COMMUNICATION, CRITICISM, AND THE POSTMODERN CONSENSUS An Unfashionable Interpretation of Michel Foucault JAMES JOHNSON University of Rochester A critique is not a matter of saying that things are not right as they are. It is a matter of pointing out on what kinds of assumptions, what kinds of familiar, unchallenged, unconsidered modes of thought, the practices that we accept rest. . . . Criticism is a matter of flushing out that thought and trying to change it: to show that things are not as self-evident as one believed, to see that what is accepted as self-evident will no longer be accepted as such. Practicing criticism is a matter of making facile gestures difficult. -Michel Foucault (1988b, 154) This is an essay of criticism in the sense Foucault seems to intend. It targets the postmodern consensus among political theorists. This consensus consists of the view, common to both its admirers and detractors, that postmodern political thought is corrosively skeptical, that it relentlessly uncouples its critical pretensions from any constructive normative commitment. Jacques Derrida once made a comment which, by its very incongruity, highlights the postmodern consensus. He confessed to an interviewer: "Indeed, I cannot conceive of a radical critique which would not be ultimately motivated by some sort of affirmation, acknowledged or not."'This remark should seem highly discordant to anyone who has witnessed the persistent jousting between critics and defenders of postmodernism.2 The postmodern consensus takes as primitive precisely what Derrida, surprisingly but rightly in my estimation, deems inconceivable. AUTHOR'S NOTE: Thanks to Joshua Clinton, Jack Knight, Richard Lynch, Tom McCarthy, Kevin Olson, Andy Rutten, Tracy Strong, Stephen White, Juliet Williams, and an anonymous refereefor helpful comments on earlier versions of this essay. All remaining errors and infelicities are my own. POLlTICAL THEORY, Vol. 25 No. 4, August 1997 559-583 0 1997 Sage Publications, Inc. 560 POLITICAL THEORY I August 1997 I hope to lend substance to Derrida's intuition by reconsidering the work of Michel Foucault, specifically his "analytics of power," and the fashionable reception of that work. I show that Foucault does not confirm the expectations of theorists who subscribe to the postmodern consensus. This strategy entails some pretty obvious perils. First, Foucault's relation to "postmodernism" is not a simple one. He surely does not articulate all postmodern themes. Nor do other postmodern theorists share all of his preoccupations. It nevertheless seems hard to address postmodern political thought in a way that fails to accord him serious c~nsideration.~ I sympathize with those who complain about critics who indiscriminately and dismissively have "lumped" Foucault and others together as postmodernists. Yet, insofar as he confounds the expectations of those who subscribe to it, Foucault himself can perhaps help us see clear of the postmodern c o n s e n ~ u s . ~ Second, Foucault was not only a prolific but--due to his simultaneous commitment to "genealogical researches" and aversion to "globalizing" theory-an enigmatic ~ r i t e rI. make ~ no claim to have gotten him "right" in any final sense. Nor do I seek to impose unwarranted theoretical unity on his work.%ather, I seek to highlight themes in Foucault's work that are important and that are systematically neglected by his critics and defenders alike. Foucault's critics misconstrue his work. But they do so for reasons that his postmodern defenders will be reluctant to accept.' To see why, it is necessary to explore what Foucault means when he remarks, in an apparently offhand manner, that for analytical purposes it "is necessary . . . to distinguish power relations from relationships of communication which transmit information by means of a language, a system of signs, or any other symbolic medium" (Foucault 1983,217).Within his own analytics of power, Foucault in fact sustains precisely this distinction between power relations and relations of communication. Moreover, he tacitly affirms the normative significance of communicative relations in ways that illuminate otherwise perplexing aspects of his work. My aim in this essay is not merely to prompt political theorists to reassess Foucault, although I do hope to do that. I aim, in addition, to challenge the postmodern consensus by suggesting why it is facile-at least in Foucault's case and in his sense-and how it systematically diverts political theorists in unfruitful directions. This critical ambition, in turn, is animated by a further, affirmative goal. I hope to reiterate the crucial importance of communicative relations for a normative assessment of politics and to draw unlikely support from Foucault for that purpose. The remainder of the essay consists of four sections. In the next section, I sketch the postmodern consensus as it has sprung up around Foucault. I indicate why it may seem warranted by his own postures. I then identify one Johnson I FOUCAULT AND POSTMODERNISM 561 particular version of the consensus which is most clearly reflected in the complaint, regularly voiced by his critics, that the source of Foucault's skepticism is his unwillingness to properly accord normative force to communicative relations. I argue in the remaining sections of the essay that this reading of Foucault is seriously flawed.* In the second section, I return to two now-familiar themes in Foucault's work that political theorists regularly depict as affirmative anchors for his show how both Foucault's promise of a dialogical ethics critical e n t e r p r i ~ eI. ~ and his concept of resistance comport poorly with the orthodox interpretation of his analytics of power. Each presupposes a commitment to the normative force of communication of the sort which, according to the postmodern consensus, Foucault refuses to make. This conclusion poses a choice. Either hoist these potential affirmative anchors or reassess the orthodox interpretation of ~ o u c a u l t . ' ~ In the third section I pursue the second option. There, I examine Foucault's analytics of power in some detail. I show how, for analytical purposes, he differentiates communicative relations from relations of power and how, having done so, he accords normative significance to the former. In this way, I directly challenge the now-conventional interpretation of Foucault. The interpretation I advance locates, in a manner that the postmodern consensus forecloses, a source for the affirmative themes in his work. It consequently holds out the possibility of locating Foucault less problematically in the critical tradition with which he explicitly identifies. Further, and at a more basic level, it enables us to make new sense of Foucault's analytics of power. In the final section I briefly advance an interpretation ofFoucault's broader project that acknowledges the privileged status he grants to relations of communication. I then suggest how, by attending to relations of communication and the various ways that they can be disrupted, political theorists might move beyond the impasse that the postmodern consensus engenders. THE ORTHODOX FOUCAULT I heard someone talking about power the other day-it's in fashion. -Michel Foucault (1980, 207) There is an orthodox reading of Foucault, one shared by both his critics and his defenders. Nancy Fraser captures the general tone of the critics. She worries that "Foucault . . . adopts a concept of power that permits him no condemnation of any objectionable features of modern societies."" William 562 POLITICAL THEORY I August 1997 Connolly, writing from a position considerably more sympathetic to Foucault, reiterates Fraser's concern. He offers the following options: "We can be democrats or nihilists; we can criticize the present from the perspective of alternative ideals or join Foucault in repudiating every ideal imaginable today as the tyrannical extension of 'our participation in the present system.' " I 2 Here, then, Foucault's critics and defenders converge. They both worry that he subverts the very possibility of social and political criticism. On first reading, at least, it appears that this common assessment is warranted. Consider Foucault's remarks on power relations. "Power," he tells us, "is everywhere" (Foucault 1978, 93). Insofar as it functions in positive, pervasive, and insidious ways, "power produces . . . reality" (Foucault 1979, 194). I will