History of Structuralism
Volume
2: The Sign Sets, 1967-Present
Francois
Dosse
Part III
Structuralism bet-ween Scientism, Ethics, and History
This page intentionally left blank
The protest against the structural paradigm eventually tarnished the term "structuralist." Everyone ardently claimed to have never partaken of the festivities. Researchers presented their work as all the more singular despite the fact that only yesterday they had tried to situate it within the collective current of the structuralist renewal. Some sought even greater formalization in order to access the very essence of structure, whereas others undertook to deconstruct this for-malization and give free rein to an increasingly literary inspiration that slowly but surely edged away from the initial, ambitious efforts at codification.
The first response-a formalist one-eame in the field of linguistics with the founding of the Paris School, which inevitably recalled the Prague School, and which, moreover, fit into this historical line: "This was the Paris School and not the French School of Semiotics, because Paris is a place where many foreign researchers come and realize that they share a certain number of things."! Born at the International Semiotics Association, the brainchild of Roman Jakobson and Émile Benveniste, drawing its inspiration from the Russian formalists and the work of the Prague, Copenhagen, and Geneva schools, the association was essentially the offspring of European linguists, despite the involvement of the patron of American semiotics, Thomas A. Sebeock.
The association sought, among other things, to give Eastern European researchers the opportunity to leave the Marxist vulgate on the other side of the Iron Curtain and renew intellectual energies of the thirties in Central and Eastern Europe. The selection of Warsaw as the site of the association's second symposium was symbolic in this respect, and the Poles played a decisive role in it. At the same time, this reunion had the ring of an impossible challenge since it took place in the summer of 1968 against the backdrop of Soviet tanks invading Czechoslovakia, hardly a propitious context in which to undertake the establishment of productive ties between East and West. Thomas Sebeock, of Hungarian descent, considered the situation so dangerous that he canceled his trip.
The Paris Semiotic Circle was established a year later. "We talked with Levi-Strauss to decide who might make up the nucleus of the French Semiotics Association. Finally, it was composed of Benveniste, Barthes, Levi-Strauss, and me. Lacan wasn't serious enough for Levi-Strauss, and Foucault seemed frivolous."2 Unfortunately, Benveniste, who was named president of the circle, did not have time to affect the orientation of its work, because shortly thereafter he suffered a stroke that left him a hemiplegic. His intellectual disappearance and Barthes's growing lack of interest in semiotics-he was leaning more and more clearly toward literature-meant that the activities of the circle depended increasingly on Greimas, who was in the Social Anthropology Laboratory at the College de France, run by Levi-Strauss, "If Benveniste had actually lived longer intellectually, the balance would have been different."! Therefore, Hjelmslevian linguistics, the most formalist of linguistic approaches, carried the day in Paris. During the same year, the association launched the new review Semiotica, which was overseen by Julia Kristeva and Josette Rey-Debove. "Ben-veniste and Jakobson needed someone young and dynamic, and they asked me to take over the secretary-general's duties."4
In the first issue, Benveniste recalled the historical origins of the concept of semiotics, borrowed from Locke, and especially from the American philosopher Charles Sanders Peirce (1839-1914), who had wanted to construct a "universal algebra of relationships."5 But Benveniste did not adopt Peirce's ideas; on the contrary, he found the view that language was everywhere and nowhere to be too loose because, in his view, it ran the serious risk of condemning any meaningful research to the abysses of the infinite. He proposed the Saussurean legacy: "Somewhere, the universe must acknowledge the difference between the sign and the signified. Every sign must be taken and understood in a system of signs. That is the condition of signification."6 Thus the Paris School adopted Peirce's notion of semiotics while remaining faithful to Saussure's methodological legacy. Distinguishing between semantic interpretation and a semiotic level was a way of broadening the analysis of the life of signs to the whole of social life. It was a matter of systematizing the Saussurean trajectory, which, starting with language, intended to study the other sign systems: "Language contains society. The interpreting relation, or semiotics, is the opposite of articulation, which is sociological."? Language would therefore interpret society, according to two principles that made it possible to place different semiotic systems into relationship with each other: nonredundancy between systems, and the fact that "there is no transsystemic sign." 8
This semiotic orientation did not as yet include Kristeva's distinction between a symbolic level of language in the linguistic sense of a homogeneous and articulated structure, and a semiotic level, which she understood as an unconscious process, something like a drive, observable in the interstices of language as so many marks of undecid-ability and heterogeneity. That would come later.
The Paris Semiotic Circle initially presented itself as the meeting place between structural anthropology and Saussurean semiology. Because Levi-Strauss favored the group, he invited his semiotic partners into his Social Anthropology Laboratory at the College de France. But he in no way tolerated Greimas's intention of creating a better symbiosis between Saussure's linguistic legacy and the semiotic study of myths: "This linguistic domination was acceptable for many, including anthropologists, insofar as it remained discrete and offered conceptual tools, but it became intolerable when it became a semiotic enterprise with pretensions of covering many areas."? So Levi-Strauss quickly gave his colleagues their leave, and "Greimas was forced to leave the office that he had at the College de France."lO
As a result, Greimas's influence grew and the school became hermetic in its increasingly rigorous, self-enclosed formalization, drawing its model more than ever from the hard sciences. Ever since he had published Structural Semantics, Greimas was convinced that he could reach total meaning and the complete meaning of the structure. In this configuration, the sign became "the transcendental site of the condition of the possibility of meaning, of signification and of reference." 11 Greimas argued that this site could be reestablished with the semiotic square, a veritable open sesame for any sign system. This dream of formalization took structuralism as its emblem, a crystal whose low temperature prevented the dispersion of molecules and nourished hope that, by reducing humanity to a degree zero, the transcendental keys to the conditions of its possibility could be found. "The structuralist dream would be death by refrigeration. "12
The school produced a number of semiotic studies of literary objects, including Algirdas Julien Greimas's on Maupassant.P Jacques Geninasca's on Gerard de Nerval!" Michel Arrivé's work on Alfred Jarry,15 and Jean-Claude Coquet's work, which had a more general thrust.te Literature for the semiotician, however, was nothing other than a signifying practice like any other, without any particular valorization: "Literature as an autonomous discourse with its own laws and intrinsic specificity was almost unanimously rejected."17 "For semiotics, literature does not exist!" 18
Philippe Hamon considered the character of the novel from this angle, pulverizing it from a semiological viewpoint. He worked out a grid of critical analysis of what he considered to be the manifest trace of humanist ideology. When he dissolved the notion of hero, he did so by applying many concepts that made it possible to establish a general theory specifying a semiology of the character and "distinguishing this semiology from the historical, psychological, psychoanalytical, or sociological approach."19 He defined the character as a sort of morpheme that was doubly articulated by a discontinuous signifier (I, me, to me ... he, julien Sorel, the young man/our hero/... ) and a signified, which was also discontinuous (allomorphs, amalgam, discontinuity, redundancy, etc.). The character's meaning was clear only with respect to the other characters of the utterance, and not by a simple accumulation of characteristics. The study would therefore have to define the pertinent semantic axes and attempt to hierarchize them. "We would thus see classes of distinctive characters, defined by the same number of semantic axes."20 This enormous construction supposed an immanent approach to the literary text, conceived as a construction, and not as a given. Literary tales were then studied in their literariness, cut off from exogenous determinations and confined within their internal logic. A number of semantic categories, such as isotopy, for example, girded up the analysis. "By isotopy we mean a redundant set of semantic categories that make a uniform reading of the story possible. "21
What was clearly an evolution in semiotic analyses of literature between the sixties and the seventies paralleled the changes in linguistics during the same period as it moved from a "linguistics of states to a linguistics of operations," according to Philippe Hamon.s- Such a shift made it possible to go from a closed conception seeking to point out the specificities of complete systems to a much more open approach to discerning the characteristic constraints of this or that communication situation. As we have seen, this evolution led to taking the utterance into account in different interlocutive situations. But the period was also characterized by a broadening of the semiotic field of analysis, which went beyond the literary terrain to apply to all kinds of texts, including legal, biblical, political, musical, and advertising.P
Semiotics was particularly influential in biblical exegesis. The wealth of work in this area doubtless made it possible to resist the general ebb of structuralism in the late seventies. Musical language was one area of predilection for applying the structural approach. "Music alone could have justified the hypothesis of structuralist work. "24 Roland Barthes in particular wrote an article on Robert Schumann's Kreisleriana in which he distinguished between a first formal semiology, and a second affective semiology which he believed revealed the way sounds were set into relationship with each other in terms of dissonance and consonance.v
Serge Martin, the author of a work on musical serniotics.w took a much more Hjelmslevian approach. Unlike Barthes, he wanted to discover how meaning is produced within the system itself, by comparing the major and minor modes, rather than in any external form of scales and their intervals.
For me, the system represents what we might call being in the world, in Heideggerian language. It's a schematization with very deep affective roots .... What Heidegger says about Kant's arrangement corresponds completely to the musical system, meaning that the system is a structure in the logical sense of the term. But at bottom this structure points to a deep affective relationship with the world, and that's the reason that music is its expressionP
Thus it was this absent structure, at once essential and unrevealed, that semiotics hoped to restore in its signification. It was even possible to precisely discern the supreme importance of structure in the creation of the Viennese School, where no tonal polarization existed any longer. "Contrary to what tonal music meant, musical language is taken here to come first, with its formal rules of transformation."28 This sketch of the theory of musical language carried the three essential axioms of Hjelmslev's semiotics onto a musical plane.
In 1970, the term "semiotics" replaced "semiology" and "structuralism." It was also at the beginning of the seventies that Lacan dissociated himself from structural linguistics and looked to formalize his thinking to a larger extent by using topology and mathemes.
It will appear, I think, here that the claim that structuralism can give us a way of understanding the world is one more imputation to the clown that is givenus as representing literary histoiy, and that is the issue. But the boredon which it inspired in me, albeit in the most agreeable way because I was in the best of company, is perhaps not what gives me reason to be satisfied.s?
Like the other participants at the structural banquet whose company he esteemed, Lacan, who did not want to get caught red-handed, shunned a dubious label. Instead, he turned to mathematics to lead him to higher levels than Saussurean linguistics could reach. At that point, he managed to draw together Levi-Strauss's mytheme, the Greek term mathema (meaning knowledge), and the root of the notion of matherne, which implied mathematics. Lacan hoped to definitely quit what he henceforth called linguistery, which he considered to be too descriptive, and by means of total formalization reach the pure Signifier, that initial gaping space out of which are formed the knots that, since 1972, he called Borromean. Having temporarily stitched the fate of psychoanalysis to that of the social sciences, Lacan sought out the hard sciences. "The only thing that remained, the sole nourishment for the hermit in the desert, was mathematics.V?
Lacan gave more and more seminars on topological figures, including graphs and tores, and on stage he used string and ribbons of paper, which he snipped into smaller and smaller pieces to demonstrate that there was neither inside nor outside in these Borromean knots. The world was fantasy, and sat beyond intraworldly reality; its unity was accessible only through what is missing in languages. "Mathematization alone achieves a reality, a reality that has nothing to do with what traditional knowledge has sustained, that is not what it thinks it is, not reality but fantasy."3! Lacan was attempting to conceive of the totality and the interiority of what was lacking in reality, working from within to eliminate the categories of inside and outside, interior and exterior, and of any spherical topology. He tried to use a twist as the basis of his model of the knot that eludes all attempts at centering. Deeply plunged into a universe of pure logic unfolding from the priority granted the symbolic void, "Lacan claimed to escape sub-stantification through recourse to topology."32 With the quest for a matheme, the system of rules and the combinatory belonging to a pure system of logic made it possible, more so than had linguistics, to firmly hold the referent, affect, and lived experience at bay.
Some considered Lacan's use of topological figures to be purely pedagogical, a way of teaching psychoanalysis. "The matheme concerned the idea of transmission; it was not a question of making psychoanalysis into physics. "33 But even beyond the possible didactic interest of this topological phase, which frustrated more than one listener, we might imagine that having run into a dead end with linguistics, Lacan refused to totally disseminate his reading of the unconscious, as Derrida had done, because it would have taken psychoanalysis toward an infinite interpretation in which it would have lost itself. He preferred to suggest another direction, with that of the matheme and the Borromean knots, ostensibly a metaphor for the need for a basic and as yet undiscovered structure. "Interpretation is not open to every meaning."34 Closer to structure in the mathematical sense, Lacan took one more step toward abstraction and the idea of a distinct object tied to the operation of specific ideation through which one could deduce the general properties of a group of operations and define the area where demonstrative utterances engendered their properties.
Was this recourse to mathematics and to modelization just a metaphor or was it a heuristic and operational move? Andre Regnier asked about the transition from group theory to The Savage Mind. 35 He analyzed Levi-Strauss's use of the concepts of symmetry, inversion, equivalence, homology, and isomorphism, in his Mythologiques, concepts that he had borrowed from the logical-mathematical realm of knowledge, and whether the use of such metaphors was not in some way dangerous. This, however, was not the case when these ideas, like that of a transformational group, played a central role in Levi-Strauss's scheme of things. "They (totemic institutions) are thus based on the postulate of a homology between two systems of difference, one of which occurs in nature and the other in culture."36
Levi-Strauss had a very broad notion of transformational class and used the term quite freely, focusing on one or another relationship in the syntagmatic chain depending on the needs of his demonstration. He also claimed "my right to choose myths from various sources, to explain a Chaco myth by means of a variant from Guyana, or a Ge myth by a similar one from Colombia."3? Regnier questioned the scientific nature of the demonstration, which would mean adopting nonarbitrary codes and justifying the correspondences: "To understand why, if a being is a sign, it has this rather than that meaning____
Finally, the 'logics' in question have a rather evanescent existence: they are rules imposed on the links but we do not know what they are. "38 He included Levi-Strauss in this illusory scientificity expressed by a belief in formalization in the human sciences.
Gilles Gaston-Granger, however, did acknowledge the relatively successful use of formalization, for example, as when Levi-Strauss analyzed kinship relationships. His model worked, was pertinent, and let us understand how marriages were structured by prescriptions and proscriptions. "But what I would criticize in Levi-Strauss is his attempt to show us that transformations in the mythic mind create a relationship in the same way that algebraists understand a relationship. I don't believe that to be the case."39 Levi-Strauss nevertheless continued to firmly defend modelization. From the mathematics of kinship to the logical-mathematical treatment of the units that constitute myths, he reiterated his confidence (in The Naked Man, the final volume of Mythologiques) in "structuralism [which] proposes an episte-mological model for the human sciences that is incomparably more powerful that what they have had until now. "40
This use of modeling for examining kinship relationships got a second wind with Francoise Heritier's work. A student of Levi-Strauss, she considered herself "lucky to find Claude Levi-Strauss, the director of the Social Anthropology Laboratory. "41 She was able to use a whole range of material gathered on kinship in Upper Volta (Burkina Faso), and she reconstituted the genealogies of the inhabitants of three villages in Samoaland. Modelization and informatics led her to theoretical generalizations based on ethnographic material. "The computer became an indispensable means for getting to the realities of the way marriage worked in societies."42 Informatics aided her in reconstituting how society with semicomplete kinship and marriage structures worked (called the Crow-Omaha system): "Confirming Levi-Strauss's intuition, it seems that a semicomplex system of the Omaha type functions endogamously like the Aranda supersystem, which belongs to a system of elementary marriage structures. The choice of partners takes place in the fourth generation following the common ancestor to two lines of descendants of blood relations."43 With this thesis and the progress that it made possible in going from the study of elementary to semicomplex kinship structures, Francoise Heritier demonstrated the powerful potential of the structuralist paradigm when applied in a limited field in the social sciences, and proved that beyond the variations of intellectual modes, structuralism did allow for true conceptual progress, even if it was often accompanied by the mirage of the purest formalization, that of mathematical language.
From Explosive Literary Mourning to the Pleasure of the Text
Structuralism drew its inspiration from the most formalized of the hard sciences. But at the same time, it was part of a new literary sensibility trying to redefine traditional novelistic storytelling. With the crisis of the novel as an intangible mode of expression, literary theory and literature drew closer and gave rise to the New Novel. A new literary avant-garde quickly grew up in response and became the criterion of modernity. Boundaries between critical and creative activity were muted so that the true subject-writing and textuality-eould unfold indefinitely. As Philippe Hamon wrote, "To consider the concept of literature between 1960 and 1975 is to write the history of a dissolution."! The structuralist theoretical apparatus, and particularly the linguistic approach, would fully participate in the new literary adventure, which took the form of reappropriation of language in its very essence, beyond any boundaries between genres.
New Criticism and the New Novel: Symbiosis Some structural themes were familiar in the founding principles of the New Novel: the elimination of the subject, with the exclusion of the classical novelistic characters; a preference for a space where observed objects were cast in different configurations; a defiance of dialectical time in favor of a suspended temporality, a slack presence that dissolved as it revealed itself.
