such a choice. Access to this mobility in the historian’s gaze constitutes an important conquest of history during the last third of the twentieth century. Jacques Revel even adopts the phase “interplay of scales” to greet the exercise of this methodological freedom, which we shall assign to the part of interpretation implied in the search for truth in history when the time comes.57
The approach to microhistory adopted by some Italian historians stems from this interplay of scales.58 By taking a village, a group of families, an individual caught up in the social fabric for their scales of observation, the practitioners of microhistoria not only have made clear the relevance of the microhistorical level with which they work, they have also brought up for discussion the very principle of a variation in scale.59 We shall not focus on the defense and illustration of microhistoria as such, but rather on an examination of the very notion of a variation in scale, in order to evaluate the contribution of this original problem to the history of mentalities or representations, which we have seen threatened in turn from within by collapse and intimidated from without by demands for rigor that its use of fuzzy concepts make it incapable of satisfying.
The key idea attached to the idea of a variation in scale is that, when we change scale, what becomes visible are not the same interconnections but rather connections that remained unperceived at the macrohistorical scale. This is the sense of the magnificent aphorism from Pascal’s Pensées that Louis Marin, whose name will return below, liked to cite.60
The notion of scale is borrowed from cartography, architecture, and optics.61 In cartography, there is an external referent, the territory that the map represents. What is more, the distances measured by maps of different scales are commensurable according to homothetic relations, which authorizes us to speak of the reduction of a terrain to a given scale. However, from one scale to another we observe a change in the level of information as a function of the level of organization. Think of a roadmap. We see the primary axes of circulation in large scale, the distribution of dwellings on a small one. From one map to another, space is continuous, the territory is the same, hence a small change in scale shows the same terrain. This is the positive aspect of a simple change in proportion. There is no room for an opposition between scales. The counterpart is a certain loss of details, of complexity, hence of information in the passage to a larger scale. This double feature—proportionality of dimensions and heterogeneity of information—has to affect geography, which is so dependent on cartography.62 A discordant geomorphology appears with a change of scale within the same geopolitical setting, as can be verified in detail by rereading the first part of Braudel’s The Mediterranean. The term “Mediterranean” situates the object of inquiry at the level of what Pascal calls landscape—all that enveloped by the name Mediterranean, we might say at the end our reading!
The role of the idea of scale in architecture and in urban planning is also relevant to our discussion. Proportional relations comparable to those in cartography are posited along with the balance between gain and loss of information depending on the scale chosen. But unlike the relationship between map and territory, the architect’s or urban planner’s plan has as its referent a building, a town, yet to be constructed. What is more, the building or the town have varying relations with their contexts scaled in terms of nature, the landscape, communication networks, the already constructed parts of the town, and so on. These characters belonging to the notion of scale in architecture and in urban planning concern the historian inasmuch as the historiographical operation is in one sense an architectural one.63 Historical discourse has to be built up in the form of a set of works. Each work gets inserted into an already existing environment. Rereadings of the past are in this way reconstructions, at the price sometimes of costly demolitions: construct, deconstruct, reconstruct are familiar gestures to the historian.
It is through these two borrowings that reference to the optical metaphor becomes operative in history. Behaviors linked to the accommodating of this gaze are not noted inasmuch as the nature—or even the beauty—of the uncovered spectacle leads to forgetting the focusing procedures of the optical apparatus used at the price of the learned manipulations. History, too, functions in turn as an eyepiece, a microscope, or a telescope.
What the notion of scale includes within itself in the use historians make of it is the absence of commensurability of the dimensions. In changing scale, one does not see the same things as larger or smaller, in capital or lower case letters, as Plato puts it in the Republic about the relationship between the soul and the city. One sees different things. One can no longer speak of a reduction of scale. There are different concatenations of configuration and causality. The balance between gains and losses of information applies to the modeling operations that bring into play different heuristic imaginary forms. In this regard, what we can reproach in macrohistory is its failure to notice its dependence on a choice of scale with its macroscopic optical point of view that it borrowed from a more cartographical than historical model. For example, we can observe in Braudel some hesitation in the handling of the hierarchy of time spans. On the one hand, an interlocking relationship is presumed between homogeneous linear time spans thanks to the inclusion of all of them in one unique calendar time, itself indexed in terms of the stellar order, despite a certain mistrust regarding the abuse of chronology committed by the history focused on short-term events. On the other hand, we can also observe a piling up of superimposed time spans with no dialectical relation between them. The history of mentalities incontestably suffered from this methodological deficiency relative to the changes in scale insofar as the mentalities of the masses were presumed to stem from the long time span, without the conditions of their diffusion on smaller scales being taken into account. Even in Norbert Elias, himself a master in his use of the concept, the phenomena of self-constraint were said to hold across clearly identified social layers—the court, the nobles de robe, the city, and so on. But the changes in scale implied in the examination of the diffusion of models of behavior and of feeling from one social layer to another were not acknowledged. In a general way, the history of mentalities, insofar as it had simply extended the macrohistorical models of economic history to social history and to phenomena of the “third type,” tended to deal with the concept of social pressure as an irresistible force operating in an unperceived fashion in relation to the reception of messages by social agents. The treatment of the relations between high and popular culture was particularly affected by this presupposition that goes with a reading that runs from the top to the bottom of the social scale. Other pairs stemming from similar binary systems were equally reinforced by this same prejudice: strength vs. weakness, authority vs. resistance, and, in general, domination vs. obedience, following the Weberian schema of domination (Herrschaft).64
Two leading works from Italian microhistoria, accessible in French, have held my attention. Carlo Ginzburg in a short and incisive preface comments that it is thanks to an exception, given “the scarcity of evidence about the behavior and attitudes of the subordinate classes of the past,” that it was possible to recount “the story of a miller of the Friuli, Domenico Scandella, called Menocchio, who was burned at the stake by order of the Holy Office after a life passed in almost complete obscurity.”65 On the basis of the dossiers of his two trials, along with other documents relative to his working life, his family, and also his readings, Ginzburg was able to lay out the “rich picture of his thoughts and feelings, of his imaginations and aspirations” (xiii). Therefore this documentation has to do with what is called “‘the culture of lower classes’ or even ‘popular culture’” (xiv). Ginzburg does not talk about scale but about cultural levels, whose existence is taken as a precondition for what become self-defined disciplines. This argument about the self-definition, almost the tautology, of social groups and professionals—such as the bourgeoisie—practiced in social history can be found in other historians unmarked by Italian microhistory, whom we shall return to below. Terms about culture—popular culture, high culture—and by implication those about dominant and subordinate classes implied in ideological quarrels linked to vulgar Marxism and anticolonial protests are reworked. The scarcity of written documents from a largely oral culture serves as the reason why. Even Robert Mandrou, whose place in the history of mentalities we noted above, is not exempt from reproach for having preferred to deal with the culture imposed upon the popular classes—we shall return to this below with regard to Certeau’s Possession at Loudon—making this an effect of successful acculturation.66 If literature meant for the people is not to cover over that produced by the people, this latter has still to exist and to be accessible. This was the case with Menocchio’s confessions, which, owing to their uniqueness, do not meet the requirements for serial, quantitative history for which number and anonymity are important.
