Eight
Violence in the French Revolution: Forms of Ingestion/Forms of Expulsion

Brian Singer

Like all writings, this chapter has a varied provenance. But the one "gestative line" that I should like to pursue by way of introduction concerns the difficulties in thinking about violence. Acts of violence immediately raise a number of moral issues, which tend to determine the direction of enquiry. To speak of violence, it is feared, is to justify violence—unless such speech is explicitly directed at its condemnation. But to condemn violence is to place it under a subtle (but no less strict) form of censure. And in order to break the censure, and thereby open up violence as a topic of discussion, violence must be tamed theoretically, as well as restricted practically. Generally, this is done by binding violence to the modes of instrumental rationality, even as violence by itself appears as what is most threatening to rationality in all its forms.

No less a thinker than Hannah Arendt has claimed that the essence of violence, its very substance, lies with its instrumentality.^[1] Violence is a means to an end. Thus violence will receive the right of entry into a utilitarian calculus. Too little violence, and the end may not be achieved; but too much violence is morally wasteful, when not simply counterproductive. Violence may be one of several means to the realization of an end, in which case its costs must be measured against those entailed by the use of the alternate means. Or it may be only means to the realization of that end, given the character of the resistance of those opposed to that end. Here the investigation of violence is limited to its uses, its effectiveness, its necessity. It is a matter of the establishment of a calculus to establish the judicious (if not the just) use of violence. An instrumental rationality brackets—though only partially—the question of ends in order to produce a morality of the means. When "good" violence can be separated from "bad" violence, violence can be rendered, at least to some extent, acceptable.

Of course, the investigation does not stop here. The discussion of the "rational" employment of violence soon moves to consider the "rationality" of the actors and their motives. The "rationality" of the actors: their social background, their institutional affiliations, their relation to the law—in short, their "respectability." The "rationality" of the motivations: Are the actors consciously striving for concrete goals worth attaining?—here the question of ends returns, but in the more manageable form of (material) interests. And again, in both cases violence can be justified, if not necessarily legitimated, in the name of pragmatics.

In truth, the attempt to establish rules for the rational use of violence has a long history, dating back at least as far as Machiavelli. After all, such an enquiry does have eminently practical implications for the practice of politics. And it enters, almost necessarily, into our most casual judgments when confronting violent events. Yet the instrumentalist calculus of violence is fraught with all kinds of problems—even should the delicate balance between means and ends not be upset (when either the means—the violence—are judged unacceptable independent of the ends, or the realization of the ends is considered so imperative that all means employed on their behalf are condoned).

Suppose, however, that the balance is maintained. There remains the problem that violence can enter a utilitarian calculus only in a stable, narrowly circumscribed situation. Numerous suppositions follow: at the very least, the ends must be limited, the consequences foreseen, and the contingencies held at bay. And one will want to argue that such suppositions can never be met in toto, that the "virtues" of violence can never be entirely ensured in the face of fortuna . And what applies in general, applies with far greater force to the type of situations that we shall be examining: a revolutionary situation. For in the latter, as we shall see, the violence is not always directed at a well-defined enemy who poses some clear and immediate threat. Nor is the violence, ultimately, directed at some limited, short-term end, but to the creation of a new regime, a new humanity. And above all, revolutionary situations entail a state of continuous and extreme flux: it is not just that circumstances are always changing, or that the rules of the game are constantly being rewritten; the fact is the very definitions of what is real and unreal, possible and impossible, true and false, are all being shifted from their moorings.

But beyond the question of whether the instrumentalist perspective can deliver an adequate portrayal of the reality of violence (and without considering whether the actors understand their violent acts in purely instrumental terms), there are the moral questions raised by this perspective. In truth, the latter does not really seek to provide a pure description of reality and will admit as much in rare moments of self-reflection; on the contrary it attempts to describe violence not as it is but as it ought to be when "ration-

ally" applied. And yet at the same time, in the name of a certain "realism," it would denude that realm of the "ought" of its utopian dimension. The result is that an ideal violence is substituted for the ideals that violence sometimes claims to serve.^[2] In the attempt to be at once "realist" and "idealist," the perspective proves neither realistic nor, to be sure, idealistic: at worst, the artificially perfected image of an imperfect world with which to construct a morality for the amoral.

This is not to mention those discussions of the uses of violence which shift the terms, slipping from an analysis of the instrumentality of violence to an analysis of its "functionality." Here, where one speaks of the symbolic uses of violence in terms of cathartic release, social solidarity, or boundary definition and maintenance, or in terms of its regenerative powers, violence becomes an end in itself, opening up a possible whirlpool of apologetics.

To consider violence as, essentially, an instrument is not, as is so often hoped, to render it morally neutral. It does, however, tend to make violence theoretically transparent. Instrumentalized, violence appears without any interest in itself outside of the means/end context in which it is inserted. It is reduced to purely quantitative terms, a medium of action whose only content is the degree of force it embodies. As a tool, violence lacks any expressive capacity; as the result of a rational choice, violence, even if employed by institutions, cannot itself take on an institutional character. Within such a perspective, violence is denied any relation to culture. There can be no concern with the forms of violence, with the possibility that certain forms of violence are characteristic of certain kinds of society. In a word, to treat violence instrumentally is to choke off the possible elaboration of what might be termed a sociology of violence.