In 1950, Nathalie Sarraute published "The Age of Suspicion" in
Les Temps modernes, a title that expressed better than anything else the common cast of mind of new literary critics with respect to writers, and which she later took for a major work published by Gallimard in I956.2 More generally, it corresponded to the advancement of the critical paradigm among all the social sciences. Sarraute acknowledged the novel's crisis, and the collapse of the credibility of characters. She compared it with artisanal work based on lived experience, à la Michel Tournier, whose ethnological perspective envisaged creation as bricolage, rather than the more classical understanding of mimesis wherein the crush of details draping an inspired author's characters in a certain density made them believable.
Nathalie Saraute's work quickly came to symbolize the necessary break from the classical novel. Suspicion became the basis for a new relationship with the different forms of writing in this critical age. And yet, Sarraute's break with the psychologizing perspective of the novel was less radical than it appeared. She simply shifted its focus, deconstructing character archetypes and personalities in order to better seize their intimate, underlying beehives of activity. Subconversations and tropisms were conceived of as indefinable movements below the apparent conversational thread, reduced to a pretext in order to reach-via a relationship of psychological immediacy-the ego's infinitely tenuous nature. Although The Age ofSuspicion announced the shape of the New Novel, it remained heir to Dostoyevsky, Proust, and Joyce, the great innovators of novelistic writing.
The New Novel did nonetheless turn to the social sciences, drawing its inspiration from their decentering of the subject, their protest against Eurocentrism, and a configuration that substituted the figure of the Other for the quest for the Same. Conversely, structuralist researchers working in their specific disciplines were to use their discoveries and research areas to make literature. A whole new sensibility was coming to the fore, and it led one to think that truth was beyond the self and that in order to reach it, all the essential relays of knowledge needed to be destroyed. As a result, psychology and temporality became obstacles to truth and structuralism became the new aesthetic: Mondrian in painting, Pierre Boulez in music, Michel Butor in literature. Structure became a creative method, the fermentation of modernity. Initially outside of creation, structure slowly penetrated into the arcana considered until then to be unfathomable. The tenants of new structural criticism in fact invoked this new aesthetic and discovered their predecessors in Mallarme and Valery, because of their shared concern for the verbal conditions of literary creation. "Literature is and can be nothing other than a sort of extension and application of certain properties of language.">
Literary activity abounded. The Editions de Minuit published Michel Butor, Alain Robbe-Grillet, Marguerite Duras, Claude Simon, and Robert Pinget; the Tel Quel group invited Jean Ricardou, the theoretician of the new novel, to join Philippe Soilers, Daniel Roche, and Jean-Pierre Faye. In 1955, when Levi-Strauss's Tristes Tropiques was being heralded, the New Novel was being crowned with literary awards. Robbe-Grillet received the critics' prize for his The Voyeur, and two years later Michel Butor won the Renaudot Prize for The Modification, which sold more than a hundred thousand copies. In 1958, Claude Oilier won the Medicis Prize for his Mise en scène, and the same year, Esprit devoted a special issue to the New Novel. Each of these authors obviously had his own style, but they all embodied the desire for a new kind of novelistic writing that rejected the traditional forms of the novel. The wager implied that all writers had to go beyond those monumental literary forebears who seemed to have definitively defined literary limits: Proust, Joyce, and Kafka. Another direction for another generation, anchored in modernity, needed to be found.
The New Novel expressed the profound malaise of having to write after Proust's Remembrance ofThings Past. At the same time, it expressed a quest for a solution, which it found by placing literary creation in a mise en abime, in appealing for readerly participation, given the explicit projection of the writer's subjectivity. In Nathalie Sarraute's 1950 article, this new critical perspective was still informal, but when The Age of Suspicion came out in paperback in 1964, Sarraute claimed that these articles were a collective manifesto of the avant-garde. "These articles establish some fundamentals for what we call the New Novel today.":'
In 1957, the photographer Mario Dondero captured the image of a debonair group in lively discussion outside the Editions de Minuit. For readers, these figures represented the new novel: Alain Robbe-Grillet, Claude Simon, Claude Mauriac, the publisher jerome Lindon, Robert Pinget, Samuel Beckett, Nathalie Sarraute, and Claude Oilier. The classical character had disappeared from literary concerns and the author's attention shifted within the discursive sphere alone. His observations were the product of an immanent relationship to language, and reality was no longer considered outside language. From Balzac's descriptive mode to Albert Camus's distanced alienation, a further shift led to a reality reduced to the writer's discourse on reality. A symbiosis occurred during the sixties and seventies, during which "the essential thing is not outside of language, it is language itself."> In this, the structuralist orientation, which took phonology as an analytical model and linguistics as its guiding science, was a clear influence.
Rather early on, Alain Robbe-Grillet became aware of this encounter between literature and structuralism, and of the transition from a phenomenological to a structural approach. He adopted Jorge Luis Borges's definition of the difficult exercise of literature: "I am increasingly persuaded that philosophy and literature share the same goals."6 In 1963, Robbe-Grillet published a collection of articles that he had been writing since 1955, entitled For a New Novel,' In it, he laid out the principles he abided by as an author in his own novels-Erasers in 1953, The Voyeur in 195 5-and as a literary consultant at the Editions de Minuit, where he had been working since 1955. He proclaimed the reconciliation of criticism and literary creation that, in order to join modernity, had to be nourished on new areas of knowledge. "Critical concerns in no way hamper creation; they can, in fact, propel it."8 The New Novel was presented both as a school of the look and as a school of the objective novel. It promoted a new sort of realism no longer dependent on nineteenth-century models such as the work of Balzac. It was also a question of a passion for describing, but without describing the intentionality according to which the world only exists as a result of the meditation of characters. In this new writing, "gestures and objects are there before being something."? Just as Lacan had emphasized the importance of what is suggestive in words and the signifying chains, Robbe-Grillet attacked the myth of depth, preferring the surface of things. Description, the same structuralist rejection of the hermeneutic approach, and the same distinction between meaning and signification were all important.
The novelistic revolution shunned characters as outmoded vestiges of a bourgeois universe. The nineteenth century had naturalized the bourgeois order, but this reign of the once-celebrated individual was now outdated. A new era was upon us, an era of "license plate numbers."!" In this desertification, there was something of the desperation of the period; how could we write and think after Auschwitz? There was a desire to disengage from the world of being, and the criticism of modern technology. Hope took anchor in the universe of forms, from which humanity was decentered, a simple and ephemeral incarnation of an indefinite game of linguistic folds. The writer no longer was a bearer of values since "there are only values of the past."l1 The writer was to participate in a static and amnesiac present like the universe of the characters from Last Year at Marienbad, which unfolded without a past and in which each movement and work contained its own erasure. Structural themes resonated within this problematic exercise of literature: the negation of any search for genesis or origins, a purely synchronic approach whose inner logic was to be discovered. "In the
modern tale, we would say that time is cut off from its temporality____
The instant negates continuity."12
Roland Barthes immediately understood that this new literature, labeled literal, paralleled the principles of the new criticism he wanted to promote. In 1955, he wrote a very positive piece on Robbe-Grillet's The Voyeur,13 systematically using Robbe-Grillet's novels and Brecht's theater to promote the "reader's deconditioning with respect to the es-sentialist art of the bourgeois novel."14 The Voyeur accomplished this degree zero of literature and history that Barthes had called for as early as 1953; it portrayed a world of objects suspended on the observer's vision, and which constituted a desocialized and demoralized universe proceeding from a "radical formalism."15 This rapprochement between literary creation and a scientific reflection in language produced a new hybrid, which Barthes called writer-writing.le This new type joined the tasks of the writer (who was to absorb the world into the how-to of writing) and those of the writing, which had to be explained, and for which speech was simply an ephemeral material for demonstrating.
Barthes shifted the traditional boundaries and located the New Novel and new criticism both on the side of the writer, and therefore on the side of creation. Thanks to this division, the critic and writer were united in a common effort to probe the phenomenon of writing and the different possibilities of language. Structural literary theory and the practices of the New Novel constantly interacted; for both, the referent, and the various figures of classical humanism, were marginal. The New Novel abandoned sociological verisimilitude in storytelling in order to concentrate on establishing potential tales and their variations.
This symbiosis between a new literary writing, the New Novel, and a new literary criticism would evolve, in the case of the relations between Barthes and Robbe-Grillet, in the direction of a growing distance from formalization and the construction of an objective realism, and a literal literature. Just as Barthes turned toward the pleasure of the text from 1967 onward, Robbe-Grillet shifted from an objective to a subjective realism as the expression of his own subjectivity became increasingly important in his work."? He too practiced an infinite game of reflections, the mise en abime of characters, plots, staged autobiographical themes, mixed registers. He even reproached Barthes for having misinterpreted his work in 1955, and, to the contrary, demanded total subjectivity: "I have never spoken of anything other than myself."18 According to Robbe-Grillet, Barthes was desperately looking for a degree zero of writing, and his work offered its ostensible realization: "My alleged whiteness-which came at the right time to give balast to his discourse. I saw myself crowned as the 'objective novelist,' or, worse yet, one who was trying to be that." 19
Just as Levi-Strauss considered that a myth was constituted by the whole of its variants, the new novel progressed by repetitions and variations based on which the different laws of the series were played out, but always disturbed by the accident that made the tale lurch forward on the basis of its open structure. This new perspective gave literature a certain autonomy which no longer needed to be demonstrated, committed, or reflected, but that had its value in and of itself. At the same time, according to Barthes, literature could respond to new philosophical questions, by no longer asking the question of whether or not the world had any meaning, but rather the following: "The world is here: is there any meaning in it? ... An undertaking that, perhaps, no philosophy has managed, and that could therefore truly belong to literature. "20 Thus, literature would replace and serve the function of philosophy; it would be the very consciousness of the irreality of language and a veritable system of meaning, once it had been freed from all instrumentalization.
Michel Butor's work is a particularly good example of this mix of theory and practice. He was very involved in epistemology in the fifties before writing his first novel, Milan Passage, in 1954.21 He was working on an advanced degree in philosophy in 1948, under the guidance of Gaston Bachelard, on the topic "Mathematics and the Idea of Necessity." His doctoral thesis, under Jean Wahl, was entitled "Aspects of Ambiguity in Literature and the Idea of Signification." When he began writing novels, he gave up neither theory nor philosophy and considered that the novel was research, an attempt at prob-lematization. This was true for his first novel, in which he problerna-tized space based on a seven-story Parisian apartment building. In his second novel, The Use ofTime,22 time was the central character. In 1960, he explicitly returned to literary theory with his "Essays on the Novel."23 In 1962, he published Mobile,24 orienting his deconstruction of the classical novel by introducing different styles into the same story, juxtaposing sentences, quotations, press articles, collages, montages, and capital letters dispersed across the page. Barthes applauded this revolution that attacked the very idea of the book, after having deconstructed classical novelistic narration. He thought that Butor had dared to address essential elements by taking on typographical norms. "Tampering with the material regularity of the work is to take on the very idea of literature. "25
With Mobile, Butor proposed a new aesthetics, that, like a river's flow, overran the banks containing the tale. Beyond the linear development that gave it an ever-increasing but predictable flow, it varied quantitatively. By contrast, he proposed an aesthetics of discontinuity and juxtaposition.
Structuralism and the New Novel shared a concern for writing per se. This was considered the means for developing critical weapons, so much so that Jean Ricardou proposed the term "scriptural-ism"26for this gush of textuality, as the common perspective of the social sciences and literature.
The Novel of the Human Sciences
Committed structuralists in the social sciences lived this rapprochement with literature so deeply that they took their work to be creative. Deeply moved by a concern for style, the great novels of the period were essentially works in the social sciences. Tristes Tropiques was initially a novelistic project and Levi-Strauss was acutely concerned with the formal aspects of his work, conceived as a musical or pictorial enterprise. Mythologiques had the form of a musical composition with different motifs profoundly inspired by musical development. Lacan's baroque style was deeply affected by his work at Le Mino-taure, a surrealistic art review in the interwar period where he rubbed shoulders with Eluard, Reverdy, Picasso, Masson, and Dali, among others, and later was fascinated by the work of Georges Bataille, whose former wife, Sylvia, he married.
Bataille's experiments with the limits of writing and a barely communicable strangeness fascinated Lacan. Bataille was interested in liberating a constant transgression of rational, social taboos, including the figure of the Other, which reveals itself in the erasure of the self and its traps. He was also an important author for Michel Foucault's style: "Blanchot, Artaud, and Bataille were very important for my generation. "27 These authors showed how to shift literary boundaries of thinking, to move beyond limits, and to destabilize common beliefs by finding breaking points. Examining reason by looking at madness, medicine from the perspective of death, the law from the point of view of crimes, the penal code as seen from prison-these reversals were only possible thanks in part to the experiments going on in literature, and, as far as Foucault in particular was concerned, thanks to Maurice Blanchot's work.
As early as 1955, Blanchot defined The Space ofLiterature-i as that indefinite space within which a work exists in itself, revealing nothing more than its own existence. Like the New Novel, Blanchot rejected the idea of a dialectical relationship with time: "The time of the absence of time is not dialectical. What appears is the fact that nothing appears."29
Foucault paid homage to Blanchot in 1966 as the writer of an impersonal literature with which he completely identified, along with the current of structuralist thinking that defended literariness.>? "The breakthrough in the direction of a language in which the subject is excluded ... is an experiment that is taking place today in a number of different cultural sites."3! Blanchot's writing of exteriority, which places the reader in an initial emptiness, achieved what Foucault wanted to pursue in philosophy: not to use negation dialectically but to make the object of discourse move outside itself, to the other side of observation, in its interior, in "the trickle and the distress of a language that has always already begun."32 Blanchot and Foucault's shared critical activity took the form of suspended meaning, absent from its present, perceivable in its lack. It was no longer a question of seeking an ultimate, profound meaning. Both writers often used the rhetorical figure of the oxymoron, whose effect is both critical and aesthetic. We also see the structuralist and formalist givens, the refusal of all instrumentalized and ordinary language. To the contrary, the work was to try and "accomplish itself in its own experience,"33by rejecting the notion of values in order to reach a level where history was abolished and the present was heralded.
Blanchot and Foucault both bespoke Nietzschean influence. Both rejected reigning values and feared being eo-opted. A double negation resulted: negation of values, and negation of the negation, which led to frequent use of the oxymoron: "pregnant emptiness," "placeless space," "unfinished accomplishment," and so on.34Textualism sheared of values, common to the new novel and to structuralism, found a source of inspiration and a particular aesthetic here. Like the literary avant-garde, philosophical formalist practice could boast of having no external finality and could thus claim to offer a discourse that reconciled logic and aesthetics. It could also shift the boundaries between literature and rational thinking.
When "the being of literature is nothing other than its technique,"35 as Barthes put it, nothing separates critical structuralist activity from a writer's creativity. So we can see how structuralist works could be read, despite their author's denials, as novelistic enterprises. But we can also see how certain structuralists, disappointed or wearied by the search for the fundamental structure or the ultimate code, moved toward pluralization, especially after I968, in order to give freer expression to their literary inspiration.
Disseminating Philosophical Discourse As we have already seen, Jacques Derrida had actively questioned the boundaries separating philosophy from fiction. Deconstruction sought to reveal textual polysemism, the equivocation of what was said, using the undecidables that explode the boundary safeguards and make it possible to disseminate a liberated writing. Derrida turned philosophical discourse toward language and oriented it toward a greater and greater aestheticization.
In the sixties, Derrida was interested in hunting down logo-centrism and phonologism, especially in the work of those claiming to be structuralists. Over time he became increasingly motivated by the pleasure of writing: "I am trying to find a certain economy of pleasure in what we call philosophy."36 This was the pleasure of literary inventiveness and and it was at the very center of the transgression of limits. In I972, Derrida put his own textual work beyond the limits of the canon: "I will say that my texts belong neither to 'philosophy' nor to a 'literary' register." 37
Disseminating philosophical discourse barely distinct from literary writing was particularly palpable in Derrida's 1974 Gfos,38We find the same deconstructive vision of the book as a hermetic universe as in Michel Butor's work, the juxtaposition of different typographies, of parallel columns with different contents. With neither beginning, nor end, nor story, nor characters, Glas was primarily a formal search that was joining in the adventure of the New Novel. "The rain chased away the spectators who scatter in all directions. What is the issue, finally? To talk about the Scotch broom for pages and pagesr-? To interpret or execute it as a piece of music? Whom are we trying to fool?"40 Derrida also tried to open up Jean Genet's work't! by taking the philosophy/literature confrontation as far as it could go in a mosaic of separate texts with words dismantled into a true puzzle-for example, cutting the word gla from viaux two pages later.v' Speculative considerations, scientific ideas, and "autobiographical fragments "43 were dealt out in a sort of self-analysis that took the text as a pretext to destabilize the basic oppositions of Western thought. "A signature maintains nothing of what it signs. Plant a Scotch broom there, the inscription on the tombstone, the funerary monument is a broom plant: who writes, or rather speaks, without an accent ... 'Your name?' 'Genet.' 'Plantagenet?' 'Genet, I say.' 'And if I want to say Plantagenet, what's it to you?' "44
In this new discursive economy, structure was open, plural, and shattered. The notion of difference and Other, which lay at the root of early structuralism and research in structural anthropology, henceforth moved toward disseminating the very idea of structure.