But how are we to avoid falling back upon anecdotes and the history of events? A first answer is that this objection is directed principally against political history. Another, more convincing one is that it is the latent and dispersed properties of an available historical language—which the computer misses—that the historian brings to light and organizes into discourse. What this historian articulates are the readings of a man of the people, meaning almanacs, songs, pious books, lives of saints, and brochures of all kinds that this miller put together in his own unique way. By leaving behind quantitative history, one does not fall into noncommunication. What is more, these reformulations express not only the reorganizing power of a man of the people’s reading, they bring to the surface traditions and sleeping heresies that a surviving situation in a way brings back to life. The consequence for our problem of the history of mentalities is that the very concept of mentality has to be set aside inasmuch as this history, on the one hand, emphasizes only “the inert, obscure, unconscious elements in a given world view” (xxiii) and, on the other, retains the “interclass” connotation of a common culture—a presupposition that even Lucien Febvre did not escape in speaking of “men of the sixteenth century.” The great French historian nevertheless did resist the presuppositions so strongly attached to the unhappy heritage of the sociological concept of a “collective mentality.” Menocchio, for his part, cannot be situated in terms of that illiterate line, coming as he did after the invention of printing and the Reformation, which he must have read about and discussed.67
The other book that caught my attention is by Giovanni Levi, Le Pouvoir au village: Histoire d’un exorciste dans le Piémont du XVIe siècle,68 with a preface by Jacques Revel: “L’Histoire au ras du sol.” Here we are on the terrain worked on by Norbert Elias. But at the bottom of the scale—in the village. It is not about large numbers or about an individual. Nor does it deal with quantified indicators—prices or earnings, levels of wealth, distributions of professions—first named, then counted; nor with the regularities of a slow moving, almost immobile history, norms or common customs. The appearance and articulation of the phenomena considered are the fruit of a change of scale. Instead of aggregates followed over the long time span, we have a tangle of interrelations that need to be deciphered. But we ought not to expect from this a resurrection of the lived experience of social agents, as if history were to stop being history and link up again with the phenomenology of collective memory. Respect for that subtle boundary is important for my thesis, which never denies the implicit profession of the epistemological break separating history from memory, even collective memory. It is always interactions that are gathered and reconstituted.69 The important word “reconstruction” is pronounced, which, later, will relaunch the history of mentalities, now better called the history of representations, beyond the limited example of such microhistoria. But, before moving on to this more or less well controlled extrapolation, we need to have brought to its critical point history linked to the choice of a microhistorical scale. We have said, at a smaller, even minute scale, we see things we do not see at a higher one. But we need to add that what we do not see and must not expect to see is the lived experience of the protagonists. What we see remains social interactions, at a fine scale, but one already microstructured. I will add, with a slight hesitation, that Levi’s attempted reconstruction only partially satisfies Ginzburg’s well known “evidential paradigm” from his essay “Roots of an Evidential Paradigm.” The microanalysis practiced by Levi lacks the flair of a detective or of an expert in detecting counterfeit paintings or of any sort of psychomedical semiotics. The same operation of reconstructing the real that distances it from actual lived experience also distances the evidential, bringing it closer to the more classic operations of dividing things up, of articulation, and of confronting testimonies, all of which allows Levi to speak of an “experimental history.” But what does one experiment on? On the exercise of power in the village on a microhistorical scale. What we see at this scale are family and individual strategies, faced with economic realities and hierarchical relationships, in a play of exchanges between center and periphery, in short, the interactions that find their place in a village. With the concept of strategy, a noteworthy figure of rationality is brought to light, whose fruitfulness we shall evaluate below in terms of the uncertainty—opposed in turn to the fixity, permanence, and security—in short, the certainty—attached to the functioning of social norms on the larger scale, those quasi invariants of the history of mentalities over the long time span. It will be a legitimate question whether the forms of behavior placed under this term “strategy” have as their secret or admitted end reducing uncertainty or merely coming to terms with it.70 “The great social and political game that is the real subject of this book,” says Revel in his preface to Le Pouvoir au village, is, if you will, the same one that Norbert Elias reconstructs in his The Civilizing Process, but in the sense of, in Pascal’s words, “all that is comprehended in the word ‘landscape.’” But can we say that details that, in a way, lay out the landscape lead to recomposing it following specificable rules?
