1. Hannah Arendt, "On Violence," in Crisis of the Republic (New York: Harcourt Brace Jovanovich, 1970), p. 106.

2. As a morality of the means it would bracket the question of ends, or at least of ultimate ends, whether stemming from universal values, or the more particular values of the analyst. And yet, almost inevitably, the latter return, unannounced and unquestioned. For the question of the actor's "rationality" tends to be subject to the analyst's social prejudices; whereas that of the "rationality" of the actor's motives tends to be prey to the author's moral prejudices.

3. "Michelet's use of le peuple corresponds, of course, far more closely to the facts . . ." George Rudé, The Crowd and the French Revolution (London: Oxford, 1967), p. 232.

4. From Reflections on the Revolution in France , cited in ibid., p. 2.

5. Indeed Professor Rudé is able, by virtue of his archival searches, to pinpoint for each of the journées révolutionnaires , which categories of sans-culottes were present, and from which sectors of Paris they came, thus shedding considerable light on the individual character of each of the events.

6. Rudé, The Crowd and the French Revolution , pp. 192-196.

7. Ibid., pp. 219-221.

8. Ibid., p. 199.

9. Ibid., p. 225.

6. Rudé, The Crowd and the French Revolution , pp. 192-196.

7. Ibid., pp. 219-221.

8. Ibid., p. 199.

9. Ibid., p. 225.

6. Rudé, The Crowd and the French Revolution , pp. 192-196.

7. Ibid., pp. 219-221.

8. Ibid., p. 199.

9. Ibid., p. 225.

6. Rudé, The Crowd and the French Revolution , pp. 192-196.

7. Ibid., pp. 219-221.

8. Ibid., p. 199.

9. Ibid., p. 225.

10. Charles Tilly, From Mobilization to Revolution (Reading, Mass.: Addison-Wesley, 1978), pp. 145-149.

11. In truth, the book is not entirely consistent in this regard. The "rationality" of a "defensive reaction" cannot but be "bounded" by its failure to adapt to history and its ''innovations." In a later book, Ideology and Popular Protest (New York: Pantheon, 1980), even the rationality of "proactive struggles" (those of unions, for example) will be considered limited, and limiting. Rationality here—and it is truly a revolutionary ''rationality"—would consist not in adapting to history but in mastering it. However, notes George Rudé, echoing Lenin if not Lukács, those subject to history (i.e., the popular classes) are of themselves in no position to become subjects of history.

12. Richard Cobb's work in particular is invaluable for charting the social and geographical topography of these rumors and fears.

13. The term septembriseur was not the least of the epithets used to discredit one's opponents in the factional infighting that was to consume the revolutionary camp. An excellent essay on the "difficulties" in using the archival evidence to study the journées révolutionnaires is the first piece in Richard Cobb's The Police and the People. French Popular Protest 1789-1820 (London: Oxford University Press, 1970), pp. 3-48.

14. Rudé, The Crowd in the French Revolution , p. 111.

15. Ibid., p. 110. On these events one should consult the classic work of Pierre Caron, Les Massacres de septembre (Paris: Maison du Livre Français, 1935); and the more recent piece by Frédéric Bluche, Septembre 1792. Logiques d'un massacre (Paris: Robert Laffont, 1986).

14. Rudé, The Crowd in the French Revolution , p. 111.

15. Ibid., p. 110. On these events one should consult the classic work of Pierre Caron, Les Massacres de septembre (Paris: Maison du Livre Français, 1935); and the more recent piece by Frédéric Bluche, Septembre 1792. Logiques d'un massacre (Paris: Robert Laffont, 1986).

16. There is some evidence, however, that the rumor only developed after the massacres had begun. See Bluche, September 1792 , pp. 29-30.

17. " Surprisingly , only one quarter of the prisoners were priests, nobles, or 'politicals' . . ." Rudé, The Crowd in the French Revolution , p. 150 (my emphasis).

