Foucault, Conventions, and New Historicism

This article appeared as a chapter in the following collection: Barker, Stephen, ed. Signs of Change: Premodern, Modern, Postmodern. Albany: State University of New York Press, 1996. Pp. 297-307. (pagination follows the original) p. 297 Foucault, Conventions, and New Historicism David F. Bell I shall begin with an anecdote concerning an incident that recently led me to a series of theoretical reflections on a topic I had always considered in a more or less imprecise manner, namely the notion of poaching. Not long ago I had the chance to observe an acquaintance's English mastiff. It was a rather impressive dog, as anyone who has seen a mastiff—especially for the first time—can doubtless attest. The mastiff is one of the largest dogs, and its sheer size and power can be quite overwhelming. As could be expected, I immediately inquired about its breeding origin, and my acquaintance obliged with some fascinating information. It was bred for a specific purpose: to control poachers on the property of English landowners. Consequently, traits that allowed training for a very specific type of behavior were highly prized. Those traits are quite undoglike, really. When it sees something of interest, instead of darting off to explore or chase the thing as the typical dog would do, the mastiff stops, sits, observes, and simply refuses to budge. A moment's reflection reveals this to be very useful conduct when the object is to control poachers, for the idea is most certainly not to alert them by sending a dog barking and chasing after them, thus driving them off the property to safety. On the contrary, it is much more useful and effective to immobilize them, so that they may subsequently be caught on the property, en flagrant délit. Detecting the presence of a poacher (defined simply as any strange person on the dog's property), a mastiff trained in the proper manner would stop moving about and remain still. When the poacher drew nearby, the mastiff would jump out, bring him or her to the ground (mastiffs weigh 150 pounds, generally), hold the victim down with a flat p. 298 snout pressed against the chest (the snout was also a breeding trait specifically created for that purpose), and wait until the master arrived. Only then would the poacher be allowed to stand up, caught on the property with no means of denying his or her presence there. I begin with these anecdotal remarks, because I would like to talk about poaching in the pages that follow, and the images conjured up by these brief observations on the breeding origin of the mastiff are significant, I think. Poaching is a dangerous, risky business that involves, in the first place, territories. Poachers make their living by venturing into terrains that do not belong to them and that have been laid out in a gridlike structure controlled by the gaze of a master, represented here by a delegate, the mastiff. Masters and owners can afford the luxury of waiting ponderously to see what may happen, because within their domains, organizations have been devised to eliminate adventure, that is, to exclude anything unexpected. Such organizations allow unforeseen events, should they occur, to be immediately discernable. The strength of the owner's strategic position is such that most of the time it will immobilize the interloper, freezing him or her in a gaze of recognition. Only the swiftest, nimblest poachers can cross into the forbidden space and make it back out without succumbing to what might seem at first to be an overwhelming force. The owner waits; poachers must venture out, move, feint, parry, thrust. They do not have the luxury of massing forces and applying them unilaterally and definitively. They can only look for an opening, dart forth quickly, and retreat before encountering the opposing force directly. The notion of poaching is one that Michel de Certeau used when analyzing Michel Foucault's achievements in Surveiller et punir.1 The following admiring and yet somewhat ironic characterization of Foucault's research "methods" mobilizes the notion at stake: "This kind of 'art' is easy to see at work. It is an art of telling: suspense, extraordinary quotations, ellipses of quantitative series, metonymical samples.… It also is an art of seizing the opportunity and of making a hit.… His reading is a poaching. Hunting through the forests of history and through our present plains, Foucault traps strange things which he discovers in a past literature.…"2 De Certeau's point is that Foucault does something that is no longer recognizable as traditional history in Surveiller et punir. In a playful, almost ironic manner, he engages in an activity that seems to have the trappings of historical research, but which turns the archive to another use. Foucault raids the archive, but he is not interested in the patient, ponderous treatment of archival materials that is characteristic of a certain type of historical analysis since the Annales movement. In other words, Foucault wants to steal from the archive without being immobilized by a kind of classificatory activity requiring an investment p. 299 that would transform him into a prisoner of the archive, into just another defender among many others of a marked out, well delineated territory. He does not want to become hostage to a kind of historical writing that he mimics in the beginning of Surveiller et punir only to