Gilles Deleuze's work made this shift quite palpable in his play on the notion of difference against the Hegelian notion of unity, and he proposed aestheticization in its place: "We find that the history of philosophy should play a rather analogous role to that of a collage in painting. "45 Difference and repetition replaced identity, reversing Hegelianism. Deleuze considered that this demonstrated the advent of the modern world, the world of the simulacrum, the world of a new baroque more attentive to formal invention than to variation in content. A whole rhetoric of pleasure developed and Deleuze, in a writerly way, ceaselessly produced new pleasures, continuing to play on new notions become concepts in his reading of the world. He wanted above all to escape the history of philosophy, and in this he shared the structuralist sensibility. He denounced structuralism's eminently repressive relationship to creativity, calling it a "properly philosophical Oedipus ... a sort of fucking [enculage] or, which amounts to the same thing, an immaculate conception. "46
In place of shunned Hegelianism, Deleuze also proposed a plurali-zation that must run through writing, and thinking at variable intensities that could be cut up any which way. With Difference and Repetition, Deleuze shifted toward movement in structure: "Treating writing like a flux, not like a code."47The impact of May '68 was fundamental in this determination to pluralize in order to give the desiring machines a place in relation to the One, to established thought. Improbability and uncertainties took priority, as they did in Derrida's work, but more radically still was the call for a desiring flux. "Writing is one ebb among others and has no particular privilege with respect to the others and the relationships of current, countercurrent, and collisions with other flows, of shit, of sperm, of words, of acts, of eroticism, of coins, of politics, and so on. "48
Paradoxically, these flows revealed one of the major aspects of the structuralist paradigm: the absent subject. The idea was a functioning machine, and the ego yielded to the id of the desiring machine, coupled, connected at every point. Codifications and decodifications were made and undone with neither faith nor laws, polymorphous figures, rootless, slippery monads.
The notion of closure and interpretation was violently attacked in 1972 when Deleuze and Felix Guattari published Anti-Oedipus (volume i of Capitalism and Schizophrenia), which quickly became the anti structuralist war machine, helping to speed up the paradigm's deconstruction. Immediately and fabulously successful, Anti-Oedipus was symptomatic of the changes afoot and signaled the impending decline. First there was the violent return of the Lacanian repressed. The return of Freud, which Lacan had helped, had privileged the Signifier, and the Symbolic, the notion of an unconscious gutted of its affects. Deleuze and Guattari vehemently challenged this approach, arguing against Lacan's dear Law of the Master and for the necessary liberation of desiring production. All the same, they acknowledged the merit of Lacan's work for having shown how the unconscious was woven of many signifying chains and his breakthrough in imposing the acceptance of a schizophrenic flow that could subvert psychoanalysis, particularly thanks to his objet petit a. "The objet petit a erupted at the core of the structural equilibrium like an infernal machine, the desiring machine. "49 Lacan, less than his disciples and psychoanalysis in general, was the target. Deleuze and Guattari were as sardonic about psychoanalysis as Michel Foucault and used Madness and Civilization: A History of Insanity in the Age of Reason to set psychoanalysis in a direct line with nineteenth-century psychiatry, which reduced madness to a "parental complex" and considered the figure of avowing guilt, produced by the Oedipus complex, to be important. "So instead of being part of an enterprise of effective liberation, psychoanalysis is part of the most general bourgeois repression, which has meant keeping European humanity under the yoke of Mama and Papa and endlessly having to deal with this problem."50
For Deleuze, psychoanalysis was reductive, systematically driving desire back into a closed system of representations. "Psychoanalysis only elevates Oedipus to the square, transfer Oedipus, the Oedipus of Oedipus____It's the invariable turning away of the forces of the unconscious." 51 Deleuze and Guattari differentiated between capitalism, which is enmeshed with psychoanalysis, and revolutionary movements, which make their way alongside schizoanalysis. For them, as for structuralism, there was no Signifying Subject, no specific site for any transcendance whatsover; there were only processes. To express this opposition, they compared a tree with a rhizome, whose polymorphic character could represent a different mode of thought, an operational idea for promoting a new sort of philosophical writing going in all directions without codification. Recourse to logic became meaningless; this kind of writing was obviously removed from the epistemological considerations of early structuralism, and gave free rein to unarticulated, disruptive thinking, at the whim of poetic inspiration.
Above all, Deleuze and Guattari criticized Levi-Strauss, the father of structuralism, comparing two divergent logics incarnated by the desiring machine and the anorexic structure. "What do we do with the unconscious except explicitly reduce it to an empty form from which desire itself is absent, expelled? Such a form can surely define a pre-conscious, but not the unconscious. "52 Conversely, Levi-Strauss found grace in Deleuze and Guattari's eyes because their definition of schizo-analysis echoed his evaluation of the Oedipus complex. They used the myth of reference from The Raw and the Cooked, in which Levi-Strauss demonstrated that the true guilty party in the incest story between mother and son was the father who, because he had wanted to avenge himself, was punished and killed. The authors concluded that "Oedipus is first of all a notion of adult paranoia, before being a neurotic infantile feeling." 53
Alterity elevated to a mode of thought encountered structuralism's antihistorical inspiration. Instead of history, there was a very special consideration of space, a veritable cartography of structure as an open system: "Each thing has its geography, its cartography, its diagram,"54 whereas time could not be homogeneous, and indicated an inevitable disaggregation for it is trapped in discontinuous processes that establish its contingent wrenches. "Modes of thinking about difference reject history as a simple surface effect." 55 The fact that semiotics at the beginning of the seventies was moving toward textu-ality and the concept of writing also made it easier to express poetic, creative inspiration freed from any specific model, at a time when Saussureanism, Chomskyism, and pragmatics were facing off.
Philosophical pluralization was in fact contemporary with the multiplication of models and concepts in semiotic projects. The resulting relativization and ever-deferred hope of finding the ultimate key consoled those who had taken the aesthetic path, reinforced by the perceptible crisis since the sixties. "An 'age of suspicion' among semi-oticians reiterates and reinforces that of the novelists themselves." 56 This crisis opened up writing receptive to those who substituted the pleasure of the text for the desire to codify it.
Roland Barthes adopted this philosophy of desire. For him, the tension had always run deepest in his concern for theory and the expression of affect. With S/Z and The Empire of Signs, he had already begun to pluralize codes and allow a liberated intuition to express itself in an open system. This new orientation was confirmed and the choice of aesthetics explicitly announced in 1973, with Barthes's The Pleasure ofthe Text. A page had been turned. Barthes turned his back on The Semiological Adventure; the writer Roland Barthes could now free himself from the writing Roland Barthes and reveal his taste for stylistics even further. He could reveal himself to himself without having to hide behind a theoretical discourse.
Barthes thus claimed the writing as space of pure pleasure, as proof of desire and pleasure. He fully assumed his subjectivity as much in the act of writing according to his own system of tastes and distastes as in the system of readerly reactions in which judgment depended on a completely personal textual pleasure. Giving free rein to pleasure was the ultimate means of eliminating what Barthes, since beginning his research, never stopped tracking down: the signified. "What pleasure suspends is the signified value: the (right) cause."57Of course, he remained faithful to certain of his major theoretical positions and repeated that the author, the writer, did not exist: "The author is dead."58 The author had no other function than as a mere plaything, a simple receptacle, a degree zero like the dummy in a bridge game. Binarity was used to show the difference between what Barthes called pleasure texts and texts of jouissance. The former fills up and can be spoken, whereas the latter is an experience of loss, for which there are no words. Barthes's important philosophical reference here is the same as Deleuze's: Nietzsche, who is used to explode truths based on stereotypes and old metaphors to liberate the new and the singular.
Barthes compared the foreclosure of pleasure produced by two moralities: stereotypical platitudes of the petite bourgeoisie and the rigor of groups. "Our society seems both to be satisfied and violent, and, in any case, frigid." 59 The pleasure of the text opened onto the infinite, incessant intertwining of a creative opening in which the subject undoes itself by revealing itself. "The Text means Fabric,"6o notin the sense of having to look for the truth on its reverse side, but as a texture summing up its meaning. In 1975, answering Jacques Chancel on his famous radio program on France-Inter, Radioscopie, Barthes recalled that he had started writing because he thought he was participating in a battle, but that he slowly discovered the truth of the act of writing. "We write, finally, because we like doing so, because it gives us pleasure. Finally we write for reasons of jouissance. "61
Barthes was not a pure hedonist; he still had the semiologist in him and pursued his work on textuality. But his aesthetic choice showed a major difference between the 1966 Barthes of euphoric theorizing and the Barthes of 1973. More than a singular itinerary, this break showed that the structuralist program was losing steam, that the crisis of 1967-68 had affected it, and that a solution was being sought. Barthes's new path announced a certain number of returns which would break the surface beginning in 1975. While waiting, like Hegel's aged Greek who ceaselessly questions the rustling leaves and the shiver of Nature, Barthes reflected on the shiver of meaning "by listening to the rustle of language, this language that is my own nature, a modern man. "62
Twenty
Philosophy and Structure: The Figure of the Other
Classical philosophy did not do well during the turbulent times of structuralism. During the seventies, those who did theory and epistemology tried to avoid the label of philosopher. Western Reason was yielding to an increasingly passionate quest for a different figure of alterity. Not that philosophy was dead, however; it was simply addressing the social sciences, trying to designate the Other in space (anthropology), the selfs Other (psychoanalysis), and the Other in time (historical anthropology).
The post-'68 generation, like that of the fifties, continued to be attracted to and converted to these new and promising disciplines whose success seemed to usurp the role philosophy had played in the classical humanities. But philosophy had not lost its flair, as it was primarily philosophers who spearheaded this reappropriation of the various disciplines in the sciences of man while vigorously criticizing current classifications and disciplinary divisions. Nonetheless, a certain philosophical discourse was in bad shape during this period.
The Dialectic of the Same and the Other Referring to this period-the seventies-Jacques Bouveresse lamented: "Truth no longer held any interest; we had to replace the question of what was true by what was right, as Althusser said."! Not that he abandoned his own philosophical reflection, which ran against the tide, or refrained from provocation. Indeed, he even dared to ignore
the requisite references of the moment-Michel Foucault and jacques Derrida-preferring instead Rudolf Carnap, Gottlob Frege, Ludwig Wittgenstein, Bertrand Russell, and Willard Van Orman Quine. In 1973, Bouveresse published Wittgenstein: Rhyme and Reason,2 a reflection on the relationship between science, ethics, and aesthetics. "It was a deliberate provocation since, at the time, it was practically forbidden or altogether incongruous to talk about ethics. There could only be political or psychoanalytical issues."> Bouveresse, however, situated himself elsewhere to escape the theorizing/terrorizing that amounted to tracking down philosophical discourse with two big guns-psychoanalysis and Marxism. "Ifyou objected, you were never answered, really, you were psychoanalyzed, or else your class position became an issue for analysis."4
Nietzsche, Freud, and Marx were mobilized in the quest for the Other as the underbelly of Western Reason. They formed a reading grid that continued to challenge philosophy, along with a disciplinary logic for psychoanalysis and anthropology, and to pursue their old rivalry with philosophy in order to legitimate their institutionalization. Hermeneutics, with its interpretative assumption of an ultimate textual verity, became the designated adversary. After having opposed hermeneutics with structural logic-a system of relations made autonomous from its content-infinite interpretation came increasingly to be the rule.
After World War 11, Adorno and Horkheimer had already begun to examine the conflictual, dialectic relationship between rationality and myth. In order to establish itself, rationality had had to rip itself from the ancestral terror of myth and progressively master it. But this battle had not really ended in any definitive way; rationality was constantly confronted with its other. "It was a sort of adder nurtured in its own bosom."> But Vincent Descombes underscored the confusion between the two meanings of Other: the other as other-aliud-and the other as alter ego, which gave rise to a strategy of suspicion that gripped rationality itself, caught as a stake in a general conflict between different forces, of which it was only momentarily the most powerful. "In order to recognize the gravity of modern conflict, we end up suspecting that rationality has too easily won: no one is right, reason no longer exists anywhere, there are only forces engaged in a power struggle."6
Thanks to this sort of deconstruction, the successive deaths of
God, Man, and Metaphysics could be celebrated. Dialectics, with its going beyond, could be contrasted with nihilism and its overflow, which went as far as a stylistics of rupture with the academicism set in the service of philosophical argumentation. The philosopher was to give his place over to the crowd, not of specialists in the social sciences, but of people who are discovering the Other. "Here are today's masters: marginals, experimental, and pop painters, hippies and yuppies, parasites, madmen, hospitalized types. There is more intensity and less intention in an hour of their lives than in three hundred thousand words of a professional philosopher. They are more Nietzschean than the readers of Nietzsche."? The dialectic of the same and the other was omnipresent during a moment that tended to assign all the evils of the paranoid-repressive to the figure of the same, and to see creativity and freedom on the other side of the divide.
Given the freedom that social science researchers were demanding, this interplay partially reproduced the philosophical crisis of legitimacy. Raymond Aron criticized Levi-Strauss's ambivalence toward philosophy and his insistence on the scientific character of his approach after being accused by empirical ethnologists of doing philosophy without establishing the scientificity of his structural analysis. "The answer would require that the epistemological status of structural analysis be laid out, which he refuses to do."8
Paul Ricoeur also responded to the structuralist challenge in 1970. He acknowledged the fruitfulness of the approach, but nonetheless considered that it was part of a process of understanding. "The explanatory model called structural does not exhaust all the possibilities with respect to a text."? Ricoeur felt that explanations using linguistics were complementary, whereas it was necessary to open the text so that it could reach the highest stage of interpretation by reappropriat-ing the subject of meaning. Interpretation was an act, an effectuation of meaning with respect to oneself. "The text had only one meaning, which is to say internal relations, a structure. Now it has signification." 10 However, these conciliatory efforts fell on deaf ears as the umbilical cord between the different social sciences and philosophy was being brutally severed.
A good part of the young generation continued to abandon philosophy in order to throw itself into the adventure of the social sciences and the fieldwork it seemed to make possible. Philippe Descola was a student at the ENS in Saint-Cloud in 1970, and was planning to do anthropology. At the time, he thought that his work in philosophy would simply be propaedeutic, so much so that his classmates at the École Normale "realized it and called me a featherhead."l1 He read Maurice Godelier's Rationality and Irrationality in Economics'? with interest, and when Godelier came to give a series of lectures as an alumnus of the ENS, Descola discovered that anthropology was the right way to scientifically analyze different social realities. Having passed the written part of the agrégation exam, Descola failed the orals. Discouraged at the idea of starting over, "I went to see Claude Levi-Strauss and left to do fieldwork after a year of preparation." 13
Sylvain Auroux was also a student at Saint-Cloud. He decided to work as a professional linguist, which meant that he had to deviate from the traditional philosophical path. He joined the teaching staff at the ENS in 1967 and headed a group in the social sciences that invited outside lecturers. This was how he met Oswald Ducrot and discovered pragmatics. He had not adopted the scientism and exclusion of the subject of the structuralism of the period, but did nonetheless believe that this scientistic ideology had made two decisive and positive advances possible: "On the one hand, it killed, and I think for good, the transcendental philosophical subject. In the second place, it led to asking the question once and for all of whether or not the social sciences were constructed in terms of experience."14 Once he had passed the agrégation, Sylvain Auroux was posted to Vernon High School (in the Eure), where he taught philosophy from 1972 to 1974. He was dissatisfied by the utter abstraction of a philosophical knowledge that prevented articulating philosophy and social problems in that it was "totally abstract, restricted to microproblems of historical interpretation. When my students came to ask me what I thought about abortion, I answered that it was not a philosophical problem. We refused to deal with these questions theoretically."15 This comforted him in the idea that he had to leave the well-trodden paths of classical philosophy to become involved in one particular social science-linguistics-in which he would become an eminent specialist.
The Other of philosophy as an alterity that could be observed outside of Europe, borne by the discipline of anthropology, continued in the seventies to pose a major challenge for philosophy. In 1967, Levi-Strauss declared, "Philosophers, who have enjoyed a kind of privilege for such a long time because we have recognized their right to talk about everything and at every moment, must now resign themselves to the fact that much research will take place outside of philosophy. "16
In 1973, Levi-Strauss was elected to Henry de Montherlant's chair at the Academic Francaise, dramatic evidence of structuralism's unstoppable rise. Another potential candidate, the poor prince Charles Dedeyan, who personified the most classical literary history and who had planned to present his candidacy, wisely decided to abandon the competition. Levi-Strauss was the only candidate, but his election was not that easy. To be sure, he was elected on the first round, but the majority of sixteen votes was narrow, given that the minimum was fourteen. However, the entrance of the specialist on the Bororo and Nam-bikwara into the Academic Francaise in 1974 was enough to gauge the distance that he had covered between the beginning of his career in Sao Paulo in the thirties and the consecration he received in 1974 in entering under the cupola: "By welcoming me into your ranks today, you are admitting, for the first time in your midst, an ethnologist."!"