Now the elaboration of such a sociology is no easy matter. It will, relative to the commonsense understanding of violence regarding "the logic of the social," appear to engage in a number of paradoxes. For just as violence first appears opposed to rationality, violence also appears contrary to society, or more precisely, to its existence in terms of institutions, forms, and expressions. Let me explain. First, violence as the negation of form: generally speaking, violence is seen as either the primordial chaos out of which form develops, or as the final chaos in which form is to be engulfed. Second, although violence may not necessarily be opposed to expression, it does appear as opposed to communication: either it appears as prior to language, expression without articulation, the utterance of a brute subjectivity, or as posterior to language, where communication, exhausted of its resources, breaks down. Finally, violence as contrary to institution: where the latter supposes not just form but the continuity of form, violence appears as the disruption of routine, a rupture in the temporal continuum, the introduction of an unstable, unpredictable, and uncontrollable element threatening the social order.

To elaborate a "sociology" of violence, then, is to restore to violence, or at least to some violence, its character as institution. It is to understand violence as form, as a "cultural" form if one will, as a social "ritual" as well as a collective instrumentality. And it is to examine the different forms of violence as instituted in different societies as, in each case, telling us something about that society, as expressive of that society, and of its "self-understanding." One might wish to see such a sociology as providing a hermeneutics of violence through which the larger "social text" can be interpreted. Alternately, the remainder of this chapter may be read as an introduction to a possible history of violence.

The discussion begins with a brief examination of George Rudé's well-known work on popular violence, The Crowd in the French Revolution . This book may be considered as representative, though not in any rigorous sense, of the instrumentalist approach. My own, more "sociological" approach will emerge out of a critique of this work. Because I am following here on the heels of Rudé, and because of limitations of time, space, and research, my analysis will be based almost exclusively on Parisian events. However paradigmatic the Parisian case, I recognize this as a limitation, and all conclusions should, therefore, be treated tentatively.

Rudié's Populist History

George Rudé begins his book by proposing that the revolutionary crowds cannot be treated as abstractions. Though the abstractions of the left are criticized, that is, the image of the popular movements as embodying the "People," it is soon obvious that it is the abstractions of the "right" which are his real target.^[3] Here he is speaking of the image of the revolutionary crowds as "inchoate mobs," anarchic and blindly destructive, "drawn from criminal elements or the dregs of the city population." Or to quote the rhetoric of one Edmund Burke, "bands or cruel ruffians and assassins, reeking with . . . blood" and embodying "all the unutterable abominations of the furies of hell."^[4] George Rudé seeks to undermine these "abstractions" ("stereotypes" seems the more appropriate word) by pursuing, at least implicitly, three lines of enquiry: the first is concerned with who commits the violence, their quality as social actors; the second with their motivations and intentions, the reasons for the violence; and the third with the uses of violence, its instrumentality as a means for the realization of the ends posited by the intentions and motivations. These "lines of enquiry" then can be considered as forming the book's underlying "theoretical" framework. Though in truth, this is not a theoretical work, and the underlying framework hardly constitutes its originality, let alone its importance. If we must be eternally grateful to Professor Rudé and his travails, it is for having brought empirical

data to the discussion of popular revolutionary violence, and thus proof positive to his counterdemonstration.

Consider then the more concrete lines of investigation that, roughly speaking, correspond to the above lines of enquiry. Again we can speak of three lines. In the first place, throughout the book, while relating the history of the crowd's actions, George Rudé provides a body count, showing that the revolutionary crowds did not engage in large-scale, uncontrolled bloodletting. Generally speaking, the number of victims during the journées révolutionnaires —and by victims, one understands the unarmed and defenseless, and not the Swiss Guards killed in pitched battle—were few, rarely more than a handful, hardly the stuff to fuel the fires of counterrevolutionary nightmares.

In the second place—and this in many ways is the core of the book's research—he demonstrates that the revolutionary crowds were not composed of the unemployed, the underemployed, the petty criminal, the marginal, the vagabond, those who were called, in a language still resonating with the feudal past, les gens sans aveu (literally, those who had not taken an oath, and thus had no clear place within the social order, however lowly, those therefore who were the very embodiments of a social dis-order). This "underclass," it seems, were hardly major participants on the stage of the Revolution in any real sense. The revolutionary crowds were largely composed, to be sure, of sans-culottes (literally, those without breaches, that is, those who wore trousers—the wearing of stockings up to the knee being a major mark of the social divide of the time). But what George Rudé demonstrates is that it was the most stable, most law-abiding elements of the sansculottes who participated in the revolutionary events, those whose remnants some fifty years later would be considered petty-bourgeoisie or, if one will, lower middle class. Though participants from higher classes—that is, bourgeois, rentiers , merchants, civil servants, and professionals (breaches and all)—as well as participants from the lowest classes, were not entirely absent, it was the workshop masters, craftsmen, shopkeepers, petty traders, journeymen and, more rarely, wage earners who predominated.^[5]