return to Foucault's views on power relations in some detail below. At present it is important only to note the implications he draws from his position. "It seems to me," he explains, "that power is 'always already there,' that one is never 'outside' it, that there are no 'margins' for those who break with it to gambol in" (Foucault 1980, 142).13In these and other remarks on power, Foucault seemingly confirms the postmodern consensus. It is difficult, given the ubiquity that he attributes to power relations, to discern what affirmative resources remain available to him, or how these might inform his critical stance. This portrait, while fashionable, is too stark. Foucault himself intimates that this is so. In retrospect at least, he explicitly located his work in the lineage of critical thought running from Kant and Hegel to the Frankfurt School (Foucault 1988b, 95).14 He also offers further, more substantive, if still underdeveloped, indications in his fugitive remarks on ethics and on resistance. Ultimately, however, both Foucault's self-identification and his more substantive affirmations are informed by the normative status that he confers on relations of communication. This claim may sound entirely implausible. Foucault (1980, 114), after all, insists that he is concerned in his genealogies with "relations of power, not relations of meaning." Thus, while he is centrally concerned with the functioning of discourses,he does not analyze them in terms of communication. Discourses instead are themedium within which "power and knowledge arejoined together," and they thereby are implicated essentially in the ways that power relations are established and consolidated (Foucault 1978, 100; 1980,93). Several prominent political theorists draw highly critical conclusions from such remarks. Each, for his own ends, accuses Foucault specifically, and postmodern theorists more generally, of trying to expunge communicative relations, with their attendant normative import, from social and political thought. Most notoriously, Jiirgen Habermas charges that Foucault's genealogical approach systematically depreciates "categories of meaning, validity Johnson I FOUCAULT AND POSTMODERNISM 563 and value" and "deals with an object domain from which the theory of power has erased all traces of c o m m ~ n i c a t i o n . "As ~ ~a result, Habermas thinks that Foucault deprives himself of the resources needed to effectively criticize modern social and political arrangements. Similarly, Charles Taylor argues that, in their haste "to delegitimize horizons of significance," Foucault and other postmodern theorists ignore the necessarily "dialogical" bases of authenticity. Again, the upshot is that postmodern theorists, Foucault included, are disabled in the face of what Taylor sees as the persistent "malaises" of contemporary Western d e m o c r a ~ i e s . ~ ~ i n Richard a l l ~ , Rorty depicts "Foucault as a stoic, a dispassionate observer of the present social order, rather than its concerned critic." In his estimation, this aloof posture is symptomatic of the "lack of identification with any social context, any communication" that Foucault shares with other postmodern theorists.17 Foucault's remarks and the inferences that his critics draw from them are both misleading. But while Foucault's remarks are, for reasons that I sketch in the conclusion, intentionally misleading, the conclusions of his critics are improperly so. Foucault not only provides the theoretical resources necessary to elaborate critical principles, he does so precisely by according normative weight to relations of communication. In this way, Foucault himself helps us to see clear of the postmodern consensus. ANCHORING AFFIRMATION I would like to say something about the function of any diagnosis concerning the nature of the present. It does not consist in a simple characterization of what we are, but instead-by following lines of fragility in the present-in managing to grasp why and how that-which-is might no longer be that-which-is. In this sense, any description must always be made in accordance with these kinds of virtual fracture which open up the space of freedom understood as a space of concrete freedom i.e., of possible transformation. -Michel Foucault (1988b, 36) In his analysis of modern power, Foucault focuses on what he terms the "micro-physics of power" and, in particular, on the "effects of domination" that it induces. He claims that this power operates through "techniques of subjection and methods of exploitation" which, considered as a whole, constitute a "policy of coercions that act upon the body, a calculated manipulation of its element^."'^ Nancy Fraser suggests that this sort of language, this talk of domination, coercion, exploitation, and so on, may betray an unstated affirmative stance. She finally concedes, however, that Foucault's writings afford "no clues . . . as to what his alternative norms might be." And she 564 POLITICAL THEORY / August 1997 concludes that his attempt to sustain a critical stance on the basis of a totalizing analysis of power is "normatively c o n f ~ s e d . " 'This ~ judgment is too quick. In the next section, I show that Foucault's analytics of power-not just his language, but the conceptual structure of his analysis-in fact harbors a crucial, if understated, affirmative dimension. In this section, I first want to explore two themes in his work that potentially afford escape from the postmodern consensus. Political theorists commonly present the promise of a dialogical ethics and the notion of resistance to suggest how Foucault indeed does strike, however tentatively, an affirmative stance. On this view, dialogical ethics and resistance each occupy and extend the sort of "space for freedom" that could facilitate a "possible transformation" of extant practices and institutions. I argue that both themes comport poorly with conventional understandings of Foucault's writings on power precisely insofar as they exhibit the crucial value he places on relations of communication. Consequently, both the promise of a dialogical ethics and the concept of resistance prompt us to reexamine those conventional understandings. Dialogical Ethics In his later writings, Foucault directs his efforts to "writing a genealogy of ethics" (Foucault 1984a, 356). He defines ethics, somewhat idiosyncratically, in terms of the subject's relationship to the self. And he proposes a set of distinctions aimed at explicating the forms that this relationship takes in various historical circumstance^.^^ The genealogy of ethics, then, amounts to inquiry into how, under specific historical circumstances, "the subject constituted himself, in such and such a form, as a mad subject or as a normal subject, through a certain number of practices which were games of truth, applications of power, etc." (Foucault 1988a, 10). An obvious question arises concerning this e n d e a ~ o r .I,~ 'of course, grant that Foucault's genealogy of ethics was incomplete at the time of his death. But recall his remarks, cited earlier, regarding the ubiquity of power relations. If, as Foucault claims, his genealogical researches first disclosed "a historical ontology in relation to truth," and then a related "historical ontology in relation to a field of power," what possible standpoint remains from which he might identify "a historical ontology of ethics through which we constitute ourselves as moral agents"?22How, other than by persuasive definition, does Foucault locate any domain of liberty within which such agency might operate?23Some suggest that the space for ethical agency emerges from the necessarily "dialogical" form of Foucault's ethics. But this simply displaces Johnson / FOUCAULT AND POSTMODERNISM 565 the question. How do dialogical relations escape the pervasive, insidious reach of power?24From this perspective, dialogical ethics, it seems, tacitly "assigns a very important place to relations of communication" at the same time that Foucault pointedly criticizes others for so doing.25 Resistance Even in his earlier writings, Foucault repeatedly insists that his portrait of power is not so bleak as it might appear. He cautions us not to underestimate the possibilities for resistance. For not only are "there no relations of power without resistances," but "like power, resistance is multiple and can be integrated in global strategies" (Foucault 1980, 142).26Indeed, in momentous circumstances, "it is doubtless the strategic codification of these points of resistance that makes a revolution possible" (Foucault 1978, 96). Resistance trades upon a number of affirmative possibilities. Foucault locates these possibilities within a quite specific understanding of the relations that obtain between intellectuals and political movement^.