refuse it and to escape from it with his treasures in a flight to avoid capture. I am most certainly not arguing that Foucault is careless or inattentive in his use of archives. His voracious reading habits and tireless research are legendary. Moreover, he pays conscious homage to the Annales method in the incipit of Surveiller et punir, beginning his study of disciplinary structures in the eighteenth and nineteenth centuries by instituting what I would like to call an "archive effect." The plates that inaugurate the text in the French edition are characteristic of the tactic employed. Pretextual, figural, those plates are calculated to expose the reader to the archive even before the writing begins. The archive is there to be seen and flaunted, to serve as guarantor of the serious historical nature of the argument to follow.3 And this "archive effect" is carried over into the first pages of the text of Surveiller et punir, almost entirely devoted to lengthy quotations taken from two different documents that become monuments to the two extremes of a supposedly historical development, which will be the subject of the book: the descriptions of Damien's torture, on the one hand, and the schedule of a Parisian reformatory for minors, on the other. Not only is the archive brought into the pages of the book to be seen through photographic plates, its very text is reproduced verbatim. But this homage to the archive is also a ruse, a feint, an expedition into archival territory to plunder in the hope of avoiding capture by placating authorities that are its would be protectors. Foucault is clearly playing with the archive in at least two different ways. First, although it is necessary to make decisions about what constitutes the archival source for the project of writing Surveiller et punir, it turns out to be difficult, if not impossible, to justify precisely or completely any such choices. Thus the corpus upon which Foucault bases his study must always stand outside—or at least on the fringes of—thorough methodological rationalization. When the question of his source material is posed in the interview translated into English under the title "Prison Talk," he responds as follows: Establishing a corpus of source data does indeed pose a problem for my research, but this is undoubtedly a different problem from the one encountered in linguistics, for example. With linguistic or mythological investigations it is first necessary to take a certain corpus, define it and establish its criteria of constitution. In the much more fluid area that I am studying, the corpus is in a sense undefined: it will never be possible to constitute the ensemble of discourses on madness as a unity, even by restricting oneself to a given country or period. With … prisons there would be no sense in p. 300 limiting oneself to discourses about prisons; just as important are the discourses which arise within the prison, the decisions and regulations which are among its constitutive elements.…4 The parallel with structuralist methodologies so important for Foucault's intellectual formation is made explicit here. One must recall that Saussure's inaugural gesture in attempting to create a structural linguistics was to delineate what could be considered the domain of linguistic analysis and to reject as external to linguistics certain phenomena connected with language usage. "What is the integral and concrete object of linguistics?" asks Saussure at the beginning of the Cours de linguistique générale. "Other sciences operate on objects that are already given and that can then be considered from different points of view. Our domain is nothing like this."5 Saussure then defines the domain that will be the object of his considerations. The allusion to mythological investigations sends the reader directly to Lévi-Strauss, who faced a problem similar to Saussure's. One need only take as an example the famous article entitled "The Structural Study of Myth'' to see that one of Lévi-Strauss's fundamental issues in developing his approach to mythology is to define his corpus.6 The following is a revealing remark that demonstrates what assumptions concerning the idea of a corpus are at work as Lévi-Strauss formulates his argument: "Our method thus eliminates a problem which has, so far, been one of the main obstacles to the progress of mythological studies, namely, the quest for the true version, or the earlier one. On the contrary, we define the myth as consisting of all its versions; or to put it otherwise, a myth remains the same as long as it is felt as such" (216–17). Defining an object of study is reduced to a gathering operation here with no inkling that this activity may, in fact, become endless because it may ultimately be boundless. LéviStrauss seems to think in this passage that a consensus on what might constitute versions of the same myth will be fairly straightforwardly obtained. For Foucault's project on disciplinary technologies, such an inaugural gesture is apparently much more difficult, if not simply impossible. Foucault insists instead on the notion of fluidity. The archive for the prison book is undefined; its borders poorly marked. No territory can be ultimately delineated and become the exclusive property of the researcher working on these questions. The types of territories being studied here cannot be staked out with some definitive version of a permanent explanatory