This is the whole question in the passage from microhistory to macrohistory.71 If one can reproach macrohistory for proceeding from the long time span to subordinate ones in terms of no stated rule, does microhistory have arguments that allow it to say that it can take up again the project of total history, but beginning from the bottom? Concretely, the question comes down to asking whether the village is a favorable place for identifying the intermediary forms of power, through which power in the village articulates the power of the state as it is exercised in that time and that region? Uncertainty is precisely what affects the evaluation of the forces at work. And it is the task of Levi’s book to explore these relationships when the hierarchy is viewed from below. Stated in terms of the epistemology of historical knowledge, the question becomes that of how representative is this history of a village and the interactions that take place there? Is the uncertainty of the protagonists also that of the analyst? Does it also weigh on the capacity held in reserve for generalization in what otherwise only would constitute a case history? And is this lesson generalizable to the extent that it can be opposed point by point to what Norbert Elias draws from his study of court society and other comparable societies?72 In sum, “how representative is a concatenation circumscribed in this way? What can it teach us that should be generalizable?” (Revel, Jeux d’échelles, xxx). Edoardo Grendi has proposed one formula that Revel treats as an elegant oxymoron, namely, the idea of the “normal exception.” Yet this formula is valuable because of what it sets aside: an interpretation of the concept of exemplarity in statistical terms, following the model of quantitative and serial history. Perhaps it invites us simply to compare worldviews arising from different levels of scale, without these worldviews being totalized. But from what higher mastery would such an overview of different scales stem? It is doubtful that somewhere there is a place from which to take such an overview. Are not Pascal’s two fragments first titled “diversity,” then “infinity”?
FROM THE IDEA OF MENTALITY TO THAT OF REPRESENTATION
I must now present the conceptual leap that constitutes access to the following section.
At the end of the first section we left the concept of mentalities in a state of great confusion when set against the background of the notion of total history into which that of mentalities is supposed to be integrated. There we were subject to two kinds of appeals: On the one hand, the one emanating from three kinds of discourse, highly divergent among themselves, but each in its way requiring a conceptual rigor held to be the only one possible for presiding over the reassembling of a burst-apart history; on the other hand, that of an original historiography linked to a choice apparently the opposite of the one implicit in the dominant historiography of the golden age of the Annales school, the choice of the microhistorical scale. The time has come to prudentially and modestly set out along the way to a reordering of the historical field, one where the history of mentalities will play a federating role on the condition of assuming the title and function of a history of representations and practices.
In order to get beyond the dispersed situation of history during the last third of the twentieth century, I propose to take as my guide a global approach that seems to me to satisfy in large measure the thrice called-for conceptual rigor inasmuch as it carries the notion of a variation in scale to its farthest limits. I shall attempt to show that the often unexplained replacement of the fuzzy concept of mentality by that of representation, when better articulated and more dialectical, is perfectly coherent with the uses I am about to propose concerning the generalized concept of a variation in scale.
The global approach I am referring to finds its most explicit formulation in the collection of essays titled Les Formes de l’expérience: Une autre histoire sociale, edited by Bernard Lepetit.73 The historians in this volume take as their focal reference term in the societies considered—what I would call the pertinent object of historical discourse—the instituting of the social bond and the modes of identity attached to it. The dominant tone is that of a pragmatic approach where the principal accent is on social practices and the representations integrated into these practices.74 This approach can legitimately draw on a critique of pragmatic reason where it intersects, without becoming confused with, a hermeneutic of action that itself proceeds from the enriching of the phenomenology of Husserl and Merleau-Ponty by semiotics and by the whole blossoming of works devoted to language games (or to discourse). The resolutely historical branch of this critique of practical reason can be recognized from the fact that the social bond and the changes affecting it are taken to be the relevant object of historical language. In this way, the epistemological break brought about by the Labroussian and Braudelian models is not denied. It is deliberately assumed by the new research program that posits “as its first priority the question of identities and social bonds” (13).
The continuity with preceding programs of the Annales school can be seen in that the three problems identified in the introduction to this chapter—that of the kind of change taken as most relevant (economic, social, political, cultural, and so on); that of the scale of description; and that of temporal spans—are dealt with as one interdependent block.75
Their commitment to a critique of pragmatic reason first made these historians more attentive to the increasingly problematic character of the instituting of the social bond. This is why they speak more readily of structuration than of structure, and regard norms, customs, or legal rules as institutions capable of holding societies together. Next, this spontaneous commitment to a critique of practical reason made them more attentive to the articulation of practices properly speaking and of representations that we can legitimately take as themselves being theoretical or, better, symbolic practices.76 Finally, their recourse to a critique of practical reason allows justifying the often unreflected-upon shift from the vocabulary of mentality to that of representation. I will now turn to a motivated reason for this substitution.
The semantic fuzziness for which one can legitimately reproach the idea of a mentality is inseparable from the massive, indiscriminate character of the phenomenon, which was readily assimilated to the expression of some time period, even, thinking of Hegel, to the spirit of a people. This happened because the mere juxtaposing of the mental to other components of society as a whole did not allow for the appearance of its innermost dialectic. Better articulated in terms of social practice or practices, the idea of representation will reveal dialectical resources that the idea of mentality does not allow to appear. I am going to show that the generalization of the idea of a play of scales can constitute a privileged way for bringing to light this dialectic hidden in the idea of representation when paired with that of a social practice.77
What is important in the play of scales, in effect, is not the privilege granted to the choice of some scale so much as the very principle of a variation in scale, something like what is conveyed by the aphorism from Pascal used as the epigraph to the preceding section. A variety of effects can then be attributed to the exercising of these variations. I have brought together three of them in terms of the theme of identities and the social bond. They each contribute in different ways to the recentering of historiography at the end of the twentieth century. Exercising this variation in scale can draw upon three converging lines. To the first of these, I would attribute the variations affecting the degrees of efficacy and coerciveness of social norms; to the second, those variations modulating the degrees of legitimation at work in the different spheres to which one can belong among which the social bond is distributed; and to the third, the nonquantitative aspects of the scale of social times, something that will lead us to rework the very idea of social change that presided over our whole inquiry concerning the explanation/understanding practiced in history. Following these three routes, we shall keep in mind Pascal when he says that at each scale one sees things one does not see at another scale and that each vision has its own legitimate end. At the end of this threefold consideration we will be able to confront directly the dialectical structure that calls for preferring the idea of representation to that of mentality.
The Scale of Efficacy or of Coerciveness
As microhistory has already verified, the initial benefit of a variation in scale is that it shifts the accent to individual, familial, or group strategies that call into question the presupposition of submission by social actors on the bottom rank to social pressures of all kinds, and principally those exercised on the symbolic level. This presupposition is not unconnected to the choice of the macrohistorical scale. It is not only time spans in the models stemming from this choice that appear to be hierarchical and interconnected, but also the representations governing behavior and practices. To the degree that a presupposition of submission by social agents goes with a macrohistorical choice of scale, the microhistorical choice leads to the opposite expectation, that of random strategies in which conflicts and negotiations take precedence under the sign of uncertainty.