Lastly, the author considers the motives for participation in the revolutionary crowds. It comes as no surprise that he finds that "bribery and corruption," or "the quest for loot," were not "major factors stimulating revolutionary activity."^[6] Nor does he consider the revolutionary crowds to be acting from the irrational instincts posited by the crowd psychology inaugurated by Gustave Lebon.^[7] Such right-wing shibboleths are quickly dismissed. He does admit that the revolutionary crowds acted, in part, out of political motives, having "enthusiastically supported and assimilated the objects, ideas, and slogans of the political groups in the National Assembly, Cordeliers, and Jacobin Clubs whose leadership they acknowledged and in whose interest they demonstrated, petitioned, or took up arms."^[8] But it is not simply a matter of the sans-culottes passively following the initiatives of the republi-

can bourgeoisie. The sans-culottes had their own reasons for engaging in street action, reasons that were foreign to the interests and experience of their erstwhile bourgeois allies, and which account for the force and continuity of the social ferment of the revolutionary period. The sans-culottes' primary concerns, Professor Rudé asserts, were with more basic issues: those dealing with the provisioning of cheap and plentiful food in the face of shortages and high prices. And he goes on to note that the reasons for their actions were continuous with a long history of popular disturbances, dating back at least several centuries. These were not new struggles, seeking the establishment of a new, hitherto unknown order. "[T]he sans culottes intervened" he states, " . . . to reclaim traditional rights and to uphold standards which they believed to be imperilled by the innovations of ministers, capitalists, speculators, agricultural 'improvers,' or city authorities."^[9] These were defensive reactions, or in the words of Charles Tilly, "reactive struggles"; they sought to defend tradition and community against the threats posed by the newly emergent state and economy. They were the type of popular struggle that prevailed during the period prior to the consolidation of the nation-state and capitalist economy, after which new struggles of "proactive" character would arise.^[10]

Without delving, into this last point, which, despite its considerable interest, is not germane to my discussion—and which, in fact, is not central to George Rudé's discussion, being the horizon to which the latter points toward its end—let us very briefly summarize his argument. The revolutionary crowds, he seems to be saying, were largely composed of the lower classes to be sure, but of their more respectable elements. These crowds engaged in revolutionary actions for reasons that were reasonable, if somewhat archaic. Lastly, respectable and reasonable, these actions, as measured by the level of violence, were relatively moderate. This is a sympathetic portrait. The three lines of investigation lead, both implicitly and explicity, to a defense of the sans-culottes' revolutionary actions. The book is meant to be a history from below; and indeed the bulk of the book is composed of a chronicle of the Revolution's history as presented by a history of the journées révolutionnaires. The villains of the piece are those located above. Notably, the first two estates, that is, the aristocracy, the clergy, and other counterrevolutionaries. But also, if to a slightly lesser degree, the republican bourgeoisie and their state. George Rudé follows here the historiography of Albert Soboul wherein the sans-culottes and the revolutionary bourgeoisie are posed as, depending on the period, the best of allies and the worst of friends. The Jacobin state, in particular, is condemned for completely eliminating the sans-culottes' capacity for autonomous action, thus cutting itself off from its popular base of support, and mounting a Terror that was excessive and arbitrary. In short, this is a history of popular movements that would also be a populist history.^[11]

Now, my problem with the book lies not with its populism, nor with the morality tale hidden therein. What I find difficult to accept is that the "popular" actors must, in order to be rended historically acceptable, be made into rational actors. With the exception of their context, they would be like us, or like we would like to believe ourselves to be. One would hardly think that they came from another time, let alone another culture. For them violence would be a mere resource, something to be mobilized, in itself neutral, transparent.

This is to say that it is not my intention to criticize Professor Rudé's characterization of the social composition of the revolutionary crowds. Nor will I dispute his characterization of their motives, at least directly. My intention is not to attack Professor Rudé on his chosen terrain but to shift the terrain and open up new areas for discussion.

The September Massacres

Let me begin with a brief comment concerning the images of popular violence against which George Rudé directs his argument. This imagery was not invented by right-wing historians after the revolutionary events; the perception of the revolutionary mob as composed of the lowest, least-stable elements of the population was common to the time. And it was not just the wealthy and counterrevolutionary who feared the vast underworld of the desperately poor and criminal. As George Rudé is not unaware, almost all sectors of society, including the revolutionary, and the revolutionary sansculotte, shared the same fears relative to les gens sans aveu; and almost all sectors were equally willing, when confronted with social disturbances contrary to their perceived interests, to blame the same people. In other words, the stereotyped images of the revolutionary mob constitute less a historical argument, or even a political argument, than a widely shared social prejudice, both then and now and in times intervening. As such, this prejudice must be considered part of the reality it purports to describe; for it is part of the motivations and intentions of the Revolution's actors; indeed it seems to have a social dynamic of its own, almost independent of the multiple uses to which it was put. On more than one occasion, it was rumors, almost always false, concerning the least favored part of the population, that set the popular movements in motion. What historians refer to as the Great Fear, the peasant insurrections of the summer of 1789 which swept away the old rural order, was precipitated by rumors that Paris was expelling its surplus population, casting into the countryside a confused and hungry mass that would pillage farmer and farm alike. And again it was "lesser fears" that were the catalyst of several of the journées révolutionnaires in Paris, the sans-culottes rising in self-defense against imaginary threats posed by the same desperate elements, presumed in the pay of some counter-

revolutionary conspiracy. In truth, as we shall see, this most unfortunate sector of the population was more likely to be the victims of popular violence than its perpetrators. As a target, the object of George Rudé's critique is simultaneously weak and resilient: weak in the sense that no great imagination is required to see that the stereotypes are largely false, the result of what is generally known as scapegoating; and resilient in the sense that for this very reason they will always be with us.