^' As he explains: If one wants to look for a non-disciplinary form of power, or rather, to struggle against disciplines and disciplinary power, it is not towards the ancient right of sovereignty that one should turn, but towards the possibility of anewfom of right, one which must indeed be anti-disciplinarian, but at the same time liberated from the principle of sovereignty. (Foucault 1980, 108; emphasis added) The essential political problem for the intellectual is . . . that of ascertaining the possibil. ity of constituting a new politics of truth. (Foucault 1980, 133; emphasis added) Political analysis and criticism have in large measure still to be invented-so too have the strategies which will make it possible to modify the relations of force, to co-ordinate them in such a way that such a modification is possible and can be inscribed in reality. That is to say, the problem is . . . to imagine and to bring into being new schemas of politicization. (Foucault 1980, 190; emphasis added) Even a charitable reader might justifiably find these remarks puzzling. Foucault does not develop them.28In each case they appear, almost as an afterthought, at the very end of an extended lecture or interview. This, perhaps, reflects Foucault's sense that they are, if not completely at odds with his depictions of power relations, surely in considerable tension with them. It nevertheless is clear that Foucault sees contemporary social movements as the bearers of those new political forms that resistance might articulate. He holds out feminism, for instance, as an exemplar of what he calls "movements of affirmation" (Foucault 1988b, 114 f; 1980, 219-20). Femi- 566 POLITICAL THEORY I August 1997 nism, according to Foucault, resists extant definitions of female subjectivity in terms of sex. On this interpretation, feminism advances a "new politics of truth," one which resists accepted truths about women, and which, consequently, suggests "new schemas of politici~ation."~~ Here, too, questions arise.30How do "movements of affirmation" distance themselves from the insidious, ubiquitous, continuous, plural functionings of modern power? Does power not ultimately set the agenda? In short, how, given the way that he characterizes power relations, can Foucault also insist that resistance is not "inexorably frustrated through being a compatriot of power" (Foucault 1980, 142; 1978,96)? Finally, if Foucault can answer that question, what criteria might he adduce for determining whether the "new" forms, schemas, and so on that resistance potentially establishes are improvements in normative terms? The postmodern consensus obscures the way that Foucault responds to questions of this sort. He explains that various "forms of resistance against different forms of power" have in common the feature that "they attack everything which separates the individual, breaks his links with others, splits up community life, forces the individual back on himself and ties him to his own identity in aconstraining way" (Foucault 1983,211-2). Resistance, then, aims to defend symmetrical social relations against the corrosive effects of power. Yet, this assertion surely raises pressing questions. What sustains the "links with others" that resistance aims to defend? How is the "community life" that resistance seeks to preserve itself constituted? In his analytics of power, Foucault suggests that relations of communication play a very important role here. This claim, like my earlier claim that dialogical ethics tacitly ascribes a crucial role to relations of communication, requires that we reassess his views on power relations. AN "ANALYTICS OF POWER " WITH0 UT NORMATIVE COMMITMENTS? Silence itself-the things one declines to say, or is forbidden to name, the discretion that is required between different speakers-is less the absolute limit of discourse, the other side from which it is separated by a strict boundary, than an element that functions alongside the things said, with them and in relation to them w i h n over-a11 strategies. -Michel Foucault (1978, 27) Foucault traces the evolution and transformations of what he calls "pastoral" power back from the present, through the Christian tradition and Judaism, to the "ancient Oriental societies" of the Middle East. He contrasts Johnson 1 FOUCAULT AND POSTMODERNISM 567 it on various dimensions with the sort of "political" power characteristic of the classical and modern West. In short, "political" power is "at work within the state as a legal framework of unity." The role of "pastoral" power, by comparison, "is to constantly ensure, sustain and improve the lives of each and every" individual. Here, the "theme of keeping watch is important." This broad distinction, and Foucault's preoccupation with the latter, positive or creative, "modality of power" is no doubt familiar (Foucault 1988b, 67, 60, 62; 1983, 213-6).31Indeed, it is precisely this preoccupation, so frustrating to his critics and so enthralling to his defenders, that grounds the postmodern consensus. Foucault locates "the threshold of our modernity" at the juncture during the seventeenth and eighteenth centuries where pastoral power extended its reach and started to become rationalized into what he calls "bio-power" (Foucault 1978, 139-48; 1979, 224). He explains that bio-power coalesced around "two techniques of power." One is what he calls "the regulation of populations." It consists of "comprehensive measures, statistical assessments, and interventions aimed at the entire social body or at groups taken as a whole." The other is what he calls "the disciplines of the body." These function at the level of "micro-power" to oversee, normalize, and thereby individuate the subject (Foucault 1978, 145-6; 1980, 125, 160; 1983, 215). In this section, I examine Foucault's treatment of "regulative methods" and "disciplinary techniques" in turn. My aim is to demonstrate how, within his own analytics of power, Foucault identifies relations of communication independent of power and tacitly grants normative import to those relations.32 Regulative Methods This technique of power operates on what Foucault refers to as the "social body." It functions at the aggregate level, on what we might call epidemiological or demographic variables-"birth and death rates, life expectancy, fertility, state of health, frequency of illness, patterns of diet and habitation" (Foucault 1978, 25, 139). The particular effects of regulative methods appear most clearly in Foucault's discussion of emergent doctrines regarding the "police" functions of government (Foucault 1978,23-5; 1988b, 77-83). In the seventeenth and eighteenth centuries, he tells us, these functions formed an expansive "governmental technology." Aside "from the army, justice properly speaking, and direct taxes," in this period "the police apparently sees to everything" (Foucault 1988b, 80). Most conspicuously, according to Foucault, by the eighteenth century, police functions included the regulation of sex (Foucault 1978, 24-5).33 568 POLITICAL THEORY / August 1997 Foucault draws upon several rather obscure utopian texts to illustrate how this doctrine of police regulation emerged. He considers one, written by Turquet de Mayenne in the early seventeenth century, to be especially revealing. Foucault explains that, according to Turquet, among its crucial functions, the police must "foster working and trading relations between men, as well as aid and mutual help." He then remarks: There again, the word Turquet uses is important: the police must ensure "communication" among men, in the broad sense of the word. Otherwise men wouldn't be able to live; or their lives would be precarious, poverty stricken, and perpetually threatened. (Foucault 1988b. 79) Foucault obviously takes the word communication to be especially significant. He goes on to explain why: And here, we can make out what is, I think, an important idea. As a form of rational intervention wielding power over men, the role of police is to supply them with a little extra life; and by doing so, supply the state with a little extra