Levi-Strauss did continue to circumvent philosophy in two fields: art and science. He evoked art in the following terms at the time of his election to the Academic Francaise: "There is a painter and a bricoleur
in me who take turns____Take Tristes Tropiques. ... In writing it, I felt
I was composing it like an opera. The transitions from autobiography to ethnology in it correspond to the difference between recitatives and arias. "18 At the same time, he played his scientific hand by publishing, in the same year as his election, a second collection of articles, Structural Anthropology, 19 which covered the period from his famous 19 52 text "Race and History" to his latest articles of 1973.
In this work, Levi-Strauss argued for structuralism's scientific capacity by addressing his two favorite areas: kinship structures and myths. He redefined scientific criteria in the social sciences and argued that the linguist and the ethnologist have more to say to researchers in "cerebral neurology or animal ethnology"20 than to lawyers, economists, or other political scientists. The metamorphosis was thus more to be expected from the hard sciences. Levi-Strauss paid homage to his predecessors who had helped to create a rigorous ethnology-Jean-jacques Rousseau, Marcel Mauss, and Emile Durkheim, and he called for a general humanism that only contemporary ethnology could bring, by reconciling the human and natural orders, which it favored. "Ethnology is pushing humanism to enter a third stage."21
Structural ethnology would succeed philosophy, in an ultimate, democratic, and universal phase that could finally consign philosophical humanism-whether the aristocratic and limited humanism of the Renaissance or a bourgeois market humanism of the nineteenth century-to the past. But this could only happen by eliminating humankind from the center of nature and by putting an end to his historical determination, which Levi-Strauss considered to be the prolongation of a past humanism, responsible for all the major catastrophes of the nineteenth and twentieth centuries: "All of the tragedies that we have lived through, first that of colonialism, then fascism, and finally the concentration camps, have been inscribed not in opposition to or contradiction with what claims to be humanism in the form we have been practicing it for many centuries, but, I would say, practically as its natural progression. "22
Levi-Strauss's success in 1973 made it possible to relativize the increasingly severe criticisms of his work. During that same year, Raoul and Laura Makarius collected their articles from 1967 on and published them in a volume with the deliberately provocative title Structuralism or Ethnology.23 They saw structuralism as the life preserver that ethnologists had grabbed in order to escape the decline of the functionalism linking their fate to a defunct colonialism. They criticized the negation of the reality of phenomena in favor of models that operated as if they were transcendent. Structuralism, in their eyes, led to idealism: "The search for explanations in structuralism is eliminated by eliminating everything that has to do with the concrete, empirical nature of facts."24The Makariuses correlated kinship relations, the origins of exogamy, and changes in the mode of production during the transition from gathering to hunting, whence a severe criticism of the structural point of view insofar as it presented kinship relations as temporally invariable. They criticized the elimination of lived experience, echoing the already familiar position of Edmund Leach, for whom it was "the absence of structure which normally characterizes a whole set of directly observable empirical givens."25
Cracks and variations in anthropological paradigms became apparent in France as of the seventies. In fact, they were sufficiently evident that, as Levi-Strauss was being elected to the Academic Francaise, Christian Delacampagne was writing an article in Le Monde: "We could also allege that structuralism needed this official consecration. The surprising thing, in any case, is that it came at a time when it is being increasingly challenged on all sides." Criticizing this detachment from the object in the name of radicalizing structural techniques continued, and increased, in the eighties. Thomas Pavel saw in it a simple return to pre-Spinozan practices, to precritical exegetical techniques, which thus represented a regression, including with respect to seventeenth-century humanist philology, which had dissociated mystic reading from historical exegesis. It would be a return to the cabalistic principles of reading the Torah, which allowed a free permutation of phonological or lexical units. "As in Levi-Strauss, the perceptible text is frozen in a mysterious disorder of currents of signification legitimated at a very different level."26
Challenged by the Other of primitive society, the philosopher was also contested by the selfs Other-Lacanian psychoanalysis. In 1970, Lacan had just been excluded from VIm, and therefore from the brotherhood of the philosophical elite. He announced his theoretical riposte to those philosophers who had dared to reject him, thereby repeating the gesture of the International Psychoanalytical Association, which had already made him into a rebel. He argued that the site of truth was only to be found in one of the four possible disccursesv-e-the psychoanalytic discourse-from which the three others derived: "The unconscious is knowledge, and by definition an unknowing knowledge. Only discourse can articulate the unconscious."28 We have already seen that Lacan borrowed this notion of discourse from Michel Foucault, but used it against philosophy. The first discourse, that of the master, which is particularly evident in a political context, closes off access to sublimation, directly confronts death, and only retains of the Thing the objet petit a while deluding itself about taking action. The discourse that crystallized Lacan's protest was university discourse, which set itself on a moral plane and sought mastery. It is "the gaping hole into which the subject is engulfed for having to suppose a knowledgeable author."29 The third discourse is that of the hysteric, the man of science: "Science takes its impetus from the hysteric's discourse."3o Given all of this, only analytical discourse escaped the desire of mastery and allowed unconscious knowledge to come forth, in place of truth, as the only signifying knowledge. "Lacan is finally led to identify philosophical and metaphysical discourse,"31 thereby situating analytical discourse as the discourse of discourses, the site of the truth of these discourses.
In 1970, an order by Francois Wahl almost led to a new reasoned, critical dictionary of psychoanalysis, directed by Charles Melman, which was to be a work and weapon solely for Lacan's Freudian School.
I saw quite clearly that it would be a thankless task. My idea was very simple, and that was that I knew that if there were no collective work of the FreudianSchool involving each one of the authors, there would be no FreudianSchool. My idea was to force the hand of destiny because the school was a nebula in which many galaxies were juxtaposed.P
But this would-be competitor of the Laplanche and Pontalis Vocabulary of Psychoanalysis never came to fruition. Lacan, like Levi-Strauss and Barthes, played both sides, in different registers. On the one hand, he did not think that psychoanalysis could be transmitted by teaching, like science, which made him a man of words rather than a man of the written word, someone who implicated himself constantly in what he said and did not dissociate his discourse from analytical literature. On the other hand, the more subjective he became in his speaking, the more he used mathemes, Borromean knots, and tores in order to divest himself of his pathos and set himself within a scientific perspective transmitted by working transference. "Seminars were a vital investment for Lacan because there is no knowledge without a transference mechanism. "33
We have already seen how much collective enthusiasm this analytical discourse, with its claims to be the site of truth, inspired among many philosophers, and particularly the Althusserians, who elected to join the psychoanalytical adventure. Even economists, removed though they were from such concerns, were attracted, as Hubert Brochier's 1972 entrance into the Freudian School attested. "Lacan brought many interesting things to psychoanalysis in France, an attentiveness to the unconscious, a way of manipulating people in the noble sense of the term, from their depths."34 But, as an economics specialist who had chosen the most formalized mathematics possible, Hubert Brochier looked upon Lacanian formalization negatively, if only peda-gogically. What was true for economics-this formalization was the product of academic respectability but contributed nothing in terms of tangible knowledge-was also true for psychoanalysis. For Brochier, the surface of Mobius, Klein's bottle, Borromean knots, and all of Lacan's topological manipulations, laid out with ever greater artful insistence on his seminar blackboard, contributed nothing more to understanding the unconscious than Walras's theory of general equilibrium helped understand how a real economy operated. "We don't always know what it is good for and, when one discusses it with its supporters, they tell you that it has a purely pedagogical value."35 It was, in any case, symptomatic that certain economists felt the need to set their own concepts against those of psychoanalysis, due largely to Lacan's growing influence, which had put psychoanalysis at the center of rationality in the social sciences.
A third figure of the Other, the Other in time, became the privileged object of research during the seventies. This was philosophy's third challenge, and it implied moving beyond a certain number of atempo-ral philosophical categories in order to confront history, beginning with an anthropological approach. Jean-Pierre Vernant did just that. He had also been in philosophy and in 1948 joined the philosophy commission of the CNRS and became interested in the category of work in the Platonic system. He discovered the relativity of the way we pose problems, since we habitually start from contemporary reality and too often turn our anachronistic mental gear on the past. Vernant realized, in fact, that there was no word in Plato's vocabulary to express the notion of work. He historicized his approach and discovered that going from the eighth to the sixth centuries b.c. meant going from one mental universe to another, which was the subject of his first book."
In searching for the notion of work, Vernant found, above all, the omnipresence of religion. Vernant was a Hellenist and a student and disciple of Louis Gernet, who had written an anthropology of the Greek world, and whose totalizing approach, in the tradition of Mar-eel Mauss and his "total social fact," would be represented in the ever-present concern for theory in Jean-Pierre Vernant's work. The other major influence on Vernant at the beginning of the fifties was Ignace Meyerson, a professor of historical psychology, whom Vernant had met in 1940 and who influenced his thinking about Greek man, his conceptual categories, his emotions, his "mental gear," to use one of Lucien Febvre's preferred categories. At the end of the fifties, as we have already seen, after having historicized his object, Vernant struc-turalized it with his reading of Hesiod's myth of the races.
In I958, Vernant analyzed Greek myths "on the model proposed by Levi-Strauss and Dumezil, I proceeded, therefore, as a conscious and voluntary structuralist."37 This first structuralist work on the myth of the races began on a note about Greece in which Dumezil raised the problem of trifunctionality. The Dumezilian line was important for Vernant, who was leaving the Sixth Section of the EPHE to enter the Fifth in I963. Thanks to Dumezil, he often came into contact with these issues. On the occasion of one of these visits, Vernant, who had already gone down a half-flight of stairs, heard Dumezil call down to him. "He said to me, 'Monsieur Vernant, could you come
back up?___Have you thought about the College de France? You
would do well to think about it and go to see Levi-Strauss, for there are a few of us who have you in mind.' So I went to see Levi-Strauss, who said to me, 'No problem, I will nominate yoU."'38
Presented by Levi-Strauss in I975, jean-Pierre Vernant made his entrance into the College de France. With him, historical anthropology rose to the summit of legitimation. But Clio was not in exile with Vernant. He was passionate about movement, the transition from one stage to another, and the psychology/historical anthropology he favored was a science of movement rather than a will to enclose history in any sort of statism, Marx was one of his other major references, and he considered him to be the veritable ancestor of structuralism. Not Althusser's Marx of the postepistemological rupture of the sub-jectless process, for the subject was Vernant's chief concern. "I have never laughed so much as when I read Althusser's Answer to John Lewis. To explain Stalin's crimes by saying that humanism had continued to wreak havoc was completely nuts!"39
Vernant looked at the whole of Greek life, rather than detaching any particular category in order to examine its internal and immanent logic. Heir to Louis Gernet's totalizing ambitions, he did not isolate religion, his favorite field of research-to the contrary. He analyzed political organization, something that was relatively absent from structuralist studies, and he studied its rise thanks to the reforms of Cleisthenes in Athens. The territorial principle replaced genetic organization in the city. "The center spatially translated the aspects of homogeneity and equality, and no longer those aspects of differentiation and hierarchy."40 This new space established by the polis corresponded to another relationship to time, and to the creation of civic time. This double effort at homogenization to counter the divisions, factions, and rival groupings that were sapping the city lay at the root of a radical shift of Greek mental categories. The rise of Greek philosophy, and of reason, was not due to purely contingent phenomena, as Levi-Strauss thought, but was clearly the "daughter of the city."41
A colloquium on Greek myth in Urbino, Italy, in May 1973 compared French structuralism with other interpretations of myths. This gave Vernant the opportunity to clarify his vision of structuralism. The Paris semiotic school was heavily represented, notably by joseph Courtes and Paul Fabbri. Vernant was there, along with his school of historical anthropology. Marcel Detienne presented a paper titled "Greek Myth and Structural Analysis: Issues and Problems," jean-Louis Durant on the topic "The Worker Ritual Murder and Myths of the First Sacrifice," and Vernant's lecture was titled "The Promethean Myth in Hesiod." This generated a clash at the top between the Italian school led by Angelo Brelich and the British empiricism of Geoffrey Stephen Kirk. In his final remarks, Vernant clearly argued for the coherence of his school's approach and, after having claimed that the case studies that had been presented should calm any misgivings about the elimination of history, he defended the structural program loud and clear:
We don't consider structuralism to be a premade theory, an already constructed truth that we look for elsewhere in order to apply it to Greek data. We note the changes in perspective brought by mythological studies like those of Claude Levi-Strauss in the last years, and we test their validity in our field, but without ever losing sight of the specificity of our working materials.v
In reaction to the severe criticisms made against Marcel Detienne's paper arguing that Greek sacrifice grew out of hunting rituals, and that the myth of Adonis was born of an ancient gatherer civilization that had previously existed in Greece, Vernant energetically defended the structural approach.
I would like to ask Kirk one question. Is it enough to baptize History as a reconstruction, about which the least we might say is that it is purely hypothetical, in order to find ourselves branded as conservatives and positivists? To locate myths of sacrifice in the whole of the Greek religious context, to compare the many versions of different periods at the heart of the same culture in light of a systematic order: is this more daring than to gaily make one's way from the neolithic age to fifth-century Greece? ... To my mind, that history belongs at best to science fiction, and at worst to the novel of imagination.f
Vernant created a school around himself, and the work of a whole group of researchers was oriented along the lines he laid out, among them Pierre Vidal-Naquet, Marcel Detienne, Nicole Loraux, and Francois Hartog. This anthropological research on historical data led in particular to a collective work published in 1979 under the direction of Marcel Detienne and Vernant, Sacrificial Cuisine in Greek Lands.r' The book raised the question of Greek daily life and culinary questions, in the manner of Levi-Strauss, not out of an interest in exoticism, but in order to better understand Greek society and its use of sacrifice as a way of pacifying and domesticating violence. In this democratic society, sacrifice was the work of all, but limited to citizens, meaning free men. Women were excluded from this rite, just as they were from citizenship. When they did pick up sacrificial instruments, it was to transform them into mortal and castrating weapons. Cutting up meat was, therefore, a man's job, who served his wife. Sacrifice thus offered a privileged perspective on Greek society from within, and Levi-Strauss saw in this work an important analogy with his own observations about American myths. "The work of jean-Pierre Vernant, Pierre Vidal-Naquet, and Marcel Detienne seems to show that in Greek mythology there are certain levels where we find ourselves practically on the same footing as with the American mind."45 The passionate discovery of different figures of alterity and of the Other made it possible to create a symbiosis between three different kinds of approaches: structural anthropology, historical anthropology, and psychoanalysis. All worked at understanding the other side of Western rationality; this posed a major challenge to philosophy.
The Reconciliation of History and Structure
Fernand Braudel had already reacted to the structuralist challenge in 1958 when he focused historical discourse on a practically immobile history of the long duration. In this way, he contrasted the legacy of Marc Bloch and Lucien Febvre's Annales with Claude Levi-Strauss. These historians were no strangers to the structuralist effervescence: May '68 had shaken up the antihistoricism of structuralism's early days and broadened the possibilities for history, which had already been renovated by the Annales, but reconciled with the structural point of view, with greater attention to permanent features than to changes, more anthropological than factual. For the historians, who had been excluded from the intellectual limelight enjoyed by linguistics, anthropology, and psychoanalysis during the sixties, this was sweet revenge.
It was the beginning of a veritable golden age; readers of works in historical anthropology were avid for new research. The new editorial board to which Braudel handed over the Annates was largely responsible for recuperating and adapting the structural paradigm to historical discourse. In 1969, a younger generation of historians (Andre Bur-guiere, Marc Ferro, Jacques Le Goff, Emmanuel Le Roy Ladurie, and Jacques Revel) turned from economic history to a history more attuned to the study of mentalites.
In 1971, this new team published a special issue on the theme "History and Structure." 1 The title alone clearly expressed a desire to reconcile two apparently contradictory terms, something like the marriage of fire and water. That historians wrote alongside Claude Levi-Strauss, Maurice Godelier, Dan Sperber, Michel Pecheux, and Christian Metz showed that the battles had come to an end and that the times favored collaboration among historians, anthropologists, and semiologists. At the beginning of the seventies, a vast alliance was thus established in order to promote a common research. The decade was indeed rich in interdisciplinary collaboration. Andre Burguiere, who introduced the special issue, clearly recognized the ebb of structuralism in the aftermath of the upheaval of 1967-68, and that historians should seize the opportunity to take advantage of the situation. He argued on behalf of historians for an open and well-tempered structuralist program that could demonstrate that historians were not content with perceiving manifest reality, as Levi-Strauss had said in 1958, but that they were also interested in hidden meaning, in unconscious collective practices, as were anthropologists.
Fernand Braudel had already proposed the tongue duree as a means for historians to perceive structures, and as a common language of all the social sciences. Andre Burguiere went further. He outlined a general program of cultural history and historical anthropology that could unfold on the very terrain of structural studies, that of the symbolic. This was where the structural method could best and most easily show its effectiveness. In 1971, the Annates were arguing for a structuralism for historians. Burguiere held the banner high: "A little structuralism leads us away from history, whereas a lot of structuralism brings us back."? Anthropologists had indeed challenged historians, but the entente cordiale seemed quite clear at the beginning of the seventies, thanks to the anthropologization of historical discourse. In 1971, Levi-Strauss was interviewed on the Annates radio program Mondays on History, along with Fernand Braudel, Raymond Aron, and Emmanuel Le Roy Ladurie. Levi-Strauss admitted, "I have the feeling that we are doing the same thing: the great book of history is also an ethnographic essay on past societies."!