If we broaden our gaze beyond macrohistory, we see outlined in other societies than those studied by microhistoria entanglements of great complexity between the pressure exercised by models of the behavior seen as dominant and the reception, or better the appropriation, of received messages. At the same time, every binary system opposing high to popular culture, along with their associated pairings (force/weakness, authority/resistance), totters. Opposed to them are: circulation, negotiation, appropriation. The whole complexity of social interaction has to be taken into account. Yet for all that the macrohistorical view is not refuted. We can continue to read Norbert Elias as we trace out symbolic orders and their power of coercion from the top to the bottom of societies. It is precisely because the macrohistorical vision is not abolished that we can legitimately pose the question of how representative microhistorical organizations are when considered in regard to the phenomena of power readable on the broad scale. In any case, the notion of deviation we often find in comparable contexts cannot exhaust the combinatory resources of pictures drawn at different scales. It is still higher-order systems that are considered from below.78 In this regard, the extension of the domain of representations of the models of long-time-span history remains legitimate within the limits of the macrohistorical point of view. There is a long time scale for the features of mentalities. Nothing is lost from the problem Durkheim posed at the beginning of the twentieth century precisely under the title of “collective representations,” a term significantly that has reappeared following the long use of “mentality” by those associated with Annales. The Durkheimian idea of “basic norms,” which goes with those of unperceived agreements and agreement concerning the modes of agreement, retains its problematic and pragmatic force.79 The task is rather to place these guiding concepts in a dialectical relation to those governing the appropriation of these rules of agreement about agreement. Furthermore, mere consideration of the necessary economy of the creative forces resisting forces tending toward rupture leads to giving some credit to the idea of a customary habitus that can be assimilated to a principle of inertia, even of forgetfulness.
In this spirit, and under the heading of the scale of efficacy or of coerciveness, the problems of institutions and of norms, which each obey different contextual rules, can be considered jointly.80
The major uses of the idea of an institution—as juridical-political; as an organization functioning in a regular manner; as an organization in the broad sense tying together values, norms, models of relations and behaviors, roles—lead to the idea of regularity. A dynamic approach to the constituting of the social bond will then surmount the contingent opposition between institutional regularity and social inventiveness, if we speak of institutionalization rather than of institutions.81 In this regard, the work of institutional sedimentation gains, it seems to me, in being compared to the work of archiving things we saw at work at the documentary level of the historiographical operation. Might we not speak, in an analogical sense, of an archiving of a social practice? Considered in this way, the process of institutionalization brings to light two faces of the efficacy of representations: on the one hand, in terms of identification—the logical, classificatory function of representations; on the other hand, in terms of coercion, of constraint—the practical function of establishing conformity in behavior. On the path to representation the institution creates identities and constraints. Having said this, we ought to stop opposing the coercive aspect, by preference assigned to the institution, to the presumably subversive side seen in social experience. Considered from a dynamic point of view, the process of institutionalization oscillates between the production of nascent meaning and the production of established constraints. Thus we can formulate the idea of a scale of efficacy of representations. Norbert Elias’s analyses of the relations between physical forces camouflaged as symbolic power, or those of Michel Foucault in Discipline and Punish, would need to be placed on a scale of efficacy considered as a scale of coercion. What is important is that “human beings need institutions, which is another way of saying that they make use of them as much as they serve them” (Revel, “L’Institution et le social,” 81).
In other contexts, one will prefer adopting as a conceptual device the idea of a norm, where the accent is turn by turn on the process of evaluation that marks out the permitted and the forbidden or on the modes of feelings of obligation sanctioned by punishment. The idea of a norm, too, deployed from the moral to the juridical plane, lends itself to a variation in the scale of efficacy, in the orders of identification and the qualifying of behavior as much as in that of degrees of coercion. It is along such a scale that we may place the opposed manners of approval and disapproval in procedures of legitimatizing and condemning. I shall have more to say about this when we consider the diversity of applications of the idea of a norm in a plurality of realms of interactions among forms of behavior. Here we can already observe the general dialectical structure. The figures of the just and unjust can be taken to be the basis for opposed evaluations. Those of the just mark out the modes of claimed or assumed legitimacy, those of the unjust the modes of condemned illegitimacy. From the point of view of the process as dynamic, the fundamental competence of social agents to negotiate conflicts is added to this basic polarity. This capacity is exercised as much on the plane of the qualifying of contested or assumed behaviors as on that of the levels of refused or accepted coercion.82 One interesting concept halfway between justification and condemnation would be that of an “adjustment” of action, of an action that “fits.”83
The Scale of Degrees of Legitimation
The second line along which the theme of a variation in scale lends itself to an instructive extension is that of the degree of social status that social agents may claim in the order of public esteem. But one is not great or small at no price. One becomes great when, in a context of discord, one feels justified in acting in the way that one does. High status and justification thus go hand in hand. The notion of justification adds a new dimension of intelligibility to those of the institution and the norm. Discord, conflict, disputes, differences of opinion constitute the relevant context. We laid the way for establishing this pair of high status and justification at the moment when we adopted as a general principle for reorganizing the historical field the establishment of the social bond and the search for identity that is attached to it. It is in situations of discord that social agents raise their demands for justification. The same feeling of injustice we saw at work in the stratagems of condemnation is at work in strategies of legitimation. The question is the following: How to justify agreement and manage disagreement, principally by means of compromise, without giving into violence? Here is where the question of status presents itself, which brings into play something other than a need for taxonomic classification, namely, a need for recognition that takes as its basis the scale of evaluations at work in a series of qualifying tests—a notion we also encounter in other contexts, such as heroic folktales. Luc Boltanksi and Laurent Thévenot have added a complementary component of intelligibility to that of status by taking into account the plurality of regimes of justification resulting from the plurality of types of conflict. Someone who has high status in the commercial order may not be so in the political order or in the order of public reputation or in that of aesthetic creation. The principal concept thus becomes that of “economies of standing.”84 What is important for the present investigation is to join to the hierarchical idea of status or standing, which is a variant of the idea of scale, the horizontal line of the pluralization of the social bond. Interweaving these two problematics contributes to breaking with the idea of a common mentality, too readily confused with that of a undifferentiated common good. Certainly the idea of a “common humanity of the members of the city” (De la justification, 96) is not to be rejected. It equalizes human beings as human, excluding in particular slavery or treating some as subhuman. But in the absence of differentiation this bond remains nonpolitical. To the axiom of common humanity must be added that of dissimilarity. It is what sets in movement the qualifying tests and sparks off procedures of justification. In turn, these latter are oriented toward setting in place compromises satisfying the model of an “orderly humanity” (99). The enterprise remains risky and in this sense uncertain inasmuch as “there exists no neutral measuring point, external to and superior to each world, from which the plurality of justices could be considered from on high, like a range of equally possible choices” (285).85 As a result, attempts at justification make sense in distinct “cities,” in multiple “worlds.” The difficult question raised by this work is that of the criteria of workable justifications in each city. Each criterion is linked to the question of identifying distinct spheres of action.