strength. This is done by controlling 'communication,' i.e., the common activities of individuals (work, production, exchange, accommodation). (Foucault 1988b, 79) Foucault sees Turquet's text as emblematic of a doctrine with wide relevance. Even as, in the intervening centuries, the realm of proper police activity has narrowed, the regulative function that Turquet identifies persists in the ways that modern power relations impinge on relationships of other sorts within the social body. Foucault remarks that "what characterizes the power we are analyzing is that it brings into play relations between individuals (or between groups)" (Foucault 1983, 217). He insists that a proper analysis of power relations requires that we distinguish them analytically from social relationships of other sorts, even if the two "in fact always overlap one another" (Foucault 1983, 218). Thus, while empirically they "are interwoven with" or "combine with" relations of other sorts within the social body, power relations are independent in the sense that they "have nothing to do with" those other sorts of relation (Foucault 1980, 142; 1988b, 83). This implies, crucially, that the converse also holds; other sorts of relation within the social body are independent, in the same sense, of power relations. Indeed, power relations, on Foucault's account, presuppose the existence of these other sorts of relation. As he explains, power relations both "are the immediate effects of the divisions, inequalities, and disequilibriums which occur" in these other sorts of relation "and are the internal conditions of these differentiations" (Foucault 1978, 94; 1980, 142). In short, power relations ramify and prolif- Johnson /FOUCAULT AND POSTMODERNISM 569 erate by infiltrating and exploiting other sorts of relation that, in principle, exist independently of power within the social body.34 It is in this context that Foucault, as I noted at the outset, affirms the need to differentiate power relations, not just from other sorts of relation within the social body generally, but from relations of communication in particular (Foucault 1983,217-8).~'He does not, however, simply insist on distinguishing relations of power and of communication. He identifies relations of communication as an especially important medium for the functioning of regulative methods. And, as I now show, Foucault also attributes normative significance to communicative relations. Disciplinary Techniques What Foucault (1988b, 83) calls the "art of government" deploys regulative methods to bring various aggregate social relations within the purview of power relations. So, too, it relies on the disciplines to subvert extant relations of other sorts at the individual level. The disciplines, according to Foucault, function through insidious, continuous, ubiquitous mechanisms of surveillance and normalization. "Discipline 'makes' individuals: it is the specific technique of power that regards individuals both as objects and as instruments of power" (Foucault 1979, 170). It does so, as Foucault explains, not just by continuously disrupting reciprocal social relations in general but, in the paradigmatic case, by subverting symmetrical, reciprocal relations of communication. Foucault argues that "the development and generalization of disciplinary mechanisms" during the eighteenth century "constituted the other, dark side" of those enlightenment processes that generated liberal democratic politics. In this sense, Foucault insists that the disciplines provided the necessary, unavoidable "foundation" for the advent of political "liberties" during that century (Foucault 1979, 222).36Consider how Foucault depicts disciplinary mechanisms, first in historical and then in theoretical terms. At the level of historical description, Foucault observes that the disciplines did not emerge smoothly. He explains that during this period, disciplinary power "had to solve a number of problems" that, on his view, other forms of power were ill equipped to addres3' In particular, as it emerges, discipline has to master all the forces that are formed from the very constitution of an organized multiplicity; it must neutralize the effects of counter-power that spring from them and which form a resistance to the power that wishes to dominate it: agitations, revolts, spontaneous organizations, coalitions-anything that may establish horizontal conjunc- 570 POLITICAL THEORY / August 1997 tions. Hence the fact that the disciplines use procedures of partitioning and verticality, that they introduce, between the different elements at the same level, as solid separations as possible, that they define compact hierarchal networks, in short that they oppose to the intrinsic, adverse force of multiplicity the technique of a continuous, individualizing pyramid. (Foucault 1979,219-20) Here, Foucault attributes the historical ascendance, the "political takeoff," of disciplinary power to its prodigious success, relative to extant juridical or political forms of power, at disrupting the sorts of unrehearsed horizontal relations that nourish r e s i ~ t a n c e It . ~ is ~ important to read this claim in the context of two points made earlier. Recall, first, that for Foucault (1979,95), "where there is power, there is resistance." Recall, as well, that resistance aims to establish and defend reciprocal, symmetrical social relations. From this an obvious implication follows. Insofar as disciplinary mechanisms always elicit resistances, the task of introducing asymmetries and subverting reciprocities must be notjust historical but ongoing and hence subject to more general theoretical analysis. Foucault quite famously claims that for theoretical purposes we primarily should attend neither to the locus of power nor to the aims of those who exercise it, but to its effects (Foucault 1980,97). Perhaps the most indispensable effect of disciplinary power is to produce "docile bodies." Foucault provides a lengthy, exceedingly fine-grained account of this phenomenon (Foucault 1979, 141-69). He recapitulates that account in the following terms: To sum up, it might be said that discipline creates out of the bodies it controls four types of individuality, or rather an individuality that is endowed with four characteristics: it is cellular (by the play of spatial distribution), it is organic (by the coding of activities), it is genetic (by the accumulation of time), it is combinatory (by the composition of forces). (Foucault 1979, 167) Of these four disciplinary techniques, two contribute to the task of "elaborating procedures for the individual and collective coercion of bodies" in ways that are especially relevant for present purpose^.^' I consider, in turn, Foucault's account of how "spatial distribution" and the "composition of forces" operate. Discipline generates docility, Foucault explains, "in the first instance" by distributing individuals spatially. It does this in at least four ways: through "enclosure," through "partitioning," by establishing "functional sites," and by specifying "rank" (Foucault 1979, 141-9). It is particularly important to notice how, in corollary processes, discipline partitions individuals and distributes them across functional sites. It accomplishes the former by dividing collectives into "cellular" elements. In a spatial distribution, Johnson 1 FOUCAULT AND POSTMODERNISM 571 Each individual has his own place; and each place its individual. Avoid distributions in groups; break up collective dispositions; analyze confused, massive or transient pluralities. . . . Its aim was to establish presences and absences, to know where and how to locate individuals, to set up useful communications, to interrupt others, to be able to at each moment to supervise the conduct of each individual, to assess it, to judge it, to calculate its qualities or merits. (Foucault 1979, 143; emphasis added) Once this task is accomplished, or perhaps in tandem with it, discipline demarcates architecturally functional sites in such a way that specific "places were defined to correspond not only to the need to supervise, to break dangerous communications, but to create a useful space" for, for instance, medical or military purposes (Foucault 1979, 143-4; emphasis added). Here, then, the way that discipline simultaneously disrupts and rearranges communicative relations emerges as crucially important to the task of producing docile bodies. Foucault returns to