grid. Like a poacher, the researcher must meander among archives he cannot possess or master totally, where he plunders only to move on in a manner that is uncharted and destined to remain so. Foucault most certainly overestimates the possibility for a researcher in linguistics or mythology to constitute a corpus in a precise, if not to p. 301 say scientific, manner. Linguistics itself in the poststructuralist age has discovered how important certain elements rejected by earlier linguistic methodologies can be for understanding language—enunciative contexts, for example. And as for the ethnological treatment of myth, the gathering of versions of myths eventually took Lévi-Strauss on a meandering path of increasing complexity that exposed him to all the theoretical aporias of ethnological thought. But Foucault exaggerates the parallel drawn with linguistics and mythology studies in order to set off and emphasize the tactical necessities, the uncharted practices he pursues in his own work. Now, I do not mean to suggest that historians, from whom Foucault also wants to separate himself, do not recognize the impossibility of completeness and preciseness when it comes to archives. I would submit, however, that they do not revel in this impreciseness, in the tactical maneuvers required by the undulating, ever-changing nature of the sources they must use. They are much more inclined to hide such ad hoc decisions, to present them as empirical necessities, and to pursue as an ideal some moment of stasis when things stop moving, frozen by the historian's gaze. Foucault, on the other hand, elevates such tactics, imposed by the things he wishes to study, to the level of something like an art form. The second way in which Foucault plays with the archive is in the very methodology he applies once he fastens onto archival materials in search of discoveries. Let me quote one characteristic passage from the "Prison Talk" interview mentioned above to illustrate what I mean. When asked about how he discovered the text used for the dossier he published on Pierre Rivière, Foucault answers in a very interesting way.7 The form of the question is itself extremely suggestive. The interviewer asks: "First of all, how did you come upon this astonishing text?"8 One cannot help but call attention to the expression "come upon," denoting the surprise of a fortuitous encounter, the act of tripping over something that just happens to be in one's path. In addition, the adjective "astonishing" reinforces the impression of surprise and of the find, the hit. Foucault answers: ''By chance, while systematically working through penal reports by medicolegal and psychiatric experts published in professional journals of the nineteenth and twentieth centuries" (Power/Knowledge, 48). The juxtapositioning of chance and systematicity here is extraordinarily evocative, it seems to me, and is a theoretical blind spot for Foucault at the moment he utters this comment. For, I would claim, the two notions of chance and systematicity are radically different, and the very presence of chance in the process of archival work is precisely what drives the Foucaldian project forward. It is not the systematic laying out of the archival material that resulted in the Pierre Rivière book, but, rather, the surprise of the find that was elevated to the status of p. 302 representative moment. The importance of chance is revealed in Foucault's remark, only to be instantly covered over by a comment diverting attention toward a supposed systematicity of the research that went into the project. This kind of situation is regularly visible in Foucault's work of the period and is perfectly reminiscent of what takes place in the opening pages of Surveiller et punir, where the Damien document, for instance, is elevated theatrically (it is an extraordinarily theatrical document in itself) to the level of a paradigm that speaks for itself.9 It may seem that I have said little about convention up to this point, but the contrary is the case. What Foucault is doing in Surveiller et punir might well be described as an attempt to transform a convention by developing a different practice. The grounding in and systematic use of the archive characteristic of the Annales school was a convention faced by Foucault in very much the sense of the notion of convention developed by David K. Lewis. Lewis's analysis describes convention as a coordination equilibrium problem producing regularity in the behavior of members of a population when they are agents in a recurrent situation.10 What Lewis means, simply put, is that in any given instance of a recurrent situation, everyone conforms to a certain regularity, everyone expects everyone else to conform to it, and everyone prefers to conform to it on the condition that others do so. Conventions rely on notions of expected behavior and on the idea that it is more comfortable to conform to such behavior in decision-making situations. Otherwise situations requiring decisions would become inordinately complex and lead to aleatory and therefore unsatisfactory solutions. What I am claiming is that Foucault did not use the archive in Surveiller et punir in the regular and expected way, but he was clearly aware of that expected usage, clearly in touch with a set of conventions that had grown up in French historical analysis over the period of the preceding decades. He was manifestly not only