There is another point that must be brought out here. How can Professor Rudé speak of the revolutionary actors as reasonable when those he would defend hold the very prejudices that he would attack? Or better, how can one speak, whether implicitly or explicitly, of rational actors, when the grounds of their actions lie within a social imaginary composed, at least in part, of rumors and hearsay?^[12]

Beyond all questions concerning the actors, their social composition, their "rationality," what can be said about their violence? George Rudé as suggested, leaves an impression of moderation, but only by neglecting, if not entirely ignoring, what are known as the September Massacres of 1792. The ostensible reason for his neglect has to do with the paucity of evidence relative to those who participated in the massacres. No one wanted to admit their participation after the event for, unlike the other journées revolutionnaires, there was little to gain and much to lose in any such admission.^[13] The only real data comes three years after the massacres, and one year after the fall of Robespierre and the Jacobin party, when judicial proceedings were brought against thirty-nine persons for alleged participation in the massacres—and then all but three were acquitted for lack of evidence.^[14] However, despite the lack of hard data relative to social composition, the character of the event remains clear. And the character of the September Massacres was so exceptional, that it must be considered in some detail.

What immediately strikes the observer about the September Massacres is the extent of their violence. The number of victims—again among the unarmed and defenseless—exceeded by far anything that had occurred previously. George Rudé provides a maximum estimate of 1,400 executed, and this figure has not been exceeded by more recent estimates.^[15] Admittedly, the September Massacres occurred under exceptional circumstances. Paris, it then seemed, was threatened with foreign invasion, the leaders of the invading forces had threatened widespread reprisals against the revolutionaries, and the Parisian populace was, understandably, affected by a nervous ferment. And yet the massacres appear not to have been a panic reaction. They appear to have been, at least in part, premeditated, and they unfolded in relative calm. The ostensible reason for the massacres was the presence in the prisons of numerous counterrevolutionaries, many of them rounded up during the last journée révolutionnaire, August 10, when the monarchy had fallen. There were rumors of a complot des prisons , a conspiracy amongst the

prisoners against the Revolution,^[16] and on the second of September the prisons were invaded by the revolutionary crowds. However, it must be said that the massacres were not limited to the prisons, or by what one usually understands as prisons, and the largest number of the victims was not counterrevolutionaries. At least 70 percent of those executed were common-law prisoners: common thieves, prostitutes, forgers, vagrants, and the like—that is, precisely, the gens sans aveu, the eternal scapegoats discussed above.^[17]

In this regard, it is perhaps surprising that George Rudé seeks to serve the September Massacres with reasons—reasons, one might add, not far removed from those given by the septembriseurs themselves. For he writes that "[u]nsavoury as the episode must appear in itself, the massacres . . . completed the destruction of the internal enemy some weeks before" the defeat of the external enemy at Valmy.^[18] One wonders in what sense most of the victims were enemies of the Revolution, and in what sense those who were enemies of the Revolution, being in prison, were threats to the Revolution. Even if there was a complot des prisons—for which there is no evidence—it should hardly have to be said that it would have been highly unlikely that these prisoners could have (1) escaped the prisons, (2) armed themselves, and (3) successfully confronted an armed and vigilant, revolutionary populace. Born in fear, fueled by rumor, and directed at a wide, imprecise array of victims, this journée révolutionnaire, perhaps more than any other, cannot be reduced to the terms of a means-end rationality, let alone justified in such terms. The violence was, in a sense, "too total."

Now George Rudé is not a "rational-choice" theorist in any rigorous sense. He does not seek to demonstrate that the massacres resulted from the rational pursuit of the self-interests of the massacreurs , nor that the means chosen in the pursuit of these interests were perfectly adequate to the designated ends. He does, however, implicitly believe that the revolutionary actors were reasonable and that their actions are to be understood as such. And this belief results in some rather blatant revolutionary apologetics.^[19] In claiming the "utility" of the massacres, George Rudé is perhaps justifying less the rationality of the events than the "rationality" of the Revolution itself, or at least of the "popular Revolution," understood as a whole that must be rescued from the taint of ambiguity and defended in all its peripeties. Here, most definitely, I must part company with Professor Rudé and let him continue to fight, some two hundred years later, the Revolution's battles, and with almost the very same arguments.^[20]

The importance of the September Massacres does not simply lie with the number of its victims. Many historians see the massacres as marking the beginning of the end of popular violence. Already several weeks prior to September, Danton, then the minister of justice, had declared: Que la justice des tribunaux commence, et la justice du peuple cessera .^[21] And after the massacres, almost all parties with influence on the government agreed that a special

apparatus had to be established to carry out revolutionary justice and monopolize the violence carried out in its name. Accordingly, the events of September are seen as marking the last important outbreak of what is sometimes called the Popular Terror, which was to be replaced by the official Terror of the Revolutionary Government. As such, a line is drawn after the September Massacres, distinguishing popular violence from state violence, the latter being explicitly presented as a means to stop the "popular executions," while satisfying the sans-culottes' desire for "prompt justice."

My account is going to be somewhat different. I will not situate the rupture after the September Massacres with the construction of a governmental apparatus specializing in "revolutionary justice."^[22] Without denying the value of the more common account, I shall place the September Massacres themselves at the point of rupture. For my rupture concerns not those who carried out the violence, but the forms of that violence.