Two discussions directly applicable to my thesis—which is that of the fruitfulness of the theme of the play of scales for a history of representations—are opened in this way. The first one concerns the finite character of the regressive process that, from elementary justifications to secondary justifications, leads to an ultimate justification in a given sphere. The division among cities or worlds is strictly correlative with the coherence of regimes of action thereby justified. The problem, once again, is not of a taxonomic order, but one of hierarchy in estimation. As in Aristotle’s system, we must admit the necessity to stop somewhere. The enumeration of cities—religious, domestic, retail, cultural, civic, industrial—rests on such a postulate of a finite terminal justification. This difficulty calls forth a second one. What discourse authenticates the final justification appropriate to some city? How do we recognize the ultimate argument proper to some city or world? Here our authors adopt an original but costly strategy. To identify the forms of argument at work in ordinary discussions, they place them under the aegis of more articulated, stronger arguments, where the process of justification is brought to its reflective peak. Thus the works of philosophers, theologians, politicians, and writers, backed up by training manuals meant for corporate managers and union representatives, are called upon. And in this way Adam Smith, Augustine, Rousseau, Hobbes, Saint-Simon, and Bossuet provide the founding discourses for actual discourse in everyday disputes. But the question is then of the fit between these founding discourses and the ones they justify. One may be pleased to see philosophy reintroduced within the social sciences as one argumentative tradition, something that provides an indirect justification of philosophy, and for our two authors, who are an economist and a sociologist, acknowledgment that they belong to a history of meaning. But one may also ask in return about the true nature of the tie existing between the texts read by our sociologists and the discourses practiced by social agents inasmuch as the great foundational texts were not meant for this use and, moreover, they are in general unknown to social agents or their representatives on the plane of public discussion. This objection, which one might direct against our authors’ whole enterprise, is not unanswerable in that the social space itself makes a place for another type of scale, that of levels of reading between archetypical texts and the least organized kinds of discourse. Both kinds, as written, were given to be read to a multiplicity of readers forming a chain. After all the sixteenth century Italian miller from the Friuli provided himself with arguments for his clever negotiations on the basis of his contingent readings. Yes, reading also has its scales, which get interwoven with those of writing. In this sense, the great texts that serve to explain and decipher the lesser texts of a lesser caliber of ordinary negotiators stand halfway between those written by historians when they join archetypal texts to the implicit discourses in the cities in question and those sometimes written by social agents themselves about themselves. This chain of writing and reading assures the continuity between the idea of representation as an object of history and that of representation as a tool of history.86 In its first sense, the idea of representation continues to be part of the problem of explanation/understanding; in the second, it falls under that of the writing of history.
The Scale of Nonquantitative Aspects of Social Times
I would like to end this brief examination of applications of the notion of a variation in scale with an extension to nonquantitative aspects of the temporal component of social change. The nesting of long, middle, and short time spans, well known to readers of Braudel, rests in the first place on quantitative relations between measurable intervals of centuries for the long time span, decades for conjunctures, down to days and hours for dated events. A common chronology lays out dates and indexed intervals in terms of calendar time. Owing to this, measurable spans of time are correlated with repetitive, quantifiable aspects submitted to a statistical treatment of noted facts. However, within this same well-delimited framework of what is measurable, the time spans considered present intensive aspects oftentimes disguised in extensive measures such as the speed of or acceleration in the changes considered. To these two notions, which are measurable only in appearance, are added values of intensity such as rhythm, cumulativeness, recurrence, persistence, and even forgetting, inasmuch as the reservoir of real capacities of social agents adds a dimension of latency to that of temporal actuality. We can speak in this regard of a scale of available competencies of social agents.87
Having said this, we can apply the notions of scale and variation in scale to these intensive modes of historical time. There is no reason to abandon the field of scales of time spans opened by Annales. There is also a long time span of the features of mentalities. This holds for global society, but also for those cities and worlds whose plurality structures social space. In this regard, it is necessary to learn to cross the plurality of worlds of action not just with the scales of efficacy, as said above, but with the scale of temporal regimes, as we are now attempting to do. Here too the accent has to be placed on the variation in scale and not on the presumed privilege of one or another scale.