The historians delved into the delights of slow history, the history of permanences. Historiography, in its turn, privileged the figure of the Other, with respect to the reassuring image of the same. In arguing for a structuralized history, the Annates historians hoped to federate the social sciences; this had been Emile Durkheim's goal for sociologists-by getting on the wavelength of the structural model and making history a nomothetic rather than an idiographic discipline.
The first effect of this cross-pollination was of course to slow down temporality, which became practically stationary. The history of events was rejected as belonging to epiphenomena and episodic stories; there was to be a single focus on that which is repeated and that which is reproduced. "As for the history of events, harmonizing the teaching of Braudel and Labrousse led to pushing back margins, or even ignoring them completely."4 The approach to temporality favored large stretches of immobile time, and when Emmanuel Le Roy Ladurie succeeded Braudel at the College de France, his inauguralles-son was entitled "Immobile History."> According to Le Roy Ladurie, a historian does structuralism consciously or without knowing it. "For almost a half-century, from Marc Bloch to Pierre Goubert, the best French historians were systematic systematizers and knew that they were structuralists, but sometimes they were unwittingly structuralists, but without being able to hide it."6 Le Roy Ladurie reaffirmed his admiration for Levi-Strauss's use of structural methods applied to kinship laws and to New World mythologies. Although he confined their use to other shores, Le Roy Ladurie nonetheless retained, especially, for the historian, the idea that reality had to be gleaned on the basis of a small number of variables, by constructing models of analysis. Using Roland Barthes's expression, Le Roy Ladurie presented historians as "the rear guard of the avant-garde,"7 specialists in recuperating the progress made in the other pilot social sciences that they had "shamelessly looted"8-an entirely fair observation that described well this second wind of structuralism, transformed and recuperated by historians. The curriculum Le Roy Ladurie described was overarched by the same scientistic perspective as structuralism, making history a nomothetic, scientific discipline that revealed a long immobile period stretching from the end of the Middle Ages to the beginning of the eighteenth century, or from 1300 to 1700, according to a stable ecodemographic cycle of around twenty million inhabitants on French territory.
Le Roy Ladurie also found a degree zero of history after jakob-son's degree zero of phonology, Levi-Strauss's degree zero of kinship, and Barthes's degree zero of writing. "Zero growth demography"9 gave historians access to important, stable balances. The historian's new task was not to emphasize historical accelerations and changes, but the regulatory agents that reproduced existing balances exactly. This was how microbial agents would be highlighted to explain how the ecosystem stabilized. It was "even more deeply in biological facts, far more than in the class struggle, that the motor of massive history has to be sought, at least during the period that I am studying." 10
Humanity was just as decentered here as in the structural perspective, caught in a net and able only to appropriate the illusion of change. Everything belonging to the important breaks in history was downplayed in favor of the large trends, even if these were part of a history without humaniry.U Le Roy Ladurie ended his inaugural lesson on an optimistic note for the discipline that he saw as conquering once again: "History lived for a few decades in semidisgrace, the little Cinderella of the social sciences. Henceforth, it once again finds the
eminence it merits____History had simply gone to the other side of
the mirror to hunt down the Other in place of the Same."12In the school of slow history, some, like Francois Furet, had in fact already found the necessary antidote for their Communist commitment. Struc-ruralizing history and the moment became, in this case, the lever that could disengage Communist historians from Marxism and the dialectic in favor of scientificity, "The history of inertias is not only a good discipline, but it is also a good therapy against a vision of historicity inherited from the philosophy of the Lumieres.I'P
Naturalizing a history of societies that had become static, like Levi-Strauss's immobile societies, with their simple reproductive machines, adopted the structural program against the dominant nineteenth-century historical voluntarism. Faced with the collapse of revolutions as end points and of attempts at restoration, history flowed into immobility, a static present with neither before nor after, juxtaposing the Same and the Other in space. For some, this immobilization of temporality was accompanied by a political conservatism gutted of all projects: "I willingly acknowledge that this type of history (that of long periods, of the ordinary man) is, finally, a history with a conservative vocation." 14
Above all, those who used slow history as an antidote to the philosophy of the Lumieres were those who had used Marxism as a militant war machine, much like the Stalinist vulgate so popular during the fifties and sixties. This was not true for historians who had not been politically committed and therefore had no need to exorcise past demons. They were not less seduced by structuralism, although they did not view it as an antidote or an alternative to Marxism, quite to the contrary.
Georges Duby, for example, discovered Marxism in 1937 in his philosophy class, but it was never more than an analytical, heuristic tool. In 1980, he recalled its impact on his work and his development: "Marxism had a profound influence on me. I react very violently against those who claim today, following a Parisian fashion, that Marxism did not count for the historians of my generation. It counted considerably for me, and I insist that this be said."15 Reconciling Marxism and structuralism, Duby could propose an attentive, di-achronic reading of structural phenomena. He read the work of the Althusserians closely. "Reading Althusser and Balibar was important because it let me see more clearly that, in the period of my specialization, economics could be secondary to other determinations. I had had a presentiment of this." 16 For Duby, Althusserianism could lend a certain complexity to Marxism.
At the same time, and like his entire generation, Duby keenly felt the challenge that anthropologists had raised for historians. This allowed him to go from economic questions, like the ones he raised in his thesis on the region around Macon during the eleventh and twelfth centuries, in which he studied the seigneurial revolution in the region around the Cluny monastery, to questions about the imaginary and the symbolic, without ever dissociating these two approaches or playing one off against the orher.'? "I am trying to eliminate the mechanism of causality and am speaking about correlations rather than about causes and effects. This has led me to think that everything is determined by everything and that everything determines everything. This idea of indispensable totality makes me think of that." 18 Rather than the mechanistic values of reflection, Duby proposed the coalescence of social levels in their different material and mental manifestations. And he proposed a new program for historians of a history of mentalities conceived not as a means for getting rid of social history, but as the fine point of social history.
Duby's most structuralist work and the most successful illustration of the adaptation of this method to history is The Three Orders or the Feudal Imagination.v This important book is the only one that was written without being solicited, and it shows Georges Dumezil's influence. "I owe an enormous debt to Georges Durnezil. This book would not have been written without him, but he is not a historian-he's a linguist and he is concerned with structure. As a social historian, I wanted to understand how this image operates and is articulated with material reality."20 Duby adopted Dumezil's trifunctionality of sovereignty, war, and fertility but reversed the argument that this was a peculiarly Indo-European mental structure. Where, for Dumezil, in the beginning was the myth, for Duby structure proposed while history disposed. He shifted the focus toward the emergence of myth in the historical fabric, its more or less profound penetration, and its meaning in the social practices where it was used. Yet the society that he studied was traversed by conflicts, which shifted and engendered representations of the world whose form or nature were adapted to the need to strangle conflicts. In this context, ideology did not simply reflect economic domination but produced meaning, and therefore reality, a social order. In Althusserian terminology, it even played a dominant role in feudal society by organizing the relations of production. The ideological sphere in this case played the role of the site of absence, the perfect model of imperfection.
Duby redefined the emergence of the trifunctional order in Western Europe as the product of the feudal revolution. In the eleventh century, the Carolingian empire had expanded and had come under external pressure. Ideological values were reversed: the military system, established on the borders, moved into the center of the social body. The king no longer incarnated the power to make war, but rather the power to preserve peace. Political power changed and focused on maintaining internal stability and defending holy places, churches, and monasteries. But during the same period, monarchical authority collapsed into a multiplicity of counties and principalities. Temporal power had failed, and it was tempting for the spiritual order-monks and clerics-to take it into hand. The social border shifted, and henceforth arms bearers were opposed to all others. There had to be an ideological consensus so that those who bore the burden of a militarized society could resign, but this consensus remained elusive.
The feudal revolution needed a system of legitimation and a model for distributing social labor and assuring the subservience of the greatest number. It was at this time, around 1025, that two different bishops, Gerard of Cambrai and Aldaberon of Laon, spoke of the trifunctional social order: "Some pray, others fight, and still others work" (Oratores, Bellatores, Laboratores). In the absence of political power, the clerics attempted to restore social balance, and the ternary figure seemed to be the earthly echo of celestial distinctions. Duby made clear that this imaginary model made it possible to justify the monopoly of economic and political power by a small, privileged minority and to hide in a tripartite structure the underlying dualism that threatened the system. Trifunctionality ensured not only the complicity of the first two orders, but also the primacy of clerics over non-clerics in the battle for the locus of monarchical power. This structure remained the word of clerics and had no echo during a latent period lasting until the end of the twelfth century at which point lords and soldiers had to impose the absolute distinction among the three constitutive orders of French society in the face of the rise of an urban bourgeoisie.
This structure of three orders moved from ideology to social reality by an inverse effect, whence its creative power. When Philippe le Bel called a meeting of the States General at the beginning of the fourteenth century, the celestial order became a socioprofessional order: the clergy, the nobility, and third estate, a division that lasted until the French Revolution. By plunging into the operation of the symbolic order, Duby showed that it was impossible to imagine a society based on a simple mechanics of reflection, and that a symbolic structure had to be studied through history. "The perfect model of the three orders, tied to the monarchical ideal and elevating the heads of the armies above the others, is a weapon in a polemic against those arguing for a new order, who included both heretics and the monks of Cluny."21
Restored in the conflict that had seen it emerge, structure was not a weapon against history. It offered a possible reconciliation between two approaches that had initially seemed antagonistic.
Twenty-two
Foucault and the Deconstruction of History (I): The Archaeology of Knowledge
When Michel Foucault was writing The Archaeology ofKnowledge in 1968 in Tunisia, he was trying to answer the many objections to the arguments of his very successful The Order ofThings. In particular, he was trying to answer young Althusserians in the epistemological circle of the rue d'Ulm who had just chosen political involvement and had just broken with the Communist Party leadership. The great upheaval that preceded May '68 and continued after it favored the splintering of structuralism. With this work, Foucault wanted to find a way at once to conceptualize his approach and to distance himself from his previous structuralist positions. He elected quite a singular path, suggesting a surprising new alliance with new history historians, heirs to the Annales. He was placing himself on historical grounds in order to work with historians, but engaged in history the way Canguilhem had treated psychology, in order to deconstruct it from within, à la Nietzsche, and his position therefore led to many a misunderstanding.
Foucault himself described the inflection in his thinking between his early and his later work. The History of Madness had paid too much attention to "the anonymous subject of history"; in The Birth of the Clinic, "the frequent recourse to structural analysis threatened to bypass the specificity of the problem presented";' The Order ofThings had lacked an explicit methodological framework, which made it
possible to conceive of analyses in terms of cultural totalities. This methodological framework became the object of The Archaeology of Knowledge, which initially took the form of a preface to The Order of Things. "Canguilhem and Hyppolite were the ones who said to Foucault: don't put it in the preface, you will develop it later."2 The work thus still bore the mark of the triumphant structuralism of 1966, but between its first version and the 1969 publication, not only did Fou-cault's thinking change, but so did the intellectual climate of the times. The most spectacular change was that The Archaeology ofKnowledge abandoned the notion of the episteme that had seemed to organize the breaks operating in The Order of Things. Without presenting himself as a historian, Foucault described his approach in terms very proximate to history. This was symptomatic. Defining himself as an archaeologist, he spoke about genealogy and circled around history in order to situate himself outside of it. This explained his at least ambiguous, and often conflictual relationship with the historical corporation.
Those to whom Foucault addressed himself in 1968-69 were in fact second-generation Althusserians, who had not participated in Reading Capital. They included Dominique Lecourt, Benny and Tony Lévy, and Robert Linhart, among others, who diverged from early Althusserian thinking in that they were more interested in the political aspect of political commitment than in defining a methodological framework of contemporary rationality. "We considered the team that had written Reading Capital to be contaminated by structuralism, and we looked upon that quite critically."3 For these politically committed militants, who were in the main Maoists, the one unresolved problem was that of praxis, practice. Yet the major innovation in The Archaeology ofKnowledge was precisely to consider practice, beginning with the notion of discursive practice. This important innovation allowed Foucault to inflect the structural paradigm so that it went beyond the discursive realm alone, thereby bringing it closer to Marxism. This notion of practice "created a decisive dividing line between The Archaeology ofKnowledge and The Order ofThings."4 The essential break with structuralism lay in this new affirmation, according to which "discursive relationships are not, as we can see, internal to dis-course."5 Not that Foucault was abandoning discursivity, since it remained a major focus, but he did consider it as a discursive practice limited to discourse: "Discursive relations___are not relations exterior to discourse.... They are in a sense at the limit of discourse."6
Foucault justified this historicization of the structural paradigm by using the path taken by the Annales historians, who had radically brought down the three traditional historical idols of biography, events, and politics. His Archaeology ofKnowledge began by describing the not insignificant interest he felt for the new historical orientation: "For many years now, historians have preferred to turn their attention to long periods as if, beneath the shifts and changes of political events, they were trying to reveal the stable, almost indestructible system of checks and balances."? This practically immobile history attracted Foucault, and the epistemological turn undertaken in 1929 by the Annales became an exergue in his theoretical work.
The marriage between the history of the great immobile anchors of events and Foucault's evolving preference for discontinuities and the powerful enigmatic breaks along the lines of Bachelard and Can-guilhem's epistemology of science might seem surprising. It was paradoxical to support the idea of epistemological thresholds on a slow history, but this apparent internal tension was only superficial. Foucault saw a converging evolution between the history of thought, new literary criticism, the history of science that pointed to larger and larger numbers of breaks, and the discernment of discontinuities, on the one hand, and, on the other hand, the historical discipline that makes events sink beneath the weight of structures. "In fact, the same problems are being posed in either case, but they provoked opposite surface effects. These problems may be summed up in a word: the questioning of the document."8
Underlying this was the same transformation of evidence that traditional history had considered a given, but that the new history saw as a construct. New historians took the document and sectioned it, distributing it in series. Its status changed; where the historian of yesteryear was used to transforming monuments into documents, the new historian "transforms documents into monuments:"? The historian thus became an archaeologist, echoing Foucault's archaeology of knowledge project by starting with the constructed series of knowledge, an intrinsic description within them. This led Emmanuel Le Roy Ladurie to comment that "the introduction to The Archaeology of Knowledge is the first definition of serial history."lO Indeed, Foucault announced his program in these same terms: "The problem now is to constitute series." 11 The apparent opposition between the discontinuity at work in the history of science or in new literary criticism and the
historian's predilection and valorization of long periods of immobile time was thus merely superficial. Common conceptions and approaches in fact led serial historians to favor discontinuities: "The notion of discontinuity assumes a major role in the historical disciplines." 12 The historian who must fill the holes and plug up the gaps in order to set them back into continuities from then on gave these discontinuities a heuristic value of a determined operation for defining the level of analysis. With discontinuity, the limits of the object of study could be defined and described based on its thresholds, its breaking points. Finally, rather than a history compressed around a center or a global history, it was a means of constructing "that might be called a general history," 13 which was defined, to the contrary, as a space of dispersion.
Foucault quite explicitly echoed the Annales legacy in defining the new task of an archaeology of knowledge: "What Bloch, Febvre, and Braudel have shown for history, we can show, I believe, for the history of ideas." 14 With this new alliance, Foucault could resolve the apparent dichotomy between a structural method and historical evolution, by presenting new history as one of the possible figures for structuralist studies. History cut across problems in linguistics, economics, ethnology, and literary analysis. "We can, if we want, designate these problems under the insignia of structuralism."15 New history was a privileged ground for setting into motion an open and historicized structuralism; later, this was called poststructuralism in the United States.
This historicizing of structuralism was quite clearly the second period of structuralist history since 1967. "Foucault's archaeology was very clearly distinct from taxonomic structuralism such as that of Instead of thinking about the structure and the sign, Foucault examined the study of the series and the event. But this shift toward history, perceived as a call to arms by the new historians of the Annales, who were to see Foucault as the man who could conceptualize their practice, was in fact only an illusory reinforcement. Foucault remained a philosopher in a Nietzschean-Heideggerian tradition, and he decided to deconstruct the historian's turf. He was interested in the discursive realm and not in the referent, which remained history's privileged object.
In no case did Foucault want to defend any discipline whatsoever of the science of history, however new this history. What interested him was to open the structures up to temporal discontinuity and shifts that determined the endless game of discursive practices. Deconstructing history was already part of the work of the new historians and involved abandoning the search for continuity or attempts at synthesizing different pieces of reality. On the contrary, this deconstruction offered a perspective of pluralization and atomization. As Habermas wrote, in this configuration of knowledge, hermeneutics was given its leave since understanding was no longer a theoretical objective: "The archaeologist will ensure that the speaking documents once again become silent monuments, objects that needed to be freed from their context in order to be available for a description of the structuralist type. "17 What these new historians were going to see as the best theoretical support for grounding their practice was in fact a systematic destruction of the historical discipline. A veritable quid pro quo would be at the bottom of all the misunderstandings in the difficult debates between philosophy and professional historians.