Forms of Violence

I shall try to give some idea of what I mean by "forms" by briefly describing the crowd violence that accompanied the previous journées révolutionnaires, as compared with the violence of the September Massacres.^[23]

Popular violence prior to the September Massacres made a spectacle of the victim and his mutilation. Examine the typical course of events: the sansculottes are in a state of insurrection. Crowds form on the street. Someone who, by position or reputation, excites popular hatred, usually a priest or aristocrat, is seized and threatened with being lanterné , that is, hung from the nearest lamppost. At this point someone tries to intervene, usually an agent of the government, displaying the symbols of his office and of the government's revolutionary legitimacy. This person (or persons) attempts to reason with the crowd and remove the potential victim from the clutches of his or her tormentors. While admitting the possible culpability of the object of the crowd's wrath, he speaks the language of the Law, claiming that the forms of due process must be followed and a proper trial held. This discourse is often shouted down, in any case ignored, and the person who gave it is forced to withdraw. If it was an agent of the government, a report will be written up, relating the efforts of the authorities and the failure of these same efforts—a report that can then be presented in the case of a judicial enquiry. Once the law, and those who speak in its name, have been explicitly rejected, the crowd makes good its threat. The poor wretch is indeed strung up from the nearest lamppost and, once dead, the head is separated from the body and placed at the end of a pike. An evisceration then takes place, with various organs also finding themselves at the end of pikes. These will then form the front of a somewhat macabre procession. A rope will be tied to the foot of the body, which will be dragged behind. This procession will tour the major

streets of the city, visiting key public places, as well as, on occasion, places of particular significance to the victim. The procession will be accompanied with much rejoicing, gallows humor, and threats to do similar things to similar people. Witnesses to the procession will join in, shout words of encouragement, turn away, vomit in doorways, and some women will faint. Once the procession is over, the various pieces of the corpse will go on semipermanent display, usually near the gates to the city or section, or some other public place.

At this point, one might remember that George Rudé had spoken, relative to the violence of the revolutionary crowds, of their moderation. And he is right, if one speaks of the violence in purely quantitative terms. However, what little violence there was, was maximized as spectacle. Indeed, one suspects that it was because the choreography was so violent, that its quantity was, relatively speaking, restrained.

Consider now the character of the September Massacres. Again, one speaks of the "typical" case. The massacres did not occur in the street, they were not public, in the above sense. The objects of the crowds' wrath were the inmates of the prisons and other "total institutions" of the period; and the violence was restricted to the confines of these institutions, their courtyards, and immediate vicinities. It was not simply a general slaughter, conducted in complete anarchy, or at least not as the slaughter proceeded. Again the authorities, municipal and national, sent deputations to calm the crowds and to either stop or moderate the butchery.^[24] And again they failed and were forced to retreat, sometimes under the threat of physical violence. At about the same time, however, persons who are described as strangers to the sans-culotte milieu,^[25] but who were not agents of governmental authority, and without any mandate, stopped the executions, installed themselves behind a table, and organized "popular tribunals." In effect, justice was improvised. Juries were formed, judges named, a prosecutor established, the prison records obtained. The prisoners were then brought before this tribunal one by one, their identities and the reasons for their institutional confinement verified. A short interrogation followed, after which the prisoner would be declared either innocent or guilty. If declared guilty, there was but one penalty: the prisoner would be pushed beyond the threshold of a doorway, whereupon he or she would be bludgeoned or hacked to death with whatever instruments were available. The bodies of the victims were left to accumulate in a pile. A short while after the butchery had ended, the authorities carted the bodies to the outskirts of Paris, where they were buried in pits covered with chalk.

The contrast between the September Massacres and the outbreaks of popular violence that preceded it are striking. Two points stand out: the absence or presence of the mutilation and display of the corpses, and the role of the law. Let us consider the second of these points first.

We saw that during the acts of popular violence someone, usually someone official, would seek to have the crowd respect the due process of law, and that his demands would, inevitably, be rejected. Indeed, so often does the discourse of those who speak of the law get repeated and with always so few results, that one suspects that its real purpose lay elsewhere. Rather than being an attempt to save the victim, or to have justice served (in its forms at least), one suspects that this discourse served as a ritual of exclusion whereby the authorities separated themselves from the crowd and washed themselves of all responsibility for its actions.^[26] Such a course of action, or more precisely, nonaction, is not without its uses, but the result is an explicit demonstration of the law's impotence and an implicit admission of the government's unofficial tolerance for the ensuing events. And neither could be tolerated by many of the revolutionary leaders, particularly those of a more radical persuasion. They sought—and indeed, they believed one of the Revolution's central aims to be—to reduce the distance between the law and the people; and this meant not only making the law an expression of the sovereign will of the people, but bringing the people under the sovereign authority of the law. And not the least of the implications of this conjoining of law and people was the placing of all violence committed by the people in the name of the Revolution under the authority of the Law. An imperative made all the more necessary in their minds by the increasing demands on the part of sansculotte spokesmen for the execution of the Revolution's enemies, and by the increasing likelihood of more and more extensive outbreaks of popular violence.^[27] Now, as noted, during the September Massacres the "people" would deny the law as represented by the duly constituted authority of both the National Assembly and the Commune. And yet, they would not reject the authority of the law per se, as witnessed by the establishment of improvised tribunals with their makeshift justice. One might say that they rejected the official representatives of the law but not the law itself. In effect, what one is witnessing with the September Massacres is the transformation of popular violence into "popular justice."^[28] And as an act of justice, the massacres would not differ all that much, as regards their form, from the acts of an official justice, which would also have its tribunals, juries, prosecutors, and its examination of written documents, and so on. It is in this sense that the September Massacres, rather than marking the culmination of popular violence, introduced a rupture that anticipated the official Terror to come. And as the anticipation (and reflection) of a new, more regularized form of violence, the September Massacres can be said to mark the end of popular violence. Not the least of the Jacobins' claims to legitimacy was their promise to end the outbreaks of crowd violence. And in fact, the Jacobin insurrection of 31 May 1793 was the first journée révolutionnaire not accompanied by popular executions.^[29] The cadavers would mount under Jacobin rule, but what were called the scènes de horreur , "the scenes of horror," had ended.^[30]