Approached in terms of intensive, not extensive size, the time span Durkheim attached to the notion of successful agreement is worth reexamining. “A successful agreement,” observes Bernard Lepetit, “precisely because it is successful makes itself a norm through the regularity of its imitative repetitions” (Les Formes de l’expérience, 19). The very notion of regularity is what stops being taken for granted. Paired with that of reiteration, it calls for the counterpart of behaviors that are forms of appropriation, stemming from the actors’ competence. A scale of temporalities is thereby opened to intersecting traversals. To the linearity of a slow descent from top to bottom corresponds the reordering continually at work in different uses over the course of the time span. This revision of the temporal concepts used in historiography has to be pushed quite far. It must not leave out, in the opposite direction, certain concepts that were highly privileged in the encounter with an emphasis on supposedly quasi-immobile structures under the influence of structuralism, and also Marxism. The categories of a leap, of deviation, of fracture, crisis, revolution, typically found in historical work during the last third of the twentieth century, all need to be reconsidered. The plea in their favor certainly does not lack relevance. By privileging deviation rather than structure, does not the historian reinforce his discipline over against sociology, which focuses on stable features whereas history concentrates on the instable ones? To be sure. But the categories of stability and instability, of continuity and discontinuity, along with other akin pairs of oppositions, which add a note of radicalness to the categories just enumerated, must, in my opinion, be dealt with in terms of a framework of polarities themselves relative to the idea of social change.88 This hypercategory is not at the same conceptual level as are the opposed pairs just named. It coheres with the relevant features of the basic referent of historical knowledge, namely, the past as a societal phenomenon. And it is from this same referential level that arise the dynamic aspects of the constituting of the social bond, with its play of identities, of readability, of intelligibility. In relation to the metacategory of social change, the categories of continuity and discontinuity, of stability and instability, have to be treated as opposite poles of a single spectrum. In this regard, there is no reason to leave to the sociologist the question of stability, which seems to me as worthy of reexamination as do those of continuity and discontinuity, which under the helpful influence of Michel Foucault’s archeology of knowledge have occupied the foreground of recent discussion. The category of stability is one of the more interesting ones among those stemming from the nonmetric aspects of the passing of time. It is a way of enduring that consists in dwelling. Accumulation, reiteration, permanence are nearby characteristics of this major feature. These features of stability contribute to increasing the degrees of efficacy of institutions and of norms considered above. They can be inscribed on a scale of modes of temporality parallel to the scale of degrees of efficacy and of constraint. Pierre Bourdieu’s category of habitus, which has a long history behind it going back to Aristotelean hexis, its medieval reinterpretations, and its being taken up by Panofsky and above all by Norbert Elias, needs to be placed on this scale. Below, I shall demonstrate the fruitfulness of this category within the dialectical framework of the pair memory/forgetting. But we can already say that it gains from being paired with the temporal aspects of the highly antihistorical categories used by Norbert Elias in The Court Society.
Stability as a mode of social change is to be paired with security, which relates to the political level. These are, in effect, two neighboring categories on the scale of temporal modes. They both have to do with the aspect of the enduring and permanence of the social bond, considered sometimes from the point of view of its truth force, sometimes from that of its authority. The force of ideas has multiple modes of temporalization.
When placed within a dynamic polarized field, these categories call for a counterpart on the side of the appropriation of values stemming from the field of norms. This counterpart, this counter image, may belong to the order of chance, of mistrust, of suspicion, of defection, of denunciation. The category of uncertainty, upon which microhistory sets a premium, is inscribed within this same register. It has to do with the aspect of the trustworthiness of representations on the way to stabilization. It is the most polemic category, oscillating between tearing apart and weaving together the fabric of the social bond. Strategies aimed at reducing uncertainty testify eloquently that it must not become in turn a nondialectical category, as could that of the invariant.89 “Over time,” says the author of Le Pouvoir au village, “every personal and familial strategy tends, perhaps, to become blunted enough to blend in with a common outcome of relative equilibrium” (xiii). “The strategic utilization of social rules” by the actors seems to imply a noteworthy use of the causal relation that would be the tendency toward optimization over the course of action. It plays out on both the horizontal axis of living together and on the vertical one of scales of efficacy and temporalization inasmuch as the social game affects the whole framework of relations between center and periphery, between capital and local community; in short, the power relation whose hierarchized structure is unsurpassable.90 That this strategic logic can in the final analysis be inscribed in the interplay of scales of appropriation is the most important conclusion benefiting a history of representations. The search for equilibrium can even be assigned to one precise temporal category, as Bernard Lepetit proposes, namely, the present of the social agents.91 By the present in history we are to understand something other than the short time of the hierarchies of nested time spans, a state of equilibrium. “The ravages of defection, or even of defiance and generalized imitation are contained in it through the existence of conventions that delimit in advance the field of possibilities, in this framework assuring the diversity of opinions and behaviors, and allowing for their coordination” (Les Formes de l’expérience, 277). Hence we can say: “The adjustment between the individual will and the collective norm, between the intended project and the characteristics of the situation at that moment, take place in the present” (279).92 Of course, not everything historical can be included within situations of conflict or denunciation. Nor do they all come down to situations of the restoration of confidence through the creation of new rules, through the establishment of new uses, or the renovation of old ones. These situations only illustrate the successful appropriation of the past. Inadaptation contrary to the fitting act, too, stems from the present of history, in the sense of the present of the agents of history. Appropriation and denial of relevance are there to attest that the present of history does include a dialectical structure. Still, it was not unprofitable to emphasize that an investigation into scales of time spans is not complete until we take into account the historical present.93
THE DIALECTIC OF REPRESENTATION
At the end of this journey through the adventures of the “mental” in the historical field, it is possible to explain, even to justify, the slow shifting from the term “mentalities” to that of “representations” in the vocabulary of historiography during the last third of the twentieth century.
The threefold development I have proposed for the notion of a variation in scale—beyond scales of observation and analysis—already sets us on the way of what will reveal itself to be the dialectic of representation. With regard to the variations in efficacy and constraint, the old notion of mentalities appears to be unilateral due to lack of a corresponding term on the side of the receivers of social messages; with regard to the variations in the process of justification at work across the plurality of cities and worlds, the notion of mentality appears undifferentiated, due to the lack of a plural articulation in social space; and finally with regard to the variety equally affecting the least quantifiable modes of temporalization of social rhythms, the notion of mentality seems to operate in a heavy-handed fashion, like the quasi-immobile structures of the long time span, or of cyclic conjunctures, the event being reduced to a function that indicates a break. Therefore, over against the unilateral, undifferentiated, massive idea of mentality, that of representation expresses better the plurivocity, the differentiation, and the multiple temporalization of social phenomena.