this theme in more graphic detail when he examines the ways that discipline applies "the composition of forces" to the same task (Foucault 1979, 162-7). He refers to this technique as "no doubt the highest form of disciplinary practice" (Foucault 1979, 167). It aims, by drawing on the effects of the other three disciplinary techniques, to maximize the functional effectiveness of those aggregate entities (e.g., military units, the labor force, etc.) that result when docile bodies are combined. Foucault explains: All the activity of the disciplined individual must be punctuated and sustained by injunctions whose efficacy rests on brevity and clarity; the order does not need to be explained or formulated: it must trigger off the required behavior and that is enough. From the master of the discipline to him who is subjected to it the relation is one of signalization: it is a question not of understanding the injunction but of perceiving the signal and reacting to it immediately, according to a more or less artificial prearranged code. Place the bodies in a little world of signals to each of which is attached a single, obligatory response. (Foucault 1979, 166; emphases added) In short, through the composition of forces, discipline "insidiously objectifies those on whom it is applied."40 It does so precisely by subverting reciprocal communication-that is, relations of communication that, as Foucault intimates here, revolve around explaining aims or formulating requests in such a way that others might understand them-and instituting in its stead a hierarchical, unequal, nonreciprocal, but exceedingly useful, "system of ~ommand."~' Foucault places his analysis of the specific mechanisms of disciplinary power into relief when, in a subsequent chapter, he discusses the disciplines in more general language. There, he depicts them in abstract theoretical terms as "those systems of micro-power that are essentially nonegalitarian and 572 POLITICAL THEORY I August 1997 asymmetrical." And he goes on to explain that disciplinary institutions-not just prisons, but schools, the military, factories, hospitals, mental health clinics-"have the precise role of introducing insuperable asjlmmetries and excluding reciprocities" (Foucault 1979, 222; emphases added). Jeremy Bentham's panopticon is the most telling example of the asymmetrical, hierarchical, nonreciprocal nature of disciplinary power. It is, on Foucault's account, the quintessential disciplinary i n ~ t i t u t i o nThere, . ~ ~ "Each individual, in his place, is securely confined to a cell from which he is seen from the front by the supervisor; but the side walls prevent him from coming into contact with his companions. He is seen but he does not see; he is the object of information, never a subject of communication" (Foucault 1979, 200; emphasis added).43 Disciplinary mechanisms, then, do not just render social relations less symmetrical and reciprocal but, crucially, they do so by simultaneously disrupting the communicative relations that nourish social and political agency and replacing them with patterns of thoroughgoing objectification. These passages are not terribly obscure. Indeed, they appear throughout the central chapters of a canonical text.44Yet, while familiar, they are difficult to reconcile with conventional interpretations of Foucault that, without hesitation, his critics and his defenders both accept. What Foucault seems to argue here-and what the postmodern consensus obscures-is that disciplinary power is normatively objectionable precisely because it imposes unequal, asymmetrical, nonreciprocal relations and because, in so doing, it obliterates the sorts of extant communicative relation that, potentially at least, could promote social relations characterized by equality, symmetry, and reciprocity. Here, we see clearly what Foucault leaves ambiguous in his treatment of regulative methods. He does not just distinguish power relations and relations of communication for analytical purposes. And he does not just describe how power relations infiltrate and exploit communicative relations. Within his analytics of power, Foucault portrays power relations as objectionable because they subvert relations of communication, relations of the sort that-if more fully specified-might sustain the vision of political agency that is implicit in his commitment to resistance or dialogical ethics. Those who subscribe to the postmodern consensus will no doubt object here. They might concede that, as Foucault portrays them, disciplinary mechanisms impose "insuperable asymmetries" by disrupting extant communicative relations and rearranging them into more useful forms. But they almost surely will balk at the conclusion that Foucault thereby portrays disciplinary power as norr~lativelyobjectionable. This reluctance is, I think, misguided. Indeed, the trajectory of Foucault's thought suggests that he found it impossible to evade such explicit normative judgment. Johnson I FOUCAULT AND POSTMODERNISM 573 Early on, Foucault sought to underscore the singular features of relations established by disciplinary power by contrasting them with contractual relations. He explains that "the way in which it is imposed, the mechanisms it brings into play, the nonreversible subordination of one group ofpeople by anothel; the surplus of power that is always fixed on the same side, the inequality of position of the different 'partners' in relation to the common regulation, all these distinguish the disciplinary link from the contractual link" (Foucault 1979,222-3; emphasis added). The language Foucault uses here is important, for it recurs when he subsequently clarifies his understanding of "power relations" by differentiating them from what he terms "states of domination" (Foucault 1988a).45He depicts power relations as "changeable, reversible and unstable" and claims that "they can modify themselves, they are not given once and for all." By contrast, in a state of domination, "an individual or group manages to block a field of relations of power, to render them impassive and invariable and to prevent all reversibility of movement."46 In light of this distinction it seems clear that, at least on reflection, Foucault identified those power relations induced by disciplinary mechanisms as states of d ~ m i n a t i o n . ~ ~ Foucault explicitly recognizes the normative implications of his stance. He concludes that the critical task of philosophy and of practical politics is to challenge, and hopefully minimize, processes by which power relations harden into states of domination (Foucault 1988a, 18, 20). Given the way Foucault characterizes disciplinary power-not only in its ideal or utopian panoptic manifestations but in the more mundane, concrete ways its mechanisms operate-this amounts to saying that the critical task of philosophy and of practical politics consists in large measure of challenging, and hopefully minimizing, those "insuperable asymmetries" that disciplinary power imposes as it disrupts extant communicative relations and rearranges them into more functional configurations. In short, Foucault charges us with the task of defending symmetrical, reciprocal relations of c o r n m ~ n i c a t i o n . ~ ~ COMMUNICATION AND CRITICISM As to the problem of fiction, it seems to me to be a very important one; I am well aware that I have never written anything but fictions. I do not mean to say, however, that truth is therefore absent. It seems to me that the possibility exists for fiction to function in truth, for a fictional discourse to induce effects of truth. -Michel Foucault (1980, 193) 574 POLITICAL THEORY / August 1997 Parties to the postmodern consensus, depending on their predilections, either champion or decry an approach to political theory that Derrida, as noted at the outset, deems inconceivable, and that Foucault, whatever the claims of his defenders or the complaints of critics, belies in his own practice. They thus misconstrue the range of options in contemporary political theory and, in so doing, help to sustain incessant, unproductive rounds of polemic surrounding "modernity," "postmodernity," and their v i c i ~ s i t u d e s .The ~~ irony, of course, is that Foucault himself considered polemic of this sort substantively barren and normatively objectionable (Foucault 1984b, 38 1-3). In this conclusion, I propose an interpretation of Foucault's larger theoretical enterprise that makes sense