18. Ibid., p. 112.

17. " Surprisingly , only one quarter of the prisoners were priests, nobles, or 'politicals' . . ." Rudé, The Crowd in the French Revolution , p. 150 (my emphasis).

18. Ibid., p. 112.

19. Though in truth, he does speak much later in the book of the massacres as a consequence of "mass hysteria." Ibid., p. 225.

20. Others have sought to place the "rationality" of the septembriseurs ' actions within the internal struggles of the revolutionary camp, their real purpose being, it is claimed, to terrorize the more moderate elements. Now although there is some evidence that certain elements were willing to use the massacres to advance their interests in the factional struggles, such maneuvers appear as a sideshow to the main event. Nor did any moderate die in the violence. Nor, generally speaking, were the moderates (i.e., the Girondins), as judged by their press, any more or less critical of the events, as they were taking place, than their more radical counterparts. It was only later, after public opinion had turned against the massacres, that the Girondins

sought to distance themselves from the events, while portraying their more radical enemies as morally compromised. See Bluche, Septembre 1792 , pp. 73-76, 201-205.

21. Ibid., pp. 22-23.

20. Others have sought to place the "rationality" of the septembriseurs ' actions within the internal struggles of the revolutionary camp, their real purpose being, it is claimed, to terrorize the more moderate elements. Now although there is some evidence that certain elements were willing to use the massacres to advance their interests in the factional struggles, such maneuvers appear as a sideshow to the main event. Nor did any moderate die in the violence. Nor, generally speaking, were the moderates (i.e., the Girondins), as judged by their press, any more or less critical of the events, as they were taking place, than their more radical counterparts. It was only later, after public opinion had turned against the massacres, that the Girondins

sought to distance themselves from the events, while portraying their more radical enemies as morally compromised. See Bluche, Septembre 1792 , pp. 73-76, 201-205.

21. Ibid., pp. 22-23.

22. One might wish to date the beginnings of the construction of this apparatus with the formation of the Committees of Public Safety and General Security in February of 1793, though one could also date these same beginnings in August of 1792—that is, just prior to the September Massacres—when the first hesitant steps were taken in the establishment of the guillotine permanente , the revolutionary tribunals, and the committees of surveillance.

23. What follows is indebted to Bernard Conein, "Le tribunal et la terreur du 14 juillet 1789 aux 'massacres de septembre.'" Les révoltes logiques , no. 11 (Winter 1979-1980): 2-24.

24. Though the attitude of the municipal authorities, in particular was rather ambiguous and has become the object of some controversy. See Bluche, Septembre 1792 , pp. 151-183.

25. Conein, "Le tribunal et la terreur," pp. 18-19.

26. Ibid., pp. 10-11.

25. Conein, "Le tribunal et la terreur," pp. 18-19.

26. Ibid., pp. 10-11.

27. Already in the month prior to the September Massacres, the Commune of Paris was petitioning the National Assembly for the authority to establish a special tribunal that would be responsible for judging and punishing political crimes. This tribunal was to be composed of citizen-judges elected by the sectional assemblies, the idea being to place all public executions under the authority of the Commune, while simultaneously associating the sans-culottes, or at least their politicized elements, with the new reorganization of the judicial apparatus. And in fact, on August 17, the National Assembly did agree to the formation of a new tribunal, though one composed of judges elected not by the sectional assemblies, but by special electoral bodies and on the basis of professional criteria. Ibid., pp. 16-17.

28. Indeed those radicals who sought to justify, or at least excuse, the events of September argued that because the people had organized themselves into tribunals (though the evidence suggests that the "people" did not organize the tribunals, but simply recognized tribunals organized "from without") these events were not to be represented as "massacres,'' but as an act of justice, of people's justice, the just punishment of the Revolution's opponents.

29. Conein, "Le tribunal et la terreur," p. 23.

30. In truth, there was one last outbreak of popular violence—after the fall of the Jacobins. On the first of Prairial (May 20, 1795), during the last desperate "revolt" of the sans-culottes, the deputy Fréraud was killed and his head paraded on a pike.