In this respect, the political field offers favorable terrain for a rule-governed exploration of phenomena falling under the category of representation. Under this term, or that of opinion, and sometimes ideology, these phenomena lend themselves to operations of denomination and definition in some cases accessible through the method of quantifying the data. René Rémond’s Les Droites en France offers a noteworthy example of such a systematic explanation combining structure, conjuncture, and events.94 And a counterexample is thereby brought against the heavy-handed accusation of unscientific conceptual vagueness directed against the notion of representation.95
With this threefold impetus, the notion of representation develops in turn a distinct polysemy that risks threatening its semantic relevance. In fact, we can make it in turn assume a taxonomic function—it will contain the inventory of social practices governing the bonds of belonging to places, territories, fragments of the social space of communities of affiliation; and a regulative function—where it will be the measure for evaluating or judging socially shared schemes and values, at the same time that it will mark the fault lines that sanction the fragility of the multiple allegiances of social agents. The idea of representation then runs the risk of signifying too much. It will designate the multiple trajectories of recognition between individuals and between the group and the individual. Thus it will rejoin the notion of “worldviews” that, after all, figures among the antecedents of the idea of a mentality.96
Given the threat of this hemorrhaging of meaning, it seemed to me opportune to narrow the gap between the notion of representation as an object of the historian’s discourse and two other uses of the same word in the context of the present work. In the following chapter, we shall be confronted with the notion of representation as the terminal phase of the historiographical operation itself. It will be a question not only of the writing of history, as is too often said—history is through and through writing, from archives to history books—but also of the transferring of explanation/understanding to written words, to literature, to a book offered for reading by an interested public. If this phase—which, let me repeat, does not constitute one step in a sequence of operations, but a moment that only didactic exposition places at the end—merits the name of representation, it is because, in this moment of literary expression, the historian’s discourse declares its ambition, its claim to represent the past in truth. Below we shall consider in greater detail the components of this ambition to be truthful. The historian in this way finds himself confronted with what appears at first to be a regrettable ambiguity of the term “representation,” which depending on context designates as a rebellious heir of the idea of mentality the historian’s represented object, and as a phase of the historiographical operation, the operation of representing.
In this regard, the history of reading gives the history of representations the echo of their reception. As Roger Chartier has amply demonstrated in his works on the history of reading and of readers, the modes of public and private reading have affected the sense of the understanding of texts. For example, new modes of transmitting texts in the age of the “electronic representation”—the revolution in the reproduction and in the medium of a text—lead to a revolution in the practice of reading and, through this, in the very practice of writing.97 In this way the loop of representations closes in on itself.
A hypothesis then comes to mind: Does the historian, insofar as he does history by bringing it to the level of scholarly discourse, not mime in a creative way the interpretive gesture by which those who make history attempt to understand themselves and their world? This hypothesis is particularly plausible for a pragmatic conception of historiography that tries not to separate representations from the practices by which social agents set up the social bond and include multiple identities within it. If so, there would indeed be a mimetic relation between the operation of representing as the moment of doing history, and the represented object as the moment of making history.
Furthermore, historians, little habituated to situating their historical discourse in terms of the critical prolonging of personal and collective memory, are not led to bringing together these two uses of the term “representation” in relation to what I have called a more primitive one, unless it is in the order of thematic reflection, at least as regards the constitution of the relation to time, that is, in terms of the act of remembering. This too has its ambition, its claim, that of representing the past faithfully. The phenomenology of memory, from the time of Plato and Aristotle, has proposed one key for the interpretation of the mnemonic phenomenon, namely, the power of memory to make present an absent thing that happened previously. Presence, absence, anteriority, and representation thus form the first conceptual chain of discourse about memory. The ambition of the faithfulness of memory would thus precede that of truth by history, whose theory remains to be worked out.
Can this hermeneutic key open the secret of the represented object, before penetrating that of the operation of representing?98
Some historians have thought about this, without leaving behind the framework of the history of representations. For them, what is important is actualizing the reflective resources of social agents in their attempts to understand themselves and their world. This is the approach recommended and practiced by Clifford Geertz in The Interpretation of Cultures, where as a sociologist he confines himself to conceptualizing the outlines of self-understanding immanent to a culture.99 The historian can also undertake to do this. But can he do so without providing the analytic instrument that this spontaneous self-understanding lacks? The answer can only be negative. Yet the work thus applied to the idea of representation does not surpass the privilege of conceptualization that the historian exercises from one end to the other of the historiographical operation, hence from reading the archives to writing the book, in passing through explanation/understanding and its literary organization. Therefore there is nothing shocking in introducing into the discourse on the represented object fragments of analysis and of definition borrowed from another discursive domain than history. This is what Louis Marin, Carlo Ginzburg, and Roger Chartier authorize themselves to do.
Chartier, in examining Furetière’s Dictionnaire universel (1727), discovers in it the outlines of the bipolar structure of representation in general, namely, on the one hand, the evocation of an absent thing through the intervention of a substituted thing that is its representative by default; on the other hand, the exhibiting of a presence offered to the eyes, the visibility of the thing present tending to overshadow the operation of substitution that is equivalent to an actual replacement of what is absent. What is astonishing about this analysis is that it is strictly homogeneous with the one proposed by the Greeks for the mnemonic image, the eikōn. However, to the extent that it takes place on the terrain of the image, it ignores the temporal dimension, the reference to the earlier, essential to the definition of memory. On the other hand, it lends itself to a unlimited expansion on the side of a general theory of the sign. This is the direction developed by Louis Marin, the great exegete of the Port Royal Logic.100 Here the relation of representation is submitted to a labor of discrimination, of differentiation, accompanied by an effort at identification applied to the conditions of intelligibility likely to avert mistakes, misunderstandings, as Schleiermacher will later also do in his hermeneutics of the symbol. It is in terms of this critical reflection that we can understand the use and abuse resulting from the priority accorded to the visibility belonging to the image over its oblique designating of the absent. With this point, conceptual analysis turns out to be useful for exploring the illusions resulting from the cooperation that weak belief grants to strong images, as we find in Montaigne, Pascal, and Spinoza. The historian finds aid in these authors for exploring the social force of the representations attached to power, and can thus enter into critical rapport with Norbert Elias’s sociology of power. The dialectic of representation adds a new dimension to the phenomena discussed above in terms of scales of efficacy. It is this efficacy per se that benefits from a supplementary degree of intelligibility applied to the idea of the absence of physical violence once it is both signified and replaced by symbolic violence.