The space of dispersion of Foucauldian archaeology shared some aspects of early structuralism: its protest against of the use of overly simple causalities, its use of a relational network spreading in all directions between different discursive practices. Foucault saw this space as possibly bringing these practices together in a coherent, causal whole. The archaeologist would thus also be a relativist since it was impossible to establish anything whatsoever. In this respect, Foucault broke with Althusser's scientism; he remained a historical materialist, and kept his sights set on a science freed of its ideological setting. As a good Nietzschean, Foucault sapped those apparently well-founded beliefs and apparently most legitimate sciences, arguing that nothing can be founded.
By attacking history, after having studied the example of philology, political economy, and biology in The Order ofThings, Foucault took on a major ancestral realm of knowledge and remained quite faithful to the structuralist tradition. He did not refute history's existence, but deconsructed it from within, a task that, in the Nietzschean early seventies, was far more successful than might have been imagined. Since the bases for knowledge and search for origins could not be discovered, the possibilities were fundamentally descriptive, and Foucault baptized himself a positivist, an infamous epithet for anyone who spoke in the name of a constructed science: "I am happy to be one [a positivist]."18 His method typically circumvented interpretation and let discursive practices play out explicitly and implicitly. "It is true that I have never presented archaeology as a science, or even as the beginnings of a future science." 19 The archaeologist worked like a geologist, satisfied with running his fingers over the temporally successive strata of knowledge, pointing out the discontinuities and breaks affecting their sedimentation.
Foucault Targets Analytical Philosophy Foucault was not trying to forge a new alliance with the Annales historians in The Archaeology ofKnowledge but to criticize analytical philosophy, the dominant philosophy of the Anglo-Saxon world. While writing the book, he frequently and carefully discussed his ideas with Gerard Deledalle, a French Anglo-Saxon philosophy specialist and director of the philosophy department of the University of Tunis who had invited him to teach in Tunisia in September 1966. His polemical objective of strengthening the positions in The Order of Things with criticism in line with the philosophy of language was not explicit in the first reading, and when Dominique Lecourt wrote an article in La Pensee on The Archaeology ofKnowledge, Foucault thanked him. He also let him know that he had missed something fundamental. "He said to me, 'You know, there is something that you did not grasp,' without saying anything more. Now I understand what he meant. It was the position ofstrength that he was trying to impose on analytical philosophy."21 Was The Archaeology ofKnowledge a weapon against analytical philosophy? We might suggest this given Foucault's relationship with Gerard Deledalleand Dominique Lecourt's remarks. But we might nonetheless consider that "this resistance to intentionality, meaning, and the referent most certainly concerns phenomenology more, and Foucault is familiar with that tradition. Or, we might simply say that hermeneutics is hostile to structuralism. "22
In any case, The Order of Things and The Archaeology ofKnowl-edge were fundamentally connected. Both were overarched by the structuralist legacy and equally attacked the theory of the subject; for even if Foucault was moving toward historicization, the issue was the subject, as it had been during the earliest days of structuralism. And, in the manner ofHeidegger, this subject had to be decentered:
what is being bewailed with such vehemence is not the disappearance of history, but the eclipse of that form of history that was se-cretly, but entirely related to the synthetic activity of the subject; ... what is being bewailed is that ideological use of history by which one tries to restore to man everything that has ceaselessly eluded him for over a hundred years.23
In the same perspective of The Order ofThings, Foucault addressed the master of creation-man himself. The archaeology of the social sciences showed that the many narcissistic wounds, from Copernicus to Freud via Darwin, had slowly but surely dispossessed man of any illusory sovereignty. The archaeologist was to take this evolution seriously and avoid restoring a humanistic anthropology since "man is an invention of recent date. And one perhaps nearing its end. "24 Before analytical philosophy and its pragmatic studies with their meaningless language acts, Foucault proposed an autonomous discursive sphere that focused exclusively on the interplay of utterances diffused within discursive formations. "The study of discursive formations requires a reduction of two orders. Not only must the archaeologist make truth an abstract notion ... but he must also reduce its pretension to meaning to an abstraction. "25
This work presented the now classical normalization of the signified and of the subject characteristic of structural linguists, which appeared as the necessary condition for addressing language by description alone. Describing the function of utterances and enunciation implied, according to Foucault, an absolute neutrality and exteriority to all enunciations, by contrast with analytical philosophy, which sought out and effaced the meaning. The archaeologist limited himself to describing existing utterances: "The archaeologist does not take utterances seriously. "26
Above all, rather than trying to frame discursive logic within false continuities, as biographies did, Foucault tried to point out the archaeological slices and shifts from one discursive formation to another, the lags and discordances. He tried to "describe the dispersion of the discontinuities themselves. "27 This concern for description within an autonomous discursive sphere was clearly part of the structural linguistic legacy and its rejection of meaning and the referent. "The archaeologist claims to speak without any concern for comprehen-sibility."28 There was, in fact, no signifier for Foucault, neither the speaker's intentionality nor the referential framework nor any occult meaning. He began and returned to the utterance as a moment to be lifted from its atemporality.
The archaeologist's work of decentering the subject led Thomas Pavel to compare Foucault's conceptual layout and that of distribu-tionalists such as Harris and his disciples: "The similarities have to do especially with the rejection of mentalist ideas____The intentional notions Foucault criticized included tradition, disciplines, influence, evolution, mentality, in a word, all the historical forms of coherence and continuity."29 We can better appreciate the misunderstanding between Foucault and historians who criticized the historical validity of his theses and accused him of using utterances outside of their context and their specific historical issues. For Foucault, the notion of utterance or of discursive formation had no empirical content; his approach was set in the limits of discourse in order to concentrate on the conditions of its possibility rather than on its content or the meaning of the discursive exchange, or the concrete propositions studied by an analytical philosophy that he considered meaningless.
While Foucault focused exclusively on discursive formations, he continued to reject linguistic methods for describing language. The path he defined, that of archaeology, seemed to offer a third option to the techniques of linguistic formalization-semiotics-and philosophical interpretation, or hermeneutics; it was located midway between structuralism, for which it was a theoretical framework, and historical materialism. Gilles Deleuze described this using the musical assessment of Webern's universe: "He created a new dimension, which we might call a diagonal dimension."30
Foucault resisted reductionism by systematically setting himself on edges and interstices between genres. Discourse, the central idea of The Archaeology ofKnowledge, lay between structure and event. It contained the rules of language that constitute the linguistic object of predilection, but was not limited to that, because discourse also encompassed speech. By discourse, Foucault meant the structural dimension and the event: "treating it sometimes as the general domain of all statements, sometimes as an individualizable group of statements, and sometimes as a regulated practice that accounts for a certain number of statements."31 His was a position of constant tension since he refused both hermetic discourse and its elucidation by elements external to language.
Since discourse did not refer to another order of things, Foucault emphasized the notion of discursive practice that let him get beyond the sign, but this is no way meant that he abandoned the idea of an autonomous discursive realm: "Discursive relationships remain most important."32 Foucault thus retained structuralist notions of the initial break between language and its referent; he also held that discourse was the most important object of study. But he studied it as a philosopher and not as a linguist, keeping discourses at a distance, shifting and rotating them to study them at more than just the most immediate level. Beneath the discursive surface, which he took as his starting point, Foucault let different discourses play against each other to discover other potential organizations. Beneath the game of simulacra, he clearly meant to describe the specific rules of discursive practices by loosening the links between words and things, and by avoiding using the contextual circumstances in which the discourse unfolded to explain its background. The archaeologist was not to define thought or representation beneath the discourse, "but those discourses themselves, those discourses as practices obeying certain rules."33
Unlike analytical philosophy, archaeology rejected the meaning of language acts and the reference to a subject. But, unlike the linguist who argued for the iterability of schemas belonging to a language system, Foucault took concrete utterances with respect to time. The archaeologist was to measure the degree of validity of a moving body as it evolved over time according to its position in discursive space and the precise moment of its utterance. These shifts and connections between different discursive realms led to questions about the division of sciences, disciplines, and established fields of knowledge and their own corpus and system of scientific rules. Thanks to the archaeologist, a certain dominant transversal discursive mode could be discerned over all the modes of knowledge of a given period.
His basic unit was the utterance, a true thing set in an in-between zone, with language as a system of rules on the one hand, and a corpus of discursive utterances on the other. This utterance is not the enunciation of analytical philosophy, and yet it is not hermetically sealed since "a statement must have substance, a support, a place, and a date. "34 Starting with the stuff of statements, Foucault did not intend to layout a synthesis around a subject, but rather a space of dispersion using the many ways in which enunciation functioned. What established and unified the utterance was no longer its integral unity but a law of distribution, specific constitutive rules where the level of relationship was the fundamental issue. "1 have undertaken then to describe the relations between statements. "3 5
Description and not causality was therefore the archaeologist's first task. The rules of utterances were as unconscious as epistemes but were more historicized, referring to a given time and space, a social, geographical, economic, and linguistic zone. Discursive practice had its place more within social realities by virtue of the organic institutional relationship that simultaneously established and limited it. Thus it was up to the archaeologist to discern the set of utterances belonging to the same discursive formation. For Foucault, the enuncia-tive space supposed a certain number of rules. Gilles Deleuze distinguished three successive circles around the utterance: a collateral and adjacent space; a correlative, organizing space that marked sites and viewpoints; and finally, a complementary space of nondiscursive practices: institutions, political events, and economic process.w This third space, which was in no way causal for Foucault, represented the essential flexion for getting out of a particular structuralism with its hermetic concept of discourse.
This was also the major personal change for Foucault and for his work until then. He had already replaced the episteme with the notion of discursive practice, and he went even further toward a materialist approach by integrating the relationship between discursive and nondiscursive practices into his work, even if it was only a question of a third circle conceived only as a visual limit. Starting from these three circles constituting the utterance, the archaeologist was to point out the iterative conditions of the utterance: "There must be the same space of distribution, the same distribution of particularities, the same order of places and sites, the same relationship with the established milieu: for the enonce all of this constitutes a materiality that allows it to be repeated."37 Such transitory figures were mortal languages and not universals. Thus Foucault ultimately defied any attempt at adopting any form of historicism or humanism. His conception was something fugitive and polymorphous. Discursive practice did not refer to the activity of a subject but to the rules to which it was subjected. As Deleuze put it, the approach was essentially "topological," and not typological.
The issue was to trace the speaker's status, placements, and positions, referring his or her discourse to a particular point in space. Foucault specifically asked the question of the speaker's place: "Who is speaking? Who, among the totality of speaking individuals, is accorded the right to use this sort of language? Who is qualified to do SO?"38 Medical knowledge does not operate chaotically, nor does it refer exclusively to its internal logic. For a doctor to be a doctor supposes criteria of competence. A medical decision or act takes its value from the person who performs it, from the doctor's socially recognized identity and institutional place. Specialist or generalist, intern or extern, doctor or health administrator, each status corresponds to a certain competence or practice in a medical and social hierarchy. "Medical statements cannot come from anybody."39 Discursive practice is clearly located within nondiscursive practices that should be reintegrated into the archaeologist's perspective.
This was the aspect of The Archaeology ofKnowledge that interested Dominique Lecourt most when he reviewed it in La Pensee in August 1970. A Marxist, he saw it as an important step forward and an important departure from The Order ofThings. Foucault's concept of practice, establishing a theory of discursivity structured by relationships invested in institutions, could not help but remind readers of Althusser and his followers of his shift toward practice. By devoting so much space to Foucault in La Pensee, one of the important theoretical publications of the peF, Lecourt wanted to make him better known, contrary to the party's rejection of him. "I liked Foucault enormously as a philosopher and as a man. This article was an attempt to translate what he was saying in his own terms, but in our vocabulary-ideologies, ideological state apparatuses-and to say that we could go farther, as it was fashionable to say at that time."40
Lecourt was delighted to see Foucault abandon the episteme, the cornerstone of The Order ofThings ("Here, Foucault wants to rid himself of the structuralist aspects of the episteme")," and turn toward the idea of discursive practice, and in so doing renew his ties with materialism. Since this idea was based on the materiality of the discursive order and pointed to institutions, it also pointed to AI-thusser's ideological apparatuses of the state. And yet, Lecourt considered that there was a vanishing point when Foucault strictly limited the archaeologist's task to description, with no hint of theorization: Foucault had stopped midway despite a promise to go further in the direction of a materialist theory of the formation of ideological objects. He stopped before defining the relationship between discursive and nondiscursive practices: "When the essential difficulty of the 'link' between ideology and the relations of production comes up, Foucault goes silent. "42 In Lecourt's Althusserian critique, Foucault's attempt failed because it did not articulate ideological formation and social relations, and remained the blind spot of Foucault's thinking that necessarily pointed to Althusser's concern for reconceptualizing the distinction between science and ideology.
The Archaeology ofKnowledge came out at a turning point in the structuralist paradigm, and was part of this adaptation of theoretical antihumanism to a new intellectual landscape. An expectant public awaited the successor to The Order of Things. It was well received, and sold more than 10,000 copies since 1969 (II,OOO copies in its first year, 45,000 copies by 1987). Jean-Michel Palmier, in an article in Le Monde titled "The Bell Tolls for Historical Thinking: The Death of the King,"43 describing Foucault's theoretical development, wrote that he deconstructed the beautiful philosophical dream that claimed to speak the essential things about the world, life, and morality, God and history, and offered in its place a careful and detailed reading of the past with his archaeology. In La Quinzaine litteraire, Francois Chatelet hailed the destruction of traditional history.i" Regine Robin recognized her debt to Foucault for having established the necessary relationship between discursive and nondiscursive practices, to which she was particularly sensitive as a historian receptive to linguistics and who favored a rapprochement between history and linguistics.v But this debt of historians was limited because Foucault never articulated the discursive level with the articulated whole of social formation; thus Robin's criticism echoed Lecourt's and that of the Althusserians. Jean Duvignaud was severe, insisting more on the continuity between The Archaeology of Knowledge and structuralism, Foucault having wanted "to dissolve self-consciousness in the discourse-object, "46 which speaks in us and for us but without us, and opens onto a dehumanized universe.
It is true that Foucault held to his antihumanist positions in 1969. His chief objective had been to decenter man, the author, the subject, the speaker. By plunging him into discursive regularities, he announced a new era of faceless writing, a period of pure freedom. "I am no doubt not the only one to write in order not to have a face. Do not ask who I am and do not ask me to remain the same: leave it to our bureaucrats and our police to see that our papers are in order. At least spare us their morality when we write."47 In 1969, this was a way of saying that while he continued to wage war against humanism and all theories of the subject, Foucault refused any structuralist recuperation. At a critical moment in the structural paradigm, he sought the means of freeing himself from himself and his past work by defining a third, neostructural path leading to new areas of investigation.
Foucault and the Deconstruction of History (11): Discipline and Punish'
Nietzschean deconstruction quickly took precedence for Foucault, whose rapprochement with certain of Althusser's ideas was only very fleeting. Theorizing the failure of the May 1968 clashes, Foucault began to get interested in the periphery, the margins of the system; he therefore became politically active on behalf of the often forgotten social marginals. Rather than a notion of revolutionary theory and practice, he proposed revolt. Nietzsche's influence was quite strong here, for where Foucault had forged a dialectic of discourse and power in his earlier works, he now added a third term, the body. This trilogy functioned in its extremes: the body and power reflected each other like Being and Nonbeing. Freedom faced off against constraint, desire against the law, revolt against the state, the many with the assembled, the schizophrenic with the paranoid. The subject's subjection passed through a third term, and discursivity belonged to the realm of power because it was consubstantial with knowledge.
From the Archaeology to Genealogy
The genealogical turn came in 1970-71, and it took three directions. First, on the occasion of an homage paid to Jean Hyppolite, Foucault gave an important lecture on history as genealogy, or as a carnival, taking as his starting point Nietzsche's relationship with hisrory-' He placed genealogy at the hub of the articulation between the body and history, and he proposed concentrating on this body, forgotten by his-tory and yet its basis. "The body: the surface on which events inscribe themselves (whereas language marks events and ideas dissolve them)."3 Thus Foucault laid out a veritable political economy of the body, tracing the different forms of its enslavement and unveiling its modes of visibility.