This brings us to the second point of comparison: the mutilation and display of corpses. During the September Massacres, although the dismemberment of corpses was not entirely absent, nor was it the rule. Nor was there the parade of dismembered and mutilated bodies which had accompanied the earlier journées révolutionnaires. In short, the spectacle of violence was absent.^[31] In fact, the Parisian populace only became fully aware of the extent of the executions once the carts, carrying the bodies of the victims to the city's outskirts, threaded their way through the city streets—and it was at that point that popular opinion began turning against the massacres.^[32] In truth, one suspects that had the violence of September taken the earlier form, there might have been fewer victims. Violence as spectacle is public, and the public, it seems, while not adverse to a certain frisson , would have balked at too many corpses. Besides, the execution of large numbers of people requires a more efficient use of time than allowed for by the dismemberment and display of body parts. In this regard too, the official Terror follows the pattern set by the September Massacres, where spectacle was abandoned for efficiency.^[33]

Now this, admittedly, might appear contrary to general impressions. One often thinks of the public executions of the Terror as a spectacle. Books like The Scarlet Pimpernel or The Tale of Two Cities have filled our minds with images like those of the tricoteuses , those old hags who knitted beside the guillotine, cackling with glee as the heads rolled. But although the manifestly gruesome aspects of these executions are not to be denied, this is to miss something of their historical significance. The guillotine was invented by the good doctor Guillotin as an efficient and painless method of execution and, compared to the earlier practices of the Ancien Régime, it was.^[34] Moreover, though the guillotine first operated in the central square of Paris, as the state Terror proceeded, it was moved to the outer suburbs, and finally, during the last days of what has come to be known as the "Grand Terror," most of the executions took place, like those of the September Massacres, within the prisons or their courtyards.

There is more to this change in forms, however, than merely the introduction of a more efficient, instrumentally rational mode of public execution. These two forms involve not just a relation to collective violence but a relation of "society" to that violence—a relation that reflects back on that society and can tell us something about it. For such violence is of a given society, and is thus expressive of that society, of its order, and of the modalities by which that order is affirmed. In order, then, to express something of this relation, and of the difference in this relation between the two forms, I am going to resort to two contrasting metaphors: a metaphor of ingestion and a metaphor of expulsion. In the one case, the anterior form, it is as though the social body seeks to devour its victims; whereas in the other, it is as if the victims are to be expelled from the social body.

In the first case, the violence is visible to all and thus engages the entire social body, at least vicariously. Again, the victims are cut up and cut open, the pieces mounted on the ends of metal implements and made to circulate through the major arteries of the city, with the entire populace being made to partake, if not necessarily savor, the repast. Moreover, whenever such violence is described, whether by the Revolution's sympathizers or detractors (though almost always by the literate and respectable classes, who were appalled at the "savagery" of the popular classes), the descriptions are couched in the terms of a vocabulary of digestion: those participating in such acts are inevitably called "cannibals," "anthropophagi," or buveurs du sang (blood-drinkers); and they are described as threatening to "bite off the head of an aristocrat" (or sometimes of a "bourgeois"), or to "eat their liver," "open up their stomach and eat the intestines," and so on; and, to be sure, particularly nasty sans-culottes are rumored to have actually carried out these threats (though such rumors are probably all apocryphal).^[35]

The second form of violence, by contrast, employs a vocabulary of expulsion (but note: not of excretion, the polar opposite of ingestion). Here one finds a language of exclusion, with reference to either the ancient practices of ostracism or the newer language of the contract: one is pushed "outside the city," or "outside the protection of the law," or one is "returned to the state of nature"). Or alternatively, one hears the language of the purge, often supplemented by a whole panoply of medical metaphors that speak of the need to cut off gangrenous members or surgically remove diseased tissue.

Now, this second, more sanitized vocabulary has to be related to the language of Enlightenment, by which I understand something different from but not entirely unrelated to the Enlightenment as philosophical movement. Although there is still much argument about the Revolution's relation to the philosophical Enlightenment, there can be no doubt about the Revolution's constant, and constantly reiterated, use of the language of Enlightenment.^[36] In a million and one speeches, texts, and theatrical representations the same message is repeated: society is to be enlightened; light is to shine on all society, and thereby eliminate all the areas of darkness; the entire topography of caves and crags, mountains and valleys, where shadows are cast, dark deeds committed, and diabolical plots hatched, is to be leveled. Accordingly society will form a single, smooth surface, where everyone will be visible to everyone else and where conformity to the law will result from continuous, mutual surveillance.^[37]

And beyond the language of Enlightenment, there lies the horizon of an institutional project. For this was the period, broadly speaking, of what, since the work of Michel Foucault, has been termed the "Great Enclosure," when what had come to be perceived as "deviances" were removed from society and placed within an increasingly differentiated series of enclosed institutional spaces. A point not without relevance to the September Massacres and

their tendency to confuse political with social deviancy. For not only, as noted earlier, were the victims common-law prisoners, as well as political prisoners, but with the invasion of the Bicêtre and the Saltpetrière (which were hospices as well as prisons), unemployed beggars and vagabonds, abandoned and delinquent young adolescents, as well as the insane, also found themselves before the bar of the revolutionary tribunals. In effect, the massacres tended to encompass almost all the newly emergent institutions of individual sequestration. It is as though a mechanism was being constructed which pushed the logic of social exclusion ever deeper. As though everwidening circles of those who had been excluded from the visible realms of social intercourse were to be expelled from even the invisible realms of social separation; while the violence that purged those who had been excluded was itself to be expelled from the realms of general social visibility. As if the snake could swallow its own tail.^[38]