31. These claims apply only to Paris. In the provinces, the "echoes" (and they were rather faint) of the September Massacres still involved the mutilation and display of corpses. See Bluche, September 1792, pp. 103-121.

32. Ibid., pp. 75-76.

31. These claims apply only to Paris. In the provinces, the "echoes" (and they were rather faint) of the September Massacres still involved the mutilation and display of corpses. See Bluche, September 1792, pp. 103-121.

32. Ibid., pp. 75-76.

33. Here a certain caution has to be added to Natalie Zemon Davis's claim that the "rites of violence," particularly in their more carnivalesque forms, serve to produce the "conditions for guilt-free massacre." Although they may help the killers to "forget that their victims are human beings,'' in the French Revolution at least, the real massacres occurred outside the framework of such rites. "The Rites of Violence,"

Society and Culture in Early Modern France (Stanford: Stanford University Press, 1965), p. 181.

34. Although the observation of movements of the face and body after decapitation led to a lively debate as to whether the guillotine actually delivered its promise of instant death. See "Notice historique et philosophique sur la vie, les travaux et les doctrines de Cabanis," which introduces P.-J.-G. Cabanis, Rapports du physique et du moral de l'homme (Geneva: Slatkine Reprints, 1980), pp. XXI-XXII.

35. Cobb, The Police and the People , pp. 87-88. Cannibalistic imagery in popular revolts has a longer history. See for example Emmanuel Le Roy Ladurie, Les paysans de Languedoc (Paris: S.E.V.P.E.N., 1966), pp. 398-399; and Carnival in Romans (New York: George Braziller, 1979), PP. 179-180.

36. See Brian Singer, Society, Theory and the French Revolution (New York: St. Martin's, 1986), pp. 46-47; Jean Starobinski, 1789: Les emblêmes de la raison (Paris: Flammarion, 1979), pp. 31-37; and Mona Ozouf, La fête révolutionnaire (Paris: Gallimard, 1976), p. 119.

37. Singer, Society, Theory and the French Revolution , pp. 47 and 193; Michel Foucault, "The Eye of Power," reprinted in Power/Knowledge: Selected Interviews and Other Writing 1972-77 (New York: Pantheon, 1980), p. 153.

38. Since writing this essay, I have been asked to say something about violence from the "right." In this regard, an exceptional article by Colin Lucas on the "White Terror" ("Themes in Southern Violence After 9 Thermidor,'' in Gwynne Lewis and Colin Lucas, eds., Beyond the Terror: Essays in French Regional and Social History [Cambridge: Cambridge University Press, 1983]) can provide a few comparisons. The popular culture of the southeast was not that of Paris, and the violence in the former was rooted in (the decomposition of?) youth organizations and charivaris which had long disappeared in Paris. Moreover the counterrevolutionary communal violence had an intensely local character (it being, apparently, a tacit rule that the victims and their tormentors had to know each other) that did not obtain in Parisian conditions. However, the violence was public: if the crowds did not participate in the killings, as in Paris, they were always there as witnesses; and bodies were mutilated, though curiously not by the killers but by the crowds after the event (though mutilation did not have the obligatory character it did in Paris—indeed, the victims were not always killed, but subjected to traditional forms of humiliation or restitution). What about the relation to the law? Obviously, such violence took place outside the framework of the law and its forms. Colin Lucas does not tell us if the representatives of the law remonstrated before the crowds, only to be explicitly rejected. But in another sense, he does suggest that it was the (former) representatives of the law, those who had upheld it the most rigorously, who were the "privileged" targets of the violence. For those murdered during the White Terror were not simply Jacobins, but those Jacobins who had applied the revolutionary laws the most inflexibly, who had thus, in the name of a larger collective entity and more general norms, ignored the community and its more traditional, tractable ways of settling conflicts and easing tensions. These were truly the ''reactive struggles" mentioned earlier, pitting the community against the external, encroaching, and innovative powers of the nation, with its legal universalism, and the state, with its military, fiscal, and legal centralization.