He sought forgotten, repressed, and incarcerated bodies in order to give them voice. In collaboration with others, he created the Prison Information Group (GIP) in 1971 so that his theoretical positions echoed his political practice. But at the beginning of the seventies, Foucault also had to define a curriculum when he became a member of the College de France. This was the topic of his inaugural lesson of December 2,197°, later published as The Discursive Order/: His was a hybrid program made up of rules put forth in The Archaeology of Knowledge but recast in a genealogical perspective, a noteworthy shift from the archaeologist's vocation. In particular, the question was no longer the relationship between discursive and nondiscursive practices; Foucault again focused solely on discourse, this time articulating it with the body. His genealogical program had always been set on historical terrain, which became the privileged object of his critical analysis. Foucault placed himself clearly and exclusively within the discursive sphere, for which "the character of discourse as an event had to be restored,"! by once again questioning the West's search for truth and abandoning the sovereignty of the signifier. The methodological rules previously defined in The Archaeology of Knowledge were here in the serialization of discourses, the observation of their regularity, and the conditions of their possibility. This was a turning point for Foucault, who presented his program as a critical program, in the line of the Archaeology, and announced his future genealogical work. The two perspectives cohabitated, but one took priority over the other during the course of the decade.
A genealogical orientation inspired the publications of the mid-seventies: Discipline and Punish and The History ofSexuality, volume 1 (1975 and 1976). "The genealogist is a diagnostician who looks at the relationships between power, knowledge, and the body in modern society."6 Foucault fleshed out the initial structural perspective with this corporal dimension, with the confrontation of desire and the law with disciplinary systems. But he continued to deny the validity of historical continuity or of any subject in a game where anonymous strategies of domination began with the body. In the genealogical framework, the subject was neither individually nor collectively pertinent. It could only be the object of many organizations of common centerless forces in social space. Localizing power and knowledge got special treatment in the political technology of the body, which Dreyfus and Rabinow called "biopower,"? From a genealogical viewpoint, knowledge has neither an objective nor a subjective basis, and science should be questioned in order to understand how truth effects are essentially effects of power.
Foucault's program was to track the underside of Western positiv-ities, to find the repressed figure of the Other. To do this, he exhumed the disciplinary procedures veiled by the liberating discourse of the Lumieres, the terror coiled up beneath humanism, the fundamental issues of power lying at the heart of the sciences. Thus he maintained his ascerbic critical stance with regard to Western modernity and the reign of reason, to which he contrasted the carnival of history. The notion of power, dispersed and diluted and yet everywhere present, became an instrument for deconstructing the categories of Western reason. "In Foucault's genealogy, 'power' is first of all a synonym for a purely structuralist function; it has the same position as 'differance' does for Derrida."8 For Habermas, Foucault contrasted Kantian idealism with a temporalization of the a priori, of power in reversed form. Power no longer depended on truth, but truth underlay power's domination; power was considered a founding and subjectless category. This double meaning of power was at the root of all the misunderstandings with historians, for it was at once a descriptive instrument for the various techniques used to subject the body, and an a priori category that made it possible to establish criticism. Here, Foucault's notion of power clearly included an ontologized structuralist category, irreducible to an empirical reality. "When I say power it is not a matter of spotting an institution that spreads out its network in a fatal fashion, a network that tightens its hold over individuals. Power is a relationship, it is not a thing."9
Problematizing Power
In the seventies, the main shift in Foucault's positions was that he became personally involved in his theoretical object of study. (A similar thing had happened at the same time for Barthes, but in another register.) This involvement was particularly obvious in Discipline and Punish, which came out in 1975. As Daniel Defert remarked.!" a footnote in The History of the Asylum had already forewarned of his work on prisons in 1961. But this was above all the result of Foucault's commitment to what, in the seventies, were called the secondary fronts, the peripheral battles, for want of having been able to make the center budge.
In February 1971, Daniel Defert and his Maoist comrades asked Foucault to create a commission to hold an inquest on penitentiary conditions. Not only did Foucault agree but he became completely involved in this militant initiative. He became the head of the Prison Information Group (GIP) in 1971, along with Pierre Vidal-Naquet, the classicist, and Jean-Marie Domenach, the editor in chief of the review Esprit. The GIP was housed at Foucault's home, where he received the families of prisoners and listened to their stories every Saturday from 4 p.m. on after his visits to prisons. His investment of time and his devotion were absolute, so much so that he put off working out his theoretical project, which would be published only after this militant phase was over. "Foucault's idea had been to have the prisoners talk. He did much more than I did. There was this altogether strange mix in this encounter between Foucauldian structuralism, a post-Marxist '68er in search of revolutionary forces, and an evangelical Christianity that, along with Maoists, provided most of the forces in the GIP."11 In this climate of discussing prison reforms and then of proliferating prison protests, the GIP came to play an important role. It was joined by many intellectuals, such as the Vincennes group including Jean-Claude Passeron, Robert Castel, Gilles DeIeuze, Jacques Ranciere, and an unexpected recruit who developed a very deep friendship with Foucault and who also got fully involved in this combat-e-Francois Mauriac's son Claude, a journalist at Le Figaro at the time. From 1971 to 1974, Foucault was at every mobilization for prisons and the GIP increased its types of actions: demonstrations, circulating information, personal testimonials, and critical thinking on the repressive practices of power.
Discipline and Punish only came out after this militant period. At the crossroad of many paths, it illustrated the ambitions of The Archaeology ofKnowledge of going beyond discursivity to link discursive and nondiscursive practices. But at the same time, it was a genealogical research program in search of the points where power imposed itself on the body and traced the mode of problematizing prisons at a very spe-cificmoment in Western history. For Foucault, prisons were one among many expressions of the exercise of power.
His approach took leave of a Marxist-Leninist notion of instru-mentalism and worked toward its pluralization. Power no longer had a center; it circulated, and it determined relationships. "During the structuralist period, we were between Lenin's The State and Revolution and Foucault's thinking on power. "12 Foucault made politics recede by broadening the definition of the scope of power, its extension into the furthest reaches of social life. The state disappeared as the nervous center of the social body. This was the antithesis of Hobbes's approach in the seventeenth century, where the state, the Leviathan, was the epicenter. Unlike Hobbes, Foucault wanted to restore the reality of those ignored peripheral bodies, disparaged as mere epiphenom-ena. This done, he could discover the order and hierarchy of an order beneath an apparently disordered inorganicity.
But Foucault's notions of power diluted the political dimension by dispersing it in all directions. No longer assigned to a particular class, it circulated through a network among individuals, operating in chains, transiting through each one before reassembling into a whole. Without any nodal point there could be no resistance to this omnipresent power that was in everyone and that was therefore nowhere. It was irresistible since there was nothing to resist. Foucault's analysis did not confuse power and the state, but often came at the price of negating the state's existence, to the benefit of a single concern-the body.
A condemned man's body was caught between different meanings of power networks. From expiating his crime during the era of public punishments to correction by a prison term within the panopticon, the process remained circular: increasing knowledge, incarnated by the Lumieres, and increasing power by extending the disciplinary fields. Foucault historicized the carceral procedure by studying the conditions that gave rise to prisons. But beyond this, he targeted a system of confinement that permeated social reality, in schools as well as in the factory, and in the army barracks. This new space of visibility was born at the end of the eighteenth century and became universal through concrete relationships. But Foucault never attributed it to a decision maker or to any particular causal system.
The practice of confinement appeared necessary and only later found justification. It was somewhere between an order of a specific discourse and an averting of the eyes, another mode of visibility. Max Weber had already suggested that modern society was based on the subject's self-discipline. Foucault tracked down its conditions in the multiplication and extension of the powers of normalization that, as they were multiplied and extended, touched the individual in all social spheres. From a juridical-discursive system where rules and the law were pronounced by a uniform power, another society emerged based on discipline and disciplinary norms. In an absolutist society, a crime was an act against the sovereign's person. The criminal's body was therefore punished in order to repair the momentarily weakened body of the Prince. Punishment was thus more political than judicial; the body was at the core of power. "The body, which is questioned as it is punished, constitutes the point at which the punishment is applied and the site from which truth is extracted."13 The condemned body was in fact the cornerstone of ceremonial public punishment. Execution was linked to the crime: blasphemers' tongues were pierced, the impure were burned, a murderer's hand was cut off. Justice repeated the crime and exorcised it through the offender's stunning punishment and death. The ceremony let the momentarily weakened or wounded ruler recover his sovereignty. "Punishment did not restore justice; it reactivated power."14
When royal sovereignty underwent a crisis, the right to punish changed. No longer the means of reactivating the figure of the Prince, it served as social protection. This new approach corresponded to the moment when illegality shifted from being a crime against the body to a misappropriation of goods. A judiciary system came into being in which disciplinary power tended to make itself invisible while the social body, in order to come under scrutiny, had to become visible in the most minute detail. A disciplinary system was set in place, and prisons, schools, and barracks were built. "What took shape was ... a tighter penal quartering of the social body."15 An omnipresent power that was able to punish any infraction at any moment replaced an impotent power that demonstrated its will to power through the display of corporal punishment. "The right to punish slipped from being the ruler's revenge to being a defense of society." 16 Modernity implied the surveillance of populations by specific institutions conceived for their effectiveness. This was the era of the great enclosure, according to Foucault, and it initially touched marginals: vagabonds, beggars, and madmen. But it also concerned schoolchildren, when the model of the convent became the rule, and soldiers who went from vagabonding to becoming sedentary figures stationed in army barracks.
A whole social system shifted according to a new design of visibility. Jeremy Bentham provided the model of this new disciplinary society with his panopticon, which, in the years 1830 to 1840, became the model for prisons. "It has polyvalent applications; it helps to correct prisoners, but also to heal the sick, to instruct schoolchildren, to guard the mad, to keep an eye on the workers, and to make beggars and the lazy work. It is a way of setting bodies in space."l? Once this disciplinary society was set in place, things slipped from individualization toward the lower end of the social body. In medieval society, individualization was maximal at the summit, in the ruler's body. In a disciplinary society, on the other hand, visibility was supposed to disclose an entire population's activities. Individualization was in decline, with power becoming an anonymous, functional machine.
Foucault changed things in two ways. First, he no longer perceived power negatively, but positively ("In fact, power produces; it produces reality").18 Above all, he challenged the progressive vision of history that considered the Lumieres to be a major moment of freedom and emancipation, and that occurred with the advent of modernity. Behind this emancipation and the reign of freedom, Foucault saw a progressive control of bodies, the extension of disciplinary practices, and greater repression in a repressive society: "Historians of ideas generally consider that the philosophers and legal theoreticians of the eighteenth century dreamed of a perfect society, but there was also a military social dream." 19 Foucault thus invited his reader to consider a real reversal of historical perspective. His genealogy focused on the body, on how to approach it, the shifting views of it, and the modalities of its visibility. True to his description of the conditions that gave rise to the birth of the clinic, when he was above all a structuralist, Foucault, to his great credit, took on the challenge of the historical archive itself, of reforms, and of police archives, in his study of the logic of punishment. He therefore had a specific body of work to analyze without addressing the canonical texts of the history of philosophy. He synchronized his analytical sights with discourse and vision in order to better understand the real stakes of the organization and apparatus of power.
His work was phenomenally successful. More than The History ofMadness, which had had two distinctly successful moments, Discipline and Punish corresponded perfectly to the state of mind of a generation that wanted to "get the cop out of its head, "the petty chief,"
and that saw the manifestations of power everywhere-so much so that Foucault's ideas quickly evolved beyond even their author's wishes, and became a vulgate for those fighting different forms of social control. A veritable critical weapon against disciplinary practices, Foucault's theses became instruments for the various sectorial struggles and the many secondary fronts that were opening and closing. Never had a philosopher so echoed the ideals and discomforts of a generation, that of '68. Discipline and Punish also echoed the rising numbers of prison revolts and provided a theoretical framework for analyzing the underbelly of modern society. As jean-Michel Besnier and jean-Paul Thomas put it, "Drawing the lesson of '68 in the seventies meant renouncing the beautiful simplicity of the struggle against state power without having yet mourned the practices and analyses that were resolutely revolutionary.t'-? So it was hardly surprising that such a brilliantly written book meet with such commercial success: 8,000 copies were sold in 1975 and as many as 70,000 in 1987.21
Foucault, a Historian?
As a philosopher, Foucault made significant incursions onto the historian's turf. But he also dialogued with the corporation of historians and even embarked on some common projects with some of them, in particular with Michelle Perrot and Arlette Farge, whose preferred historical object was also persons who had been excluded from traditional history-women and marginals.
Since the days of his thesis on the history of madness, Foucault had encountered-albeit unintentionally-professional historians. Philippe Ariès was his improbable champion, given his right-wing ideology and ultraconservative royalism. But this independent historian, isolated from the history of mentalities, argued for Foucault's thesis at Plan in 1961. The work was enthusiastically received, especially by historians; Robert Mandrou and Fernand Braudel heralded the birth of a great historian. But from the outset, the relationship with historians was thwarted by a misunderstanding since the book was treated as a work of social psychology that magnificently illustrated the concept of the history of mentalities of the Annales. But The History of Madness was hardly a history of mentalities. Later, historians had the impression of losing one of their best, whereas Foucault's intention had not been to trespass on historical grounds as a specialist of social history, however renovated, but rather to problematize, as a Nietz-schean philosopher, what he considered to be the carnival of history. With his works of epistemology, a certain barrier of incomprehension arose between Foucault and the historians: "Foucault was sometimes bitter about this. He felt it was a rejection. Before he was elected to the College de France, he had hoped to be at the Ecoles des Hautes Etudes. 1 don't believe he offered his candidacy, but he did expect to be asked to do so. He was never asked to apply. "22
Michelle Perrot, on the other hand, appreciated his work. A student of Labrousse with an eye for the long series of history, Perrot was a great specialist in the history of nineteenth-century workers before becoming a feminist historian who was very open to interdisciplinar-ity. At Paris VII she eo-taught a course in the early seventies on the topic "History and Literature" with Gerard Delfau. Active in a feminist group in 1972-73, she taught another course the following year on the question of whether women have a history.t- She took the opportunity to invite a number of sociologists, including Madeleine Guil-bert and Evelyne Sullerot, to speak about the contemporary feminine condition. During these early days of women's history, the first question was to exhume a hidden reality, to do the history of those who had been forgotten, and to bring the repressed to light. We can understand why Perrot and Foucault would find common ground during a period when Foucault was working on giving voice to mute prisoners and she was doing the same for women. When Discipline andPunish came out, Michelle Perrot was interested quite precisely in the history of nineteenth-century prisons. "I thought this book was fantastic."24 Using "The Historian and the Philosopher," a text by the historian Jean Leonard that was very critical of the Foucauldian method, and Foucault's response, "Dust and Clouds," Michelle Perrot, together with Francois Ewald, organized a roundtable conference between historians and Foucault focusing on these two contradictory texts.
The historians, with the exception of jacques Revel, who knew Foucault's work very well, and Arlette Farge, who was working with him, asked questions that missed the mark on his thinking. He tried to answer, but there were two parallel discourses. And when Francois Ewald and I were listening to the tape of the session, we said that it was unpublishable as is.25
They chose to highlight Foucault's remarks by assembling the different historians' remarks as if they were made by an anonymous histo-rian in a dialogue following the first two texts. This all became the stuff of the 1980 Impossible Prison,26 "but the dialogue never really happened. "27
During the debate, Foucault described his approach, and made no effort to attenuate his fundamental differences from historians. His goal was not to undertake a total analysis of society. "My goal had been, from the outset, different from the historians' goal. ... My general theme was not society, but the True/False discourse. "28 He reiterated that he was working at describing events, but that his aim was not to write social history. His placed his grid on discursive practices, which is what Jean Leonard criticized him for, pointing out his abundant use of pronominal verbs and the personal pronoun "one." The issue was power, strategy, technique, tactics, "but do we know who the actors are: whose power, whose strategy?"29 Foucault abandoned the roles of different institutions in domesticating and conditioning bodies; different social categories were also left in the cloakroom. Jean Leonard criticized Foucault's kafkaesque universe: "The vocabulary of geometry turns human society into a desert; he speaks about spaces, lines, frameworks, segments, and dispositions. "30 But Foucault responded to the grilling by saying that these were not his concerns. The issue was neither studying French eighteenth- and nineteenth-century society, nor writing a history of prisons between 1760 and 1840, but rather writing "a chapter in the history of the logic of punishment.'^! Dialogue was impossible because Foucault only traversed a few sites of history as a philosopher, whose main purpose was to show that the total reality so dear to historians was a trap that needed to be demystified.
Foucault mourned history. Along with the entire structuralist generation, he asked how, in the very cradle of Western civilization, reason could have given birth to the monsters of Nazism and Stalinist totalitarianism. At the heart of his relationship to history was this trauma, which left him dissatisfied with obvious answers, and always interested in detecting the subjugation behind the proclamations of the Lumieres, the great enclosure behind liberty, the physical slavery behind equality, and the exclusion behind fraternity. This was a dark vision of history, and a radical criticism of modernity. Foucault's historical deconstruction nonetheless led some historians to pay particular attention to the conceptualization and problematization of their object:
For me, this was altogether important. He never stopped shaping my thinking. In Discipline and Punish, everything he said about discipline helped me to understand everything that normalization could become in industrial society, and better to see what was called the formation of the working class. What was important in what Foucault said was that discipline was not simply repression; it was also consent, the internalization of values.v
Foucault's taste for archives led him to prepare historical files to address how the body could be taken as an issue of power among the many intersecting discourses clammering for it. Legal and medical structures each claimed that they knew how to deal with madmen. Pierre Riviere, a criminal Foucault discovered in The Annals of Political Hygiene and Medicine (1836), was therefore at the crossroads between many discourses of various origins and functions. He offered a pretext for battling for a position of power, for a legitimation of their scientificity.