Law and Society

It is at this point that I wish to conclude, though in another sense I have only just begun. There remains the question of the larger significance of the changes that I have sought to articulate. The violence of which I have spoken was neither a mere tool, whose only significance lies in its effectiveness relative to the purposes it was meant to serve; nor was it some sort of preinstitutional chaos out of which a regularized social life would be shaped, and which, as such, escapes interpretation, because without shape itself. To speak of the forms of violence is to speak of violence as part of society, and as expressive of the society of which it is a part. Indeed, to speak of the forms of violence may be to provide a unique perspective on society, here on both a society in eclipse and a society emergent. In what follows, I will restrict myself to some brief and very general comments, which, it is hoped, point in the direction of further enquiry.

1. One is tempted to claim that popular violence, as regards the mutilation and display of corpses, merely reflected the practices of the crown under the Ancien Régime. For the latter also made a spectacle of the dismemberment of those it had condemned to die—and in manners that could be quite extravagant. (In this regard popular violence was relatively "humane," the victim not being made to die a "thousand deaths" under torture.) Where popular violence most clearly differed from the violence of the crown, was in its relation to the law. Yet here too, one might argue, the crown's violence conformed to the principles of the Ancien Régime. According to the latter, since the law proceeded from above, only those closest to its source, that is, only those in the upper reaches of the hierarchy, could be the law's representatives; the general populace, being mired in a profane reality, was by definition incapable of participating in the formulation, pronouncement, or

execution of the law. As such, popular violence could not but be a rejection of the law. Only with the September Massacres, with what I called the transformation of "popular violence" into "popular justice," did the general populace begin to assimilate something of the modern conception of the law and its "egalitarianism."^[39] Only in a Republic, Montesquieu had said, can the people be imbued with the "spirit of the laws." In this sense, it is perhaps not entirely coincidental that the massacres occurred only weeks after the fall of the monarchy.^[40]

2. One does not want, however, to assimilate completely the violence of the revolutionary crowds to the principles and practices of monarchic justice (which, one should add, underwent major changes during absolutist rule). Popular violence did not simply reject the law, it inverted the law, turning the hierarchical order upside down in a carnival-like atmosphere. And as a carnivalesque inversion, one is tempted to see in such violence the signs of a separate popular culture. (The existence of the latter being suggested, not least of all, by the mixture of incomprehension and disgust with which such violence was greeted by its "bourgeois" critics—critics who, it must be remembered, often condoned the equally sanguinary movements of the "national razor.")^[41] Here a brief reference can be made to the book on popular culture by the Russian literary critic Mikhail Bakhtin, entitled Rabelais and His World . In this work the carnival is perceived as a truly popular festival, for unlike, say, the "ceremonies of the Law," it is not offered by, or with reference to, any external, transcendent source of power or truth. And in a hierarchical society, where all order is perceived to proceed from above, the carnival, by virtue of its "immanence," can only proceed in accordance with a certain, albeit welcome, disorder. For the duration of the festivities, the sanctity of the existing forms of the coercive sociopolitical organization are mocked, its rules and injunctions suspended and, in an atmosphere of terrible gaiety, carnival "rights and freedoms" reign supreme.^[42] Furthermore, in this atmosphere of carnival violence, as Mikhail Bakhtin writes, "the kitchen and the battle meet and cross each other in the image of the rent body."^[43] Images of culinary anatomization and dismemberment are common, and the dismemberment of individual bodies is clearly related to the dismemberment of the social body, the latter being literally turned inside out, with what Bakhtin calls the "material bodily lower stratum" acquiring ascendancy over the more heady realms of the upper strata. Finally, the individual body of carnivalesque imagery is portrayed as "grotesque," composed of exaggerated protruberances reaching out to the world, and equally exaggerated orifices through which all bodily transactions with the world pass. Mikhail Bakhtin contrasts this grotesque body to the classical image of the body which was revived in the high culture of the early modern era. With the latter, the body was presented as formed of smooth, impenetrable surfaces that "contained" it, closing it off as a separate completed phenomenon:

All signs of its unfinished character, of its growth and proliferation, were eliminated; its protruberances and offshoots were removed, its convexities . . . smoothed out, its apertures closed. The ever unfinished nature of the body was hidden . . . . The accent was placed on the completed, self-sufficient individduality of the given body . . . . The inner processes of absorbing and ejecting were not revealed.^[44]

One is tempted to transpose here the image of the individual body to that of the social body held by the revolutionaries who were, in this as in all else, very much classicists.