navigating within a region where he could expect his audience to be aware of a certain anticipated behavior, but where he could also depend on that awareness to manipulate the expectations of his readers and surprise them without creating methodological chaos. This in itself would not be a particularly interesting observation. What makes it more intriguing is that while Foucault is attempting to change the analytic conventions of historians by developing a practice that is unconventional, or not yet conventional, the very object of his analysis is itself a series of practices that are displacing conventions, namely, the techniques of power that undercut the public discourse on penality that was supposed to found penal doctrine from the end of the Enlightenment into the nineteenth century. One could claim, I think, that Foucault is groping for a new practice of writing history in order to convey, precisely, the essence of a historical moment when a certain type p. 303 of practice was itself undermining a set of existing conventions concerning penality. What if, in other words, the techniques, practices, technologies that Foucault presents as the object of his theory were instead the very techniques that go into building that theory, that is, were its guiding subject? As de Certeau remarks with great insight, "The question no longer concerns the procedures organizing social surveillance and discipline, but the procedures producing Foucault's text itself. In fact, the microtechniques provide not only the content of the discourse but also the process of its construction."11 The operations of the techniques/practices that Foucault "discovers" in the material he is studying are, in fact, comparable to the very techniques/practices that he mobilizes as tools for his own research and his own attempts to found a discursive practice that skirts the mastery of established conventions. This raises some questions about Foucault's procedures that I shall not have the space to address in detail here, but that I would at least like to mention. In particular, one must call attention to the insistence in Surveiller et punir on the set of practices Foucault characterizes as "panoptical." Why the privilege granted to the panoptical? It could be argued that the panoptical procedures Foucault claims to unearth figure metaphorically and quite strikingly in the theoretician's, that is, Foucault's, all-seeing gaze. The penetrating look into the archive that results in the find, the hit, is ultimately comparable to the peering eye that imbues penal practice in the early nineteenth century and freezes human action in its sights. The fascination with the panoptical thus exemplifies a drift in Foucault's discursive practice from the position of the poacher back to that of the master. The foray into the archive results in a theory about the transparence and clarity sought by disciplinary techniques at the end of the eighteenth century and the beginning of the nineteenth, and, moreover, this clarity is reproduced in Foucault's own writing. The notion of clarity that these practices promote becomes, as de Certeau has argued, a rhetoric of clarity at the level of Foucault's own technique of argument.12 He attempts to create a kind of narrative that ultimately aims to cover its seems, to present itself as a convincing, seductive story, persuasive because it prevents the reader from grasping the fortuitous manner in which archival finds are elevated to paradigmatic status. Foucault ultimately attempts to erase the marks of the metonymical operation he has undertaken and thereby, I would claim, in the reformulation of this rhetoric of clarity, falls back on that most ancient of French conventions, the one most pervasively imparted to every French school student, namely, the notion of the clarity of French thought and French language, la clarté française. The preceding comments are not meant to disqualify Foucault's enterprise, merely to call attention to some of its underpinnings and, p. 304 especially, to reveal its negotiations with a certain series of conventions, which form its backdrop. In attempting to give discursive existence to practices that are by nature nondiscursive, Foucault has to address, at least implicitly—but I think he does it rather explicitly, too—the discursive conventions within which he writes. His manipulation of the "archive effect" demonstrates that he is aware of the problems he faces, and he sets out to invent a practice to deal with them, a practice that I have called poaching here, but that clearly encounters its limit in the conventional notion of clarté. What I would like to suggest in the final part of my argument is that this poaching, problematized by Foucault in extensive meditations in Surveiller et punir and in other interviews and publications dating from the same period of his writing, has, in turn, become a convention in its own right in the context of what has been called in recent years the new historicism. Take Stephen Greenblatt's Shakespearean Negotiations as a first example.13 After a vaguely Foucaldian introduction concerning power structures, the text is composed of a series of chapters all built on the same principle, namely the juxtaposition of certain archival finds with the Shakespearean text. Chapter 1: Thomas Herriot's text on the Virginia colony and Shakespeare's political plays. Chapter 2: an anecdote recounted