The Pierre Riviere story dated from 1836 and was the subject of a collective file by Foucault and his seminar participants in 1973 .33 In this file, Foucault showed the relationship between a personal history written by Pierre Riviere himself, a farmer of about twenty who had just killed his mother, his sister, and his brother, and the legal documents and three types of medical reports: that of the country doctor, that of a city doctor who ran an asylum, and that of the great names in psychiatry and legal medicine. These juxtapositions around a specific case showed how psychiatric concepts began to be used in penal law. The accused was caught at the center of various tactics facing off in a judicial setting.
Foucault's sensitivity to archives was already quite unusual for a philosopher, and led him to publish several works with historians. After I, Pierre Riuiere, he and Michelle Perrot came out with a presentation of Bentham's Panopticon.r' He and Arlette Farge worked on royal letters condemning prisoners to the Bastille: "My meeting with Foucault was improbable because we were not at all working along the same lines. It happened around the material itself, and around something we generally ignore-his sensitivity to archives. He was quite affected by the aesthetics of the document. "35 This fascination with archives reversed the relationship between the historian and the philosopher, since it was Arlette Farge who managed to convince Foucault that he should present these documents, whereas Foucault
wanted to publish the letters and let them speak for themselves. "The miracle was that he could be convinced of it, and then he asked me to work with him on these texts."36
Foucault had, in fact, discovered these letters much earlier, while writing The Birth of the Clinic. He had already thought at the time that he would do something with them and had a very strong affective relationship with this material. "He is the only person to have said to me that it was also possible to work with emotion. He allowed me to see that feelings were no longer feelings in the touchy-feely sense of the word, but an intellectual tool."37 This was how Foucault came to work with Arlette Farge, a disciple of Mandrou, for two years. She was a historian of mentalities who had only discovered his work in 1975, when Discipline and Punish came out, and it influenced her decision to study the phenomena of deviation and marginality. "At the time, we were saying that we were going to let the oppressed speak."38 This perspective drew her to Foucault's interests and commitments, which, therefore, made this meeting less improbable than Farge suggested. She was also seduced by a form of thinking that contested linearity, that preferred breaking points, that problematized discontinuities, and thus made it possible to counter a top-down view of popular culture: "What interested me considerably at that point was that only the question of how was being asked, without asking the question of why. In a very rustic way, that tied in with the way in which I had continued to work, which amounted to bringing to light the most minute operations in this flux we called social life. "39
Their fruitful encounter led to a 1982 joint publication entitled The Disorder ofFamilies,4o which showed that the symbol of royal judgment and of the most abhorrent absolutism, which could imprison absolutely anyone without trial, was in fact most often used to satisfy the private ends of fathers who wanted the king to help them resolve disastrous family situations. The hundreds of men and women who, on "the order of the king" were sent to prisons (Bicetre and Salpetriere among others) were essentially the victims of obscure private family affairs, and imprisoned by their own families. This gave Foucault another opportunity to problematize something that had appeared obvious: power relationships circulated further, and were far more complicated, than any simple instrumental relationship incarnated by the king.
According to Farge, Foucault was quite sensitive to what histori-ans thought of him, "more than sensitive, quite torn."41 Discipline and Punish in fact heralded a real breakthrough with historians, who confirmed what was in fact a rapprochement, since The Archaeology of Knowledge-that is, since 1969-with the Annales school. This rapprochement occurred thanks largely to Pierre Nora, at Gallimard, and Foucault was a full-fledged participant in the new El Dorado of historians.
The Annales historians were, after 1968, the big winners of the structuralist vogue of the sixties. They were able to put the ball in their court at a point when it was becoming necessary to reevaluate events and diachrony. The structural paradigm was on the wane, fragmenting and imploding, overrun from within by those who had promoted the idea of an open and slippery structure, whereas from the outside came increasingly deep questions. The structuralist adventure thus continued and changed, borrowing pathways of history.
The historians who until then had felt only negatively concerned by an excitement that sent them back to empiricism, having already slowed down the rhythm of duration, were henceforth to jump on the bandwagon with all the triumphalism of latecomers.
From History to Histories
Michel Foucault's work and his special relationship with Pierre Nora at Gallimard were the essential link by which structuralism would nourish historical research. The title alone of the collection Pierre Nora inaugurated in 1971-"The Library of Histories"-emphasized the epistemological inflection as well as historians' adoption of decon-struction. The title of the historical collection thrilled Foucault, who was responsible for the Library of the Social Sciences. "It would have been banal to have called it the Library of History. I said to myself that 'Library of Histories' fully corresponded to what I wanted to say, to the fragmentation."1
Henceforth, history was written in the plural and with a small "h." And the discipline abandoned the idea of a program that would synthesize in order to better redeploy itself toward the many objects available for its limitless study. This idea of plural histories fully corresponded to Foucault's definition of historical practice in the introduction to The Archaeology ofKnowledge. Nora knew the text well because Foucault had asked him to read the manuscript: "He made me reread the first chapter, asking me how I reacted as a historian, and telling me that I was going to rediscover my own positions."? Pierre Nora introduced the collection with a preface that was highly influenced by Foucault's philosophy. He took up the idea of monument, and congratulated himself on it: "We are experiencing History's splitting asunder. New questions, enriched by the proximate social sciences, and a new concern for the entire world rather than the narrow historical consciousness of Europe, have phenomenally amplified the
questions that historians ask of the past____History has changed its
methods, its way of cutting up time, and its objects." Many new objects and a broadened historical terrain were so many signs of a triumph of history. Nora recalls having had a good laugh with Foucault about this short manifesto on the splintering of history, especially upon learning that Braudel was furious when he read it.
Pierre Nora even wanted to preface his collection with a small, synthetic book-manifesto condensing and promoting the theoretical positions of the new history. He discussed this with Foucault, Francois Furet, and Emmanuel Le Roy Ladurie. "Together we tried to think about what was happening to history. My idea was to point out the problems that were becoming apparent."3 This initiative took on unexpected propositions. It happened at a time when Jacques Le Goff was joining Gallimard. Since Nora needed support, he slowly delegated this project to Le Goff, who became so involved in it that he transformed the idea of a small manifesto into three thick volumes in the Library of Histories collection, Making History,4 which he eo-directed with Pierre Nora; Nora finished the volumes more or less single-handedly because, once Le Goff was elected president of the Sixth Section of EPHE in 1972, he no longer wanted to be intimately linked with Gallimard.
This enormous summa, which appeared in 1974, was a charter for the new history. It was the moment for a counteroffensive, and the historians, after having borne their lot when the young sprouts of the new social sciences were stealing the show, now planned to appropriate the promising paths of these independent-minded thinkers, assimilating their methods in order to complete the renewal of a discipline that had to abandon its unity in order to broaden its field of experimentation as much as possible. Historians were here responding to a challenge raised by the social sciences in general and by second-generation structuralism: deconstruction. "The field that it [history] alone occupied as a system for explaining societies over time was invaded by the other sciences through poorly defined borders that threatened to absorb and dissolve it."> For the authors of this trilogy, history had to be saved by abandoning its claim to universalism and promoting what Foucault called a general history, the history of a space of dispersion.
This shattering implied calling into question the Hegelian edifice that underlay historical discourse and decentering its unifying principle-humanity, as the subject of history, whether as an individual or as a collective subject. This echoed the structuralist proclamation of the death of man and the insignificance of the subject. Historians, like linguists and anthropologists, could now promote a scientific discourse by marginalizing the least manageable variable for a quantitative history. This was how Emmanuel Le Roy Ladurie came to title the fourth part of his Historian's Territory, volume I: History without Men.6 Unlike the first generation of the Annales which imagined an exclusively human and anthropological history, Le Roy Ladurie began with a concrete historical study of the climate since the year 1000, arguing that "the historian is mutilated if he is only a specialist of hu-inanity,"? Decentering was altogether essential, beyond this particular study, and Le Roy Ladurie considered it a true Copernican revolution in historical science. The historian's viewpoint was enriched by this decentering, thanks to which his scientific vocation could be affirmed.
The prevailing positivism wanted, like Foucault's position, to grasp the how more than the why in a descriptive take on archives. Such proximity to Foucault's ideas did not mean, however, that he had won the historians' trust. "Foucault was passionate about history, and at the same time considered that historians were idiots for not asking themselves enough questions about what they were doing."8 An upheaval occurred some time later over Pierre Vilar's contribution to Making History, in which he violently attacked The Order ofThings:
In his important works, Foucault generalized a method that makes
its vices more visible than its virtues. At the outset, authoritarian
hypotheses. Then, the demonstration, and, on the points where there is some clarity, we discover that dates are mixed up, texts called upon, such enormous gaps in knowledge that one has to believe that they are deliberate, and many historical mistakes."
Pierre Vilar, recalling Althusser's remarks about Michelet and his "delirium," felt that, all things considered, if he had to choose between the two deliriums, he preferred Michelet's. His charge was clearly harsh, and Foucault's reaction was not long in coming.
Unsuspecting, I answered the phone and heard Foucault's icy tones. I had sent him Making History. He exploded, saying, "I thought we were on the same wavelength and the first thing you sign insults what I do; it is a declaration of war. In this case, I don't understand why you are my publisher." So I opened the book with trembling hands, and discovered this page that left me speechless, and that had escaped both of us, Le Goff and me.t?
Foucault demanded that the page be removed from the second edition and threatened to leave Gallimard if he was not heard. Pierre Nora went to see Pierre Vilar. "Nora came to find me and he was completely
distraught.... Foucault is a great writer, immensely talented, but I
deny that he is at all serious from the point of view of historical recon-struction."ll The story got even more complicated when Pierre Vilar was supposed to have Pierre Nora join the Hautes Etudes, and as the next issue was coming out, Vilar's wife was dying. Nora did not want to bother him, and the story ended there. The first version of the book went unchanged, inasmuch as the passage of time had calmed Fou-cault's anger.
But this argument showed difficulties between Foucault and the historical corporation, even though it had largely adopted his ideas. In the same deconstructive perspective, it was no longer a matter of assembling multiple objects of history in a rational whole. Defining the historical operation, Michel de Certeau observed that history was no longer central, as it had been in the nineteenth century, and "no longer had the totalizing function that amounted to taking up philosophy's role of saying what meaning was."12 In his introduction of the three volumes of Making History in Le Nouvel Observateur, Pierre Nora admitted a discontinuity between the history written at the time of Bloch, Febvre, and Braudel and that of the seventies. "It is this notion
of total history that, it seems to me, raises problems for us today....
We are living a fractured, eclectic history, which includes curiosities that we should not deny ourselves." 13 The pluralizing of heterogeneous temporalities supported by the serial approach to time consigned the idea of globality to a metaphysical past: "Time is no longer homogeneous and no longer has a global meaning." 14 History was not to mourn total history, according to jacques Revel, for whom the fragmentation of historical knowledge pointed to a new scientific space: "The goal is no longer a total history but the construction of totally articulated objects. "15
Constructing the historical empire required deconstructing historical practice. This was the period when the computer promised historians access to scientificity because it could count all the possible objects of economic, social, or cultural history: how much wheat was produced, the number of births, marriages, and deaths, the number of times the Virgin was invoked in wills, the number of robberies committed in a given place, and so on. It could trace curves, point to limits, and to points of change. "In the final analysis, ... the only scientific history is quantifiable history. "16
The same year that the Library of Histories began, Seuil published Paul Veyne's How to Write History. This work on historical discourse shared a deconstructive perspective, was deeply influenced by Fou-cault's thinking, and disparaged the illusion of models of consciousness and totality as metaphysical. In Aristotelian fashion, Veyne saw history as belonging to the sublunar world of disorder and chance, which meant that it could claim no nomethetic ambitions. History could only restore the how, the description of what happened, and not an explanation of the why. It knew no limits: "Everything is historical, but there are only partial histories."17 The historian could only be a positivist since his discipline belonged to idiography. Everything else existed only by false continuities and fallacious reconstitution: "History with a capital H does not exist-there are only 'histories
of___.'''18 His ideas were so close to Foucault's that when his book
came out in paperback in 1978, Veyne added a long note titled "Foucault Revolutionizes History." As a historian, Veyne showed how useful Foucault's method was: "Foucault is the consummate historian, the consummation of history. This philosopher is one of the
very great historians of our time....He is the first completely posi-
tivist historian." 19
Paul Veyne specialized in ancient history. He used the end of gladiatorial combat during the century of Christian emperors to argue
Cerisy Colloquium, 1969: i, Serge Doubrovsky; 2, Algirdas Julien Greimas; 3, Gilbert Lascault;
Noël
Ronat; 9, unidentified; 10, unidentified; tt, Genevieve Idt; 12, Madame Mettra, '3, Roland Barthes; 14, unidentified; t 5, Tzvetan Todorov;
against the explanations that claimed that power was humanized, as a result of Christianization. Instead, he took an approach that, as Foucault had advocated, looked at the practice of political power. Emperors had adopted another kind of power, which was becoming paternal and therefore incompatible with gladiatorial existence. Describing these practices shed light on the sources of explanation. "Foucault did not discover something new called 'practice,' which had been heretofore unknown. He took the effort to look at practices of people as they really were: he spoke about nothing other than what all historians speak about, namely, what people do. "20 Foucault's great contribution, according to Veyne, was to show that words trick us, that they make us believe that things are natural. He adopted Nietzsche's use of invariants in order to replace rationalism with genealogy. "It remains that as far as sexuality, Power, the State, madness, and many other things are concerned, there can be neither truth nor error since they do not exist. There is nothing true or false about the centaur's digestion and reproduction. "21 Veyne was drawn to Foucault's structuralism and his sensitivity to the autonomization of discourse, which revealed no reality and which was removed from the referent. This theoretical framework looked above all at relations, the very kernel of structural thinking: "Foucault's philosophy is not a philosophy of discourse, but of relationships. Because 'relationship' is the name for what has been designated as 'structure."'22 Veyne concluded his defense of Foucault's method by considering that it was pointless to wonder whether or not Foucault was a historian because, for him, history is a false natural object.
The Historians Take the Ball and Run
The new history exploded with a vengeance from 1968-69 on and took up where psychoanalytical and anthropological publications had left off. If historical work did not wait until then to be published, the public was enormous and avid from this point on. The 1968-69 figures for publications are edifying. Fayard began its "History without Limits" collection under the guidance of Francois Furet and Denis Richet. Flammarion started three new collections all at the same time: the "Scientific Library" under Fernand Braudel; a "science" collection, which published abridged theses-Pierre Goubert's on Beauvai-sis came out in 1968, and Jean Bouvier's on the Credit Lyonnais (a banking system) and Emmanuel Le Roy Ladurie's on Languedoc in 1969; and finally, "Questions of History," a collection directed by Marc Ferro that raised the issue of history unframed within any particular chronology but by questions relating to the present. At Albin Michel the great classics were reprinted in the "Evolution of Humanity" collection: Marc Bloch's Feudal Society and Lucien Febvre's The Problem of Nonbelief in the Sixteenth Century.23 A broad public could therefore read the works of the founding fathers of the Annales. Pion started a collection directed by Philippe Ariès and Robert Man-drou called "Civilizations and Mentalities." At Gallimard, Pierre Nora had begun his Library of Histories in 1971, which became one of the most important melting pots for new history writing. In 1974, the number of volumes on history was six times what it had been a decade earlier. The Annales writers were in the key positions, particularly with the lead trio of publishing houses-Gallimard, Seuil, and Flammarion-which orchestrated the school's success.
This taste for history in the seventies was in some sense a continuation of the interest in anthropology in the sixties. Discovering the Other was still a concern, but this was no longer others in other lands, but al-terity within Western civilization, in the depths of the past. The historical sensibilities of the period leaned toward cultural history, toward the study of mentalities. Events were eliminated in favor of the constant, permanent features of human life, the calendar of repeated, daily human acts whose pulsations were reduced to biological or familial signs of their existence: birth, baptism, marriage, and death. The most spectacular success by scholarly, anthropologized history was Emmanuel Le Roy Ladurie's Montaillou, a Village in Southern France,24 which came out in 1975 and had a print run of three hundred thousand copies, exceptional for a university historian. Moreover, numerous articles on the history of mentalities came out in the Annales during this period.25 This anthropologizing of historical discourse shifted the focus away from sociographical studies, from the basement to the attic, as it were, and guaranteed the success of works on sexuality (Jean-Louis Flandrin, Jean-Paul Aron), on death (Michel Vovelle, Philippe Aries, Pierre Chaunu), on the family (Jean-Louis Flandrin, Philippe Aries), and on fear (Jean Delumeau). This level of the mentalities tended to cover the whole social spectrum, which it adopted and organized around the notion of the permanence of human nature. These were the last vital signs of a structural paradigm that would henceforth undergo an inexorable decline, one that was every bit as spectacular as its success.