3. This brings me to a third point, one that concerns what might be called the delineation of the social body or, more precisely, of society.^[45] I am not speaking here of a geographical or territorial delineation, though a few words about the latter may serve to illuminate what I am speaking of. In the Ancien Régime, the geographic extent of the kingdom stretched as far as the monarch's authority; the latter, however, tended to become weaker the further removed it was from the monarch's original domain, till it faded into a sort of no-man's-land at its farthest reaches. In a similar manner, it was held that the force of the law weakened the further removed its point of application from its point of origin.^[46] Which is to say that as the law moved from its source in divine principle, through its terrestial representatives in the first two Estates, and down to the lower reaches of the social hierarchy, it was supposed that its hold became ever more tenuous, men's actions being impelled by appetites increasingly removed from the ideals embodied in the legal order. What this suggests is that, even as the legal order defined the societal order, the order thus defined was necessarily much narrower than "society" or collective existence, particularly in its more profane aspects. With the Revolution all this changed. The law no longer moved vertically, down a hierarchy, but "horizontally," where it was to apply to all, with equal force and in an equal manner. One consequence of this "horizontal" positioning, which should be noted, was that the law, and the societal order that it was to establish, could no longer be inverted, only overthrown. The extension of the law in this sense, with its egalitarianism, its universalism and seriousness, was opposed to carnival forms and "their whole psychology of two worlds."^[47] In this sense, the law was the perfect instrument of revolution.

Another consequence, a more significant consequence for our present purposes, was that the legal order was held to define not just a societal order but "society" itself. In the utopian longings of the revolutionaries, law and society were to be coequivalent: the law was to constitute society, the pure product of a legislative project; while the society thus constituted was to be so designed as to uphold the "sovereignty of the law" and its rigors. To be sure, this revolutionary apotheosis of the law could not last. For the law retained a trace of its transcendence: by its very rationality and visibility, by the very fact that it demanded obedience, it could never be identical to society.^[48] And

as it became all too obvious that the hold of the law was (necessarily) weak, the imperative was to discover the "real" society, the society beneath the laws, constituted of a deeper order barely visible to the social actors, an order located, at least in part, outside the self-conscious workings of their rational faculties.^[49] But even as the legal order was, relative to the definition of social reality, to be replaced by either a "natural order" (variously defined) or a nonrational, normative order, the law still served, without remaining identical to society, to trace the limits of behavior in society. An exoskeleton of the permissible, failure to conform to the law placed one outside society, depriving one of the rights the latter guarantees and the protection it affords. And it is only when the boundaries of society are thus demarcated, that the logic of exclusion discussed above can begin to operate.

4. The last point: by drawing a line separating social from antisocial behavior, the law establishes what, ideally, lies outside society as well as within. And not the least of the things is to be pushed beyond the societal frontiers is all violence that originates outside the law and its application. We are faced here with what Max Weber considered one of the defining characteristics of the modern state: its monopolization of the legal use of violence (along with its corollary, the pacification of society). But what of this legal, or better, legalized, violence that establishes the boundaries and banishes the criminal and his crime outside the "gates of the city"? Is it to be part of the order it enforces, with full rights of visibility? Or do the means of enforcement belie the order to be enforced, and not just because of the resemblance of legalized violence to its illegal counterpart? Within the Ancien Régime things were clear. The violence of power was presumed necessary, the very reflection of its strength. As power moved ever deeper into the lower regions of the hierarchy, the recourse to violence was deemed increasingly necessary to enforce what could only be a minimal, grudging obedience. Moreover, power made itself visible through its violence; it had to appear spectacular and arbitrary (capable of terrible deeds, but also of acts of grace) if it was to mark its distance from common mortals. It was only when the nation was sovereign and power was said to emerge from society at large, that its violence, at least when directed internally, would appear as weakness, the sign of a division internal to its source. It was only when the law was the supposed expression of a general will that the (frequent) recourse to legalized violence would suggest that the law was not general and the people were far from infused with its spirit. At worst, the violence of the law might indicate the existence of an undeclared civil war, itself the possible sign of governmental tyranny.^[50]

Now it would be difficult to assert that the modern state had been more, or for that matter, less violent than that of the Ancien Régime. But where modern states have been violent, they have almost invariably sought to cover the extension of their violence with the simultaneous removal of that violence from the sight of, and more generally, intercourse with, society. Indeed, one

might read the history of state violence since the French Revolution as continuously attempting to improve on this logic.^[51] In this regard, the twentieth century, with its concentration camps and gulags, has been particularly fertile, with perhaps the most recent refinements coming from Argentina, that piece of Europe in Latin America. With the desaparecidos of the 1970s, not only was there an attempt to render the violence invisible (it was officially denied; those slated for execution being picked up by police and soldiers in civilian uniform). Not only was there an attempt to render the identities of the victims invisible (their identities were removed from official records, their bodies buried in unmarked graves or dropped over the ocean), as well as those of their persecutors (they gave themselves false names, sometimes false identities; the law under whose authority they operated was, when not ignored, made into an official state secret). But there were, as well, attempts to make even the sites of violence "disappear." For not only were the sites where victims were held, interrogated, and executed closed to public view, there were, apparently, attempts to render these sites mobile, lest, by virtue of the evidence of geography, they be susceptible to public exposure. This must all count as a significant improvement in the logic of exclusion; beyond the acts of incarceration, and of death by incarceration, it is the traces of such acts that are systematically eliminated.^[52]

We are here, to be sure, far removed from the events of September 1792. But something began in that month, another history, different connections, with unexpected and unrecognized lines of descent. It would, of course, be absurd, during the bicentennial of the French Revolution, to establish a separate anniversary for the September Massacres. What perhaps began with the latter is too deeply buried in our modernity, too far removed from questions of political partisanship, and too dark in its character and implications to admit of the prestige of an inaugural event. In something as epochal, total, and complex as the French Revolution, it is its deepest, most obscure aspects to which we remain the most tightly bound.