by Montaigne concerning gender confusion in the late Renaissance and Shakespeare's comedies. Chapter 3: Samuel Harsnett's book on exorcism and Shakespeare's King Lear. Chapter 4: texts by Hugh Latimer and William Strachey on anxiety and Shakespeare's Measure for Measure and The Tempest. The repetitive nature of the technique employed gives it an air of conventional regularity that, in fact, belies its problematic character. But it is not just this repetitiveness that shows us we are dealing with a practice that is, in reality, no longer a practice, but a convention. It is also the lack of transcendental claims for the method. Greenblatt seems content to remain at the level of his textual juxtapositionings without any great desire to mobilize them into a more generalized questioning that might include, for example—even principally—the working through of what is involved in the implicit methods underlying the act of choosing the very archival materials he uses and inserts. What he ultimately does, it seems to me, is to lean on a practice problematized by Foucault as he was inventing it for Surveiller et punir, but without taking up in his (that is, in Greenblatt's) own way all the problems of tactics and maneuver within the conventions of history writing, which were always of implicit and explicit interest for Foucault. In other words, Greenblatt seems to me to be pretending that the juxtapositioning of archival materials that was at the heart of the Foucaldian project at a certain moment needs no further justification, since it is validated simply by Foucault's success and has thus definitively entered into the possibilities p. 305 of what one might call conventional critical writing. But this always depends, of course, on forgetting that for Foucault, juxtapositioning was a practice, a maneuver, a tactic in search of its own discursive theory. It would be necessary, of course, to nuance any comparison and to identify differences in approach between Greenblatt and Walter Benn Michaels, another of the forceful proponents of new historicism. At a first level, however, their practices are similar. The book in which Michaels sets forth his own method of cultural interpretation and exhibits the elements of his adherence to Greenblatt's approach is The Gold Standard and the Logic of Naturalism.14 Once again a structure of loosely joined chapters treating a variety of literary texts and cultural questions juxtaposed with little attempt to create any synthetic coherence is apparent. Michaels conducts his own raids on literary and cultural archives in a manner that inevitably recalls Foucault's pioneering work. This tactic, however, raises a series of questions related to those highlighted by Greenblatt's writing on Shakespeare. Brook Thomas formulates those questions suggestively: But the very dazzle of Michaels's readings creates problems for his argument as a whole. What is fascinating about them is their ability to bring together details from a variety of seemingly random social practices and to establish connections between them. As such they satisfy modernist aesthetic tastes that value a poet's ability to fuse diverse material. But placed together the essays call attention to a problem raised by Pound's Cantos. Relying on the modernist technique of montage, Pound hoped that the bits and pieces he juxtaposed would ultimately cohere. But they did not. Neither do the essays Michaels places next to one another under the same cover.15 One inevitably has the impression when reading Michaels that the tactic of juxtaposing is an end in itself, that it has become a substitute for thinking theoretically about the justifications for choices, about the context in which this procedure is devised and the theoretical turn it supposedly represents. The "dazzling" impression created by the readings suggests connections among disparate phenomena, but the ultimate synthesis that the connections imply but never fully state simply does not occur, and one is left again with bits and pieces whose presence together is never fully explained. Fredric Jameson's recent essay on immanent criticism in the collection entitled Postmodernism approaches some of these problems from a different perspective, one that brings out the parallels apparent between new historicist conventional practices and the technique of structural homologies at work in structuralist analysis.16 Jameson's exemplary new historicist text is Walter Benn Michaels's book. His diagnosis is one to p. 306 which I would adhere in large part and one formulation of which I would like to quote here: This is also an aesthetic (or a writing convention, or mode of Darstellung) for which a formal rule emerges governing something like a ban or taboo on theoretical discussion and on the taking of interpretive distance from the material, the drawing up of a provisional balance sheet, the summary of the "points" that have been made. Elegance here consists in constructing bridge passages between the various concrete analyses, transitions or modulations inventive enough to preclude the posing of theoretical or interpretive questions. Immanence, the suppression of distance, must be maintained during these crucial transitional moments in such a way as to keep the mind involved in detail and immediacy. Whence, in the most successful of such artifacts, that sense of breathlessness, of admiration for the brilliance of the