of the distinctive status he accords to relations of communication in his analytics of power. I then indicate how that interpretation suggests an alternative theoretical agenda that may prove at once more fruitful and less objectionable than the postmodern consensus. Foucault claimed that he had "never written anything but fictions." He also insisted that those fictions could contribute to the production of "truth." In light of these remarks, there are at least two ways to interpret the critical impulse that animates his analytics of power. One might argue that Foucault aims to disclose how the society we currently inhabit, in fact, is infiltrated through and through by "bio-power." On this view, his fictions are designed to dramatize this state of affairs.50The problem with this line of argument is that it overlooks the privileged role he attributes to relations of communication in his analytics of power. It consequently generates difficulties analogous to those that surround his treatment of resistance and dialogical ethics. In short, this line of argument partakes of the postmodern consensus and hinders any effort to see clear of it. By contrast, one might argue that Foucault's fictions are purposeful misrepresentations, that while they depict a world thoroughly permeated by biopower, he knowingly presents that world as c o ~ n t e r f a c t u a l .On ~ ~ this interpretation, Foucault adopts a rhetorical strategy of exaggeration aimed at establishing a critical perspective from which to assess modern social and political arrangements, one that might provoke the sort of resistance or induce the sort of ethical attitude that could, in practice, impede or perhaps actually foil intrusions of power in diverse areas of social relations. This interpretation allows us to characterize his enterprise, like he himself does in a late interview, as an attempt to formulate questions, to pose problems, in a manner "necessary to make the future formation of a 'we' possible" (Foucault 1984b, 384-5).52 And in this respect, it underscores the crucial practical task of fostering the sort of egalitarian, reciprocal relations of communication of which we remain capable and upon which we might draw to sustain vigilance against various encroachments of power.53 Johnson /FOUCAULT AND POSTMODERNISM 575 By illuminating how Foucault conjoins criticism and affirmation, this interpretation of his broader enterprise is a first, tentative step away from the postmodern consensus.54It also clearly is an insufficient one. For, spelled out in this way, his critical enterprise presupposes both that we can more fully specify the ways that communication works-how it coordinates social and political relations as well as how it nourishes social and political agency-and that we can identify the distorting factors to which communication is susceptible. Yet, Foucault's critical project is radically underspecified in this respect. If it is difficult to reconstruct his views on power with any great confidence, it is, I suspect, perhaps impossible to reconstruct how Foucault understands relations of communication. Indeed, Foucault leaves us very much in the position Dewey adopted when he observed that communication "is a wonder by the side of which transubstantiation pales."55 He surely fails to afford anything approaching an adequate means to adjudicate the rival claims of, for example, postmodern theorists who proclaim the "agonistic" character of all communicative relations and critical theorists, like Habermas, who envision such relations as a much more cooperative endeavor.56 The difficulty confronting contemporary political theorists is that even once we, as do both Foucault and his critics, accord normative significance to relations of symmetrical, reciprocal communication, considerable work remains if we hope to extricate such relations from the realm of miracles. This predicament is intensified by the common but, I think, mistaken presumption that analysis of communicative relations and their vicissitudes is a narrowly philosophical concern. Political theorists need to abandon this presumption. To justify a critical posture by reference to relations of equal, reciprocal communication, we must treat communication as something other than an enigma. Our task, in my estimation, requires that we forge systematic, discriminating connections between our philosophical pronouncements and the contributions of the social sciences. Here, I give one brief, admittedly underdeveloped example of how we might do so. Studies of the pragmatics of language use disclose that "wherever some convention or expectation about the use of language arises, there will also therewith arise the possibility of some non-conventional exploitation of that convention or e~pectation."~~ Political theorists who aspire to anchor their criticism of modern social and political arrangements in some vision of communicative relations need to trace the implications of this unavoidable dynamic. Exploitation in the relevant sense can be inadvertent and so carries no necessary pejorative connotations. Yet, since it also may be intentional, it opens the possibility that conventional expectations can be subverted or displaced for purposes of strategic advantage. 576 POLITICAL THEORY I August 1997 Consider, in this regard, the austere, counterfactual world that game theorists capture in their models.58This world is populated by strategic agents who interact under circumstances unconstrained by such things as social or political institutions. Game theorists explore this world in hopes of discovering unique equilibrium outcomes generated solely by the interaction of the agents who populate it. However, they actually demonstrate three very different, though relevant and quite remarkable, things. First, they show that strategic interaction in so stark an environment generates rampant indeterminacy in the sense that many games yield either no equilibrium outcomes or a multiplicity of such outcomes.59Second, they show that under a wide range of conditions, communication, even when uncoupled from the payoffs of relevant players, can, if not entirely eliminate, significantly constrain this indetermina~y.~'Finally, they demonstrate how strategic considerations, especially the extent to which their interests diverge, can prompt relevant players to use language in novel, surprising ways that undermine whatever salutary expectations communication might otherwise establish. Game theorists, in short, systematically show how communicative relations coordinate social and political interaction in ways that-at the same time-are unexpectedly robust and surprisingly v~lnerable.~' Here, at the poorly understood nexus of expectation and exploitation around which they pivot, communicative relations present social and political agents a chance to generate and then replicate power relations. Those agents need not seize this opportunity. Frequently, in pursuit of strategic advantage, they do.62 One of Foucault's virtues, on the reading I advance, is that he reminds us, quite forcefully, of this. NOTES 1. Keamey (1984, 118). 2. For purposes of my argument, it is irrelevant whether Demda intended this remark sincerely, as another playful incitement to current academic discourse, or otherwise. 3. In good postmodern fashion, Docherty (1993) confounds this assertion precisely by failing to accord Foucault serious consideration. See, however, Dews (1987), Megill (1985), and White (1991). 4. Here I have in mind Connolly (1993, 368). I borrow the felicitous term in quotations from him. 5. For his comments on this commitment and aversion, see Foucault (1980,81-7). 6. For warnings about the risks that such an attempt entails, see Connolly (1985) and Gutting (1994). So, for example, I have very little directly to say about Foucault's genealogical method per se or about his relation to other members of the critical tradition with which he explicitly identified. Johnson / FOUCAULT AND POSTMODERNISM 577 7. I have in mind here, for instance, such writers as Dumm (1988), who in an especially (to use a word he applies to others) "pungent" essay advances an aestheticized interpretation of Foucault, and Keenan (1987), who suggests that to fully appreciate the difficulty of Foucault's postmodem politics we must overlook the extent to which that difficulty is an artifact of his postmodem commitments. 