39. One usually understands this egalitarianism as "equality before the law"; but

it also encompasses an "equality behind the law," whereby everyone participates, in principle equally, in the formulation and carrying out of the law. It is the very meaning of popular sovereignty.

40. Admittedly, during the religious strife of the sixteenth century, the crowds often claimed for themselves the role of the magistrate (and cleric). But beyond the fact that the religious and political authorities often participated in the crowd's actions, and were almost always willing to justify them, the crowds were, by acting in the name of the sacred, presumably approaching the transcendence of the law. Perhaps in a hierarchical society, the explicit rejection of the law and its forms only occurs when it is the hierarchy itself that is being attacked. See Davis, "The Rites of Violence," pp. 164-169.

41. Incomprehension, if not always disgust, almost always suggests the copresence of two incommensurable cultures. Too often historians have searched for the existence of a separate popular culture in what turns out to be the fragments of an earlier elite culture, reworked for consumption by the lower strata. Such seems to have been the fate of the research into the bibliothèque bleue —a fate that was not entirely unpredictable, given that, by definition, a bibliothèque of whatever color will be tied, however tenuously, to the world and culture of the literate. See Robert Mandrou, De la Culture populaire aux XVII et XVIIIe siècles (Paris: Stock, 1964); and Geneviève Bollème, La Bibliothèque bleue (Paris: Julliard, 1971).

42. Mikhail Bakhtin, Rabelais and His World (Cambridge, Mass.: MIT Press, 1968), pp. 9-11.

43. Ibid., p. 197. See also pp. 192-194 and 207.

44. Ibid., p. 29.

42. Mikhail Bakhtin, Rabelais and His World (Cambridge, Mass.: MIT Press, 1968), pp. 9-11.

43. Ibid., p. 197. See also pp. 192-194 and 207.

44. Ibid., p. 29.

42. Mikhail Bakhtin, Rabelais and His World (Cambridge, Mass.: MIT Press, 1968), pp. 9-11.

43. Ibid., p. 197. See also pp. 192-194 and 207.

44. Ibid., p. 29.

45. In a sense, the Revolution will mark the displacement of the social body by society. For once the former is no longer tied to the image of the individual body of the monarch, collective existence can no longer be represented quite so easily in corporeal (if not necessarily, organic) terms.

46. For purposes of clarity I am exaggerating here: the absolutist regime sought a certain, partial consolidation of its territory, and of the law enacted therein.

47. The expression is taken from Natalie Davis, "The Reasons of Misrule," in Society and Culture in Early Modern France , p. 120.

48. At least not without being internalized, and thus losing its character as "law." Singer, Society, Theory and the French Revolution , pp. 110-123.

49. In truth, one should add that this search is complicated by the existence of another, parallel search. For the society, born of the continuous reversibility of is and ought, is held to exist not just as an object to be investigated in its deep structures, but as a subject to be constantly interrogated as regards the surface movements of will formation, public opinion, and fashion.

50. Singer, Society, Theory and the French Revolution , pp. 86-87.

51. And what about antistate violence? Terrorism is spectacle or it is nothing, without, however, being popular. Generalizations are dangerous, but an extraordinary example of truly popular violence has recently been provided by the overthrow of the Duvalier regime in Haiti. Here the populace took to the streets in a festival atmosphere to search out ex-members of the paralegal police force of the ancien régime, the tonton macoutes. Macoutes who were found were executed, and, as in the French

Revolution, their dismembered bodies were borne in triumph by the crowds. Moreover, the events were filmed, presumably by participants, and video cassettes were distributed underground throughout Haiti and the Haitian diaspora. However, unlike the French Revolution, the crowd violence was not limited to a few "symbolic" victims. But then a "revolution" had not really occurred; and once, under the pressure of "liberal" allies, the killings stopped, momentum was lost and forces of the old regime were able to reassert themselves. I owe this information to a personal communication by Nina Schiller.

52. Paul Virilio/Sylvere Lotringer, Pure War [New York: Semiotext(e), 1983], pp. 88-89.