Carlo Ginzburg, in responding to Chartier’s article in “Représentation: Le mot, l’idée, la chose,” fleshes out the dialectic of substitution and visibility pointed out by Furetière through a wide-ranging deployment of examples resulting from his erudite researches.101 Essentially, it is a question of ritual practices linked to the exercise and manifestation of power, such as the use of a royal mannequin in royal funerals in England, or the empty coffin in France. Ginzburg sees in these manipulations of symbolic objects the simultaneous illustration of substitution in relation to some absent thing—the dead person—and the visibility of the present thing—the effigy. Little by little, voyaging in time and space, he evokes the funerals of images in the form of incinerated wax figurines in Roman funerary rites. From there, he passes to the modes of relationship as much toward death—absence par excellence—as toward the dead, those absent ones who threaten to return or who are endlessly in quest of a definitive resting place, by way of effigies, mummies, “idols of a colossus,” and other statutes.102 Unable, as a historian, to give an overall interpretation of this “changing and often ambiguous status of the images of a given society” (“Représentation: Le mot, l’idée, la chose,” 1221), Ginzburg prefers to respect the heterogeneity of his examples, aside from ending his essay with an unanswered question regarding the very status of his own research project: “Do we have to do with the universal status (if there is one) of a sign or image? Or rather with a specific cultural domain—and, in this case, which one?” (1225). To conclude, I want to discuss this hesitation on the historian’s part.
One of the reasons for his prudence has to do with his recognition of one troubling fact: “In the case of the status of the image, there was, between the Greeks and us, a deep rupture, which I am going to analyze” (1226). This rupture was a result of the triumph of Christianity, which opened between the Greeks, the Roman emperors, and us the break signified by the cult of martyrs’ relics. One can to be sure speak in general terms of the close association between images and the beyond, but the opposition remains strong between forbidden idols, to which Christian polemics reduced the images of the ancient gods and deified persons, and the relics proposed to the faithful for devotion. The heritage of medieval Christianity concerning the cult of images has in turn to be taken into account, and, thanks to a detour through an absorbing history of iconography, a distinct place must be reserved for the practice and theology of the Eucharist, where presence, that major component of representation, is charged with signifying not only something absent, the Jesus of history, but the real presence of the body of the dead and resurrected Christ, beyond its memorial function as regards a unique sacrificial event. Ginsburg does not dwell on this complicated history, but rather ends his inquiry into the Eucharist at the first third of the thirteenth century. Nevertheless, in ending he sets up a narrow bridge between his exegesis of the effigy of the king and the real presence of Christ in the sacrament.103
Here is where Louis Marin takes over.104 He is the irreplaceable exegete for what he takes to be the theological model of the Eucharist in terms of a theory of the sign at the heart of a Christian society. Port Royal was the elect place where a semiotics or logic of the statement—“this is my body”—and a metaphysics of real presence were constructed and exchanged values.105 But Marin’s contribution to the vast problem of the image is so considerable that I am resolved to take it up in a more complete way in the following chapter, inasmuch as it clarifies the use of representation in historiographical discourse in a brighter light than the self-understanding that social agents have of their own practice of representation.
We can observe in the works that precede Marin’s last book, Des pouvoirs de l’image, a hesitation between two uses of a general theory of representation.106 The double definition of representation he proposes would fit as well with a theory of the represented object as with that of the representing operation. This definition recalls that of Furetière: on the one hand, “presentation of the absent or the dead” and, on the other hand, “self-presentation instituting affect and meaning in the subject of gaze” (18). This proposition fits equally well with the literary expression of historiography, concerning which I shall say more below, and the social phenomena that previously were put under the heading of a history of mentalities. One can say in the first place that the historian seeks to represent to himself the past in the same way social agents represent the social bond and their contribution to this bond to themselves, in this way making themselves readers of their existing and their acting in society, and in this sense historians of their own present. However, it is the efficacy of the social image that prevails in Des pouvoirs de l’image: “The image is both the instrument of force, the means of strength, and its foundation as powerful” (18). But in linking the problematic of power to that of the image, as was already evident in Portrait of the King, Marin clearly tips the theory of representation toward the side of examining its social efficacy.107 We are back in a region visited in another way by Norbert Elias, that of symbolic struggles where the belief in the force of signs has been substituted for the external manifestation of force in a fight to the death. Pascal can again be recalled here, no longer in terms of the aura of the semiotics of the Eucharist and the real presence, but in the wake of a denunciation of the “apparatus” of the powerful. In this regard, the sketch of a theory of imagination in the Pensées was already that of a theory of symbolic domination. It is here that the theory of the reception of written messages, with its episodes of rebellious and subversive readings, could allow the theory of symbolic violence laid out in Des pouvoirs de l’image to rejoin the investigations proposed above on the responses of social agents to the pressure of injunctions projected in their direction by different instances of authority. In this regard, does not the kind of forgetting linked to the replacing of brute force by the force of images metonymically attached to the exercise of such authority not constitute the implacable corollary to this “power of the image”? Marin’s last book opens another way, where the competition between text and image comes to the fore. The theory of representation tips once again toward the side of the literary expression of the historiographical operation.
I want to interrupt, rather than to conclude, this section by expressing a perplexity: Can a history of representations by itself attain an acceptable degree of intelligibility without openly anticipating the study of representation as a phase of the historiographical operation? We have seen Ginzburg’s own perplexity, caught between a general definition of representation and the heterogeneity of examples that illustrate the competition between the evoking of absence and the exhibiting of presence. This confession is perhaps what fits best with a treatment of the represented object, if it is true as I am supposing here that it is the historian’s actual reflection on the moment of representation included within the historiographical operation that leads to the explicit expression of the understanding that social agents have of themselves and of the “world as representation.”