8. Of necessity the argument of this essay is heavily exegetical. Throughout dus essay-but especially in the third section-I rely, as much as is possible, on Foucault's books and essays. In this way I hope to avoid the qualms of those who attribute some criticisms of Foucault to misunderstandings created at least in part by his "incautious statements in interviews" (Hoy 1986, 13). 9. I borrow the anchor metaphor from Shapiro (1988, 19-21), who notes the presence of "hermeneutical anchors" in Foucault's writings. 10. Postmodern theorists such as Connolly (1991, 60) might find the way that I pose t h ~ s choice "amusing and strange." But if I am right, there is perhaps more "coherence" to Foucault's normative project than such theorists-who celebrate the "code of paradox" they claim to find in his work-may wish to acknowledge. We might then ask what sort of "anxiety" prevents postmodem theorists from recogmzing that coherence? 11. Fraser (1989,33). For similar assessments, see Benhabib (1989,369-70), Taylor (1984, 152), Walzer (1986,64), White (1991, 18). and Wolin (1988, 186, 193-4). 12. Comolly (1987, 107). See also Connolly (1984, 15-6). Connolly, of course, goes on to wonder whether this opposition is overly austere. Indeed, while his subsequent efforts to defend a "Foucauldian" ethic seem to me to be a systematic effort to deflate it, his assessment of the theoretical impasse nonetheless captures the postmodem consensus. See Comolly (1991,1993). For similarviews, see Dreyfus and Rabinow (1983,206), Hiley (1984), Hoy (1986, 12-3), and Rouse (1994,99). 13. It seems to me that Comolly seeks to identify precisely the sort of "margins" whose existence Foucault denies. Connolly grounds his "discursive ethic of cultivation" in a "capacity for ethicality" that "exceeds the bounds" of discursively constituted "identity." And he grounds that capacity, in tum, in the universal exigencies of" 'life' . . . that which exceeds the organization of a discourse nonetheless unable to proceed without it." In dus sense, Connolly's Foucauldian educ is animated less by the dialogic process-the sort of agonistic communicative relations-in which it is embodied than by the sort of self-reflection that is prompted by the universal pressure to acknowledge "the tenacity of life and the contingency of death." What Connolly characterizes as a discursively embodied "agonistic respect for difference" follows from-in the sense of being made possible by-such reflection rather than providing the precondition for it (Connolly 1991, 166-8, 176, 171). I am grateful to Georges Borchardt, Inc. on behalf of Editions Gallimard for permission to quote from Discipline and Punish and The History of Sexualily (Vol. I), both by Michel Foucault. 14. McCarthy (1991,43-9) nicely draws out some commonalities and differences between Foucault and other representatives of this tradition. All quotes from Power Knowledge by Michel Foucault, edited by Colin Gordon Copyright (c) 1972, 1975, 1976, 1977 by Michel Foucault, collection Copyright (c) 1980 by Harvester Press, are reprinted by permission of Pantheon Books, a division of Random House. 15. Habermas (1987,286). 16. Taylor (1991,66-7). 17. Rorty (1991, 173-4). 18. Foucault (1979.26, 171, 138; emphases added). 19. Fraser (1989,29-31). Compare Taylor (1984, 1985a). 578 POLITICAL THEORY 1 August 1997 20. Foucault does not view individuals as literally inventing themselves but, rather, as defining themselves in terms of, or in opposition to, received social and cultural categories (Foucault 1988b, 11). Understanding this process, on his account, requires that we trace the connections and disjunctions between (i) "ethical substance," (ii) "the mode of subjection," (iii) "self-fonning activity," and (iv) "the telos" at work w i h n a historical ethical system. It is this framework that he deploys in his history of sexuality (Foucault 1984a, 352-5). 21. My argument here parallels that made by Bemstein (1992, 152-66). 22. Foucault (1984a, 351). 23. This, of course, is what Foucault tries to do in his late interviews. Even there, however, he insists that "in human relations, whatever they are . . . power is always present" (Foucault, 1988b, 11). And, as I note below, his efforts in the same interview to differentiate power and domination comport poorly with his earlier discussion of disciplinary mechanisms. Thus, "Foucault not only fails to explicate this sense of agency, his genealogical analyses seem effectively to undermine any talk of agency which is not a precipitate of powerhowledge regimes. Who or what is left to transgress historical limits?" (Bernstein 1992, 164). 24. In an otherwise insighrful cssay, Coles (1992) conveniently neglects to ask how what he calls "Foucault's dialogical artistic ethos" fits with what Foucault terms his "analytics of power." This is questionable insofar as, even in his later writings, Foucault continues to insist that "[tlhe analyses I have been trying to make have to do essentially with the relationships of power" (Foucault 1988% 3). The "discursive ethic of cultivation" that Comolly (1991, 167) depicts is susceptible to the same criticism insofar as it is predicated upon the possibility that subjects can self-consciously reflect on the universal exigencies of "life." 25. Foucault (1988% 18). 26. See also Foucault (1983, 211-2; 1988b, 84, 122-3). 27. His views on those relations are not important for present purposes. 28. Keenan (1987, 19-28) lucidly explicates how Foucault simultaneously criticizes and affirms "rights." Insofar as I understand the remainder of his essay, he seems ultimately to concede that Foucault creates a paradox for himself by simultaneously subverting and invoking the language of rights. Keenan simply asks that we appreciate how "difficult" are the postmodern politics that result from Foucault's argument. 29. This is a special instance of the more general "political, ethical, social, philosophical problem of our days" which is "to promote new forms of subjectivity" (Foucault 1983, 216). From this perspective, Foucault, it seems, sees his genealogy of e h c s and his treatment of resistances as related. The former, recall, is concerned with how, under particular historical circumstances, individuals constitute themselves as subjects. The latter, in turn, "question[s] the status of the individual" in the sense that it challenges "a form of power which makes individuals subjects" (Foucault 1983, 211-2). 30. For variations on these questions, see, for example, Dreyfus and Rabinow (1983,206-7), Fraser (1989,29), and Habermas (1987,283-4). 31. Weberman (1995) nicely presents these familiar themes. He stresses the ways that Foucault depicts power (i) as productive rather than repressive, (ii) as relational rather than an object that is possessed, (iii) as not subjectively exercised, and (iv) as ubiquitous. I will not rehearse these themes here. 32. To reiterate, interpreters uniformly overlook this point. See, for a very recent example, Rouse (1994). He defends Foucault against critics such as Fraser, Habermas, R o o , and Taylor by advancing a "dynamic" rendering of the powerhowledge relation. Yet, he pays literally no attention to the crucial role of inde~ndent,normatively binding- relations of communication in Foucault's analytics of power. In this, Rouse is not unique among recent writers. See, for instance, Haber (1994), Simons (1995), and Weberman (1995). Johnson / FOUCAULT AND POSTMODERNISM 579 33. More specifically, Foucault (1980, 125) explains that "sex is located at the point of intersection of the discipline of the body and the control of the population." So too, he notes that police regulation also relied upon disciplinary techniques (Foucault 1979, 213-5). 34. Here, of course, we encounter the notorious claim that power relations operate in a way that is simultaneously "intentional and nonsubjective" (Foucault 1978, 94). I will not explore this notion except to note a subsequent remark. "For let us not deceive ourselves; if we speak of the structures or the mechanisms of power, it is only insofar as we suppose that certain persons exercise power over others" (Foucault 1983,217). See also Foucault (1979,225; 1988b, 83). 35. If Turquet uses communication as an encompassing category, Foucault himself explicitly understands it less expansively and sets it off from relations of, for instance, sex, production, and exchange (e.g., Foucault 1983,217-8; 1988a, 11; 1988b, 83). 36. The term "foundation" is important here. Insofar as liberties stand in necessary relation to the sovereign state, Foucault leaves no doubt that the latter is, in turn, necessarily dependent on disciplinary power. "The state is superstructural in relation to a whole series of power networks." It "can only take hold and secure its footing where it is rooted in a whole series of multiple and indefinite power relations that supply the necessary basis for the great negative forms of power." In this sense, the disciplines provide "the conditions which make it possible for" sovereignty to operate (Foucault 1980, 122, 105-6, 187). 37. The most striking example of such problems is the sort of spontaneous solidarity that public puni~htnents(whether by execution on the scaffold or by condemnation to work on a chain gang) generated as an unintended and uncontrollable by-product among the populace (Foucault 1979, 61-3,68-9, 257 f). As I note below, on Foucault's account, disciplinary power is efficacious precisely insofar as it proves adept at obstructing the emergence of such solidarity. 38. Foucault (1979, 221). 39. Foucault (1979,169). Foucault himself recommends that political theorists attend to "the specificity of mechanisms of power" (Foucault 1980, 145). Yet, commentators by and large do not heed this advice. They proclaim the creation of "docile bodies" without attending to the specific mechanisms by which docility is generated. In short, political theorists continue to take what Foucault (1980, 98) would consider "too distant a view" of how power relations operate. See, for instance, Dreyfus and Rabinow (1983, 153-60), Rouse (1994, 94-6), Haber (1994), Simons (1995), and Weberman (1995). 40. Foucault (1979, 220). 41. Foucault depicts this as a system of "signals" arranged in such a way that they elicit a measured behavioral response as they are transmitted by superiors to the objectified occupants of subordinate positions in the disciplinary hierarchy. Among the examples of this phenomenon that he offers, the following is especially striking. "The training of school children was to be canied out the same way: few words, no explanation, a total silence interrupted only by signals--bells, clapping of hands, gestures, a mere glance from the teacher, or thatlittle wooden apparatus used by the Brothers of the Christian Schools; it was calledpar excellence the 'Signal' and it contained in its mechanical brevity both the technique of command and the morality of obedience" (Foucault 1979, 166). 42. "The exercise of discipline presupposes a mechanism that coerces by means of observation" (Foucault 1979, 170; emphasis added). 43. Compare the way Foucault summarizes the rules of quarantine propounded in the seventeenth century to control outbreaks of plague. An infected town is to be sealed off and divided internally. Inside each district, stringent "surveillance" is to be imposed. "It is a segmented, immobile, frozen space. Each individual is fixed in his place." Even food distribution is to be arranged in such a way as to allow "each person to receive his ration without communicating with the suppliers and other residents." Foucault claims that such measures, 580 POLITICAL THEORY / August 1997 taken in response to an "exceptional situation," emerge rationalized "at the distance of a century and a half," as the "general model of functioning" represented by the panopticon (Foucault 1979, 195,205,216; emphasis added). 44. These chapters are central in theoretical terms. They occupy the third and longest part of Discipline and Punish. In this sense, they follow the chapters of the first two parts of the book in which Foucault documents the general transition in modes of punishment and they lay the theoretical groundwork for the chapters of the final part in which Foucault documents the disciplinary nature of the prison. In these chapters of part three, Foucault addresses the "how" and the "why" of the transitions he has earlier documented and identifies and examines the disciplinary mechanisms that allow him to subsequently present the prison as the paradigmatic disciplinary institution (Foucault 1979, 131, 194,228, 249). 45. Foucault is at this point concerned with establishing a basis of freedom or liberty that can inform his notions of resistance and ethics. In the process, he makes a remarkable theoretical reversal. In the late 1970s he hypothesized that "power is coextensive with the social body; there are no spaces of primal liberty between the meshes of its network (Foucault 1980,142). When, in the mid-1980s, he endeavors to differentiate power and domination, however, he insists that "if there are relations of power throughout every social field it is because there is freedom everywhere" (Foucault 1988a, 12). 46. Foucault (1988a, 12,3). 47. See also Foucault (1980,95-6). The alternative would be to completely revise the portrait of disciplinary power he offers in Discipline and Punish. 48. Note that this conclusion implies nothing about the nature of such communicative relations. Specifically, it does not entail that such relations be "transparent" or that they be aimed at establishing "consent." So, for example, I am not prejudging the qualms that Foucault repeatedly expresses about Habermas's work (Foucault 1984c, 377-80; 1988a, 18-9). 49. For a recent but hardly unique example, see Villa (1992). See also the subsequent comment by Johnson (1994a) on the fruitlessness of polernic and the response by Villa (1994). 50. Megill (1985, 243-7.342-6) and Rouse (1994, 99) endorse this approach. 51. Hoy (1986, 13-4) advocates this sort of interpretation. Elsewhere, however, he seems to embrace the first interpretation sketched above (Hoy 1988). 52. Foucault claims to be "writing the history of the present" (Foucault 1979,31; 1988b, 88-9). He sees his writings as continuing the tradition of philosophical critique inaugurated by Kant that is preoccupied with deciphering the present and our place in it (Foucault 1984d; 1988b, 86-95). When they are framed like this, his fictions converge in quite specific ways with the tradition of Western political thought. In this tradition, political theories typically function to provide "a 'model' for understanding the period, but a very inadequate guide in any depth." They commonly are "both false and illuminating" in the sense that "such theories, false even though in large part they may be, are an indispensable means in enabling political agents to locate themselves within their contemporary political landscape" (MacIntyre 1983,33). 53. Foucault repeatedly claims that his intent is to alert us to the dangers lurking in our present situation and thereby to provoke a response (Foucault 1984a. 343; 1988b, 168). From this perspective, his genealogies constitute less an accurate historical portrait than a warning and, in reading them, "we must seek not a meaning, but a precaution" (Foucault 1979, 139). Yet, this interpretation of his enterprise has an important implication. "Foucault paints the picture of a totally normalized society, not because he believes our present society is one, but because he hopes we will find the picture threatening. He could hope for this effect on us only if we have not been completely normalized" (Hoy 1986, 14). 54. For example, this line of argument allows us to recognize the rhetorical aspects of Foucault's work (Connolly 1985, 368-9) and to concede that his rhetoric necessarily trades on established normative commitments (Bernstein 1992, 153-7). Johnson / FOUCAULT AND POSTMODERNISM 581 55. Dewey (1958, 166). 56. Neither position is articulated very persuasively. Foucault (1983,22) and Lyotard (1984, 10,15-7), for example, proclaim the agonistic character of linguistically mediated social relations. But they do little more than that. By contrast, Habermas has not established his thesis that human language is animated by a telos of understanding, that this telos is embodied in a system of criticizable validity claims, that these represent the pragmatic presuppositions of human interaction, and that, therefore, in social interaction communicative reason is prior to strategic rationality (Johnson 1991, 1993). For a sense of where the theoretical disagreements between these two views currently stand, see Kelly (1994) and Hoy and McCarthy (1994). Moreover, these two positions do not exhaust the available theoretical possibilities. The field grows increasingly less tractable once we entertain additional contenders such