performance, but yet bewilderment, at the conclusion of the essay, from which one seems to emerge with empty hands— without ideas and interpretations to carry away with us. (Postmodernism, 188) I would underline the centrality of two notions in this passage. First, that of convention: "an aesthetic (or a writing convention…)," says Jameson. The writing convention of new historicism is more than that, as Jameson suggests by equating it with an aesthetics, literally a mode of perception of the world. In other words, the convention itself expresses a fundamental theoretical take on the objects of study on which the so-called method focuses. The method turns out to be an "artifact," the second significant notion, one that seems to summarize the whole of Jameson's hesitations with respect to new historicism. An artifact is an object produced by analysis, that is, one that results from the application of analytic rules that can in some cases be sufficiently faulty to produce a false object. Here the problem seems to be that the rules are enlisted without sufficient reflection about how they function or what they mean. They are themselves artifacts capable only of producing further artifacts. As such they become aberrations destined to be swept aside as inaccurate creations once a more careful theoretical approach is devised, one that would reveal and attempt to account for the unthought practices of the previous analytic process. What I am arguing, then, is that in the space of hardly more than a decade, a very selfconscious and experimentally savvy attempt made by Michel Foucault to finesse, one might say, the conventions of historical writing in the French mode has turned into a reigning convention in its own right as it has been mobilized by the first "classics" of new historicism. I insist in closing that Foucault's attempt to outmaneuver the conventions was itself problematic, and I trust I have stressed this point sufficiently. Running away from one convention, he runs directly into p. 307 another one—perhaps the overriding convention of French culture, namely the notion of clarity. And that clarity is meant to be a persuasive, pervasive, and unquestioned effect of his argument. What is missing in new historicism, it seems to me, is a necessary reflection on the act of juxtapositioning that would confront its conventionality and attempt to deal with what its use signifies in the context of its American appropriation. Not that the convention in itself is to be condemned necessarily, but it clearly needs to be theoretically appropriated—or reappropriated. As Jameson has suggested, if we let it stand in the place of theory, then what we are giving up is the possibility of engaging in theory—a situation that, not surprisingly, is precisely the one Walter Benn Michaels has attempted at times to impose. One might say that the writing of Greenblatt and Michaels is a sort of poaching to the second degree. Not only do they navigate in the archive in a manner that recalls Foucault's poaching, but they construct their arguments in implicit reference to a style employed by Foucault. This confusion of territories is doubtless something Foucault would have appreciated. I close with that suggestion and with a growing sense that the gaze of some mastiff is alighting on me from somewhere in this territory upon which I have trespassed. Notes 1. Michel Foucault, Surveiller et punir; naissance de la prison (Paris: Gallimard, 1975). 2. Michel de Certeau, Heterologies; Discourse of the Other, translated by Brian Massumi (Minneapolis: University of Minnesota Press, 1986). 3. The choice made by Foucault's American editor to place the illustrations in the middle of the text simply misses the point. 4. Michel Foucault, Power/Knowledge: Selected Interviews and Other Writings 1972–1977, edited by Colin Gordon, translated by Colin Gordon, Leo Marshall, John Mepham, and Kate Soper (New York: Pantheon Books, 1980), 38. 5. Ferdinand de Saussure, Cours de linguistique générale, edited by Tullio de Mauro (Paris: Payot, 1975), 23. The translation is my own. 6. Claude Lévi-Strauss, "The Structural Study of Myth," in Structural Anthropology (New York: Basic Books, 1963), 206–31. 7. Moi, Pierre Rivière, ayant égorgé ma mère, ma soeur et mon frère…: un cas de parricide au XIXe siècle, presented by Michael Foucault (Paris: Gallimard/Julliard, 1973). 8. Foucault, Power/Knowledge, 48. 9. The art of making the archive speak is one that Jacques Rancière worked to develop during the same period, although his projects came to fruition slightly later. See La nuit des prolétaires (Paris: Fayard, 1981) and Le maître ignorant: cinq leçons sur l'émancipation intellectuelle (Paris: Fayard, 1987). 10. See David K. Lewis, Convention: A Philosophical Study (Oxford: Basil Blackwell, 1986). 11. Heterologies, 189–90. 12. Heterologies, 191–92. 13. Stephen Greenblatt, Shakespearean Negotiations: The Circulation of Social Energy in Renaissance England (Berkeley: University of California Press, 1988). 14. Walter Benn Michaels, The Gold Standard and the Logic of Naturalism: American Literature at the Turn of the Century (Berkeley: University of California Press, 1987). 15. Brook Thomas, The New Historicism and Other Old-Fashioned Topics (Princeton: Princeton University Press, 1991), 132. 16. Fredric Jameson, Postmodernism, or, The Cultural Logic of Late Capitalism (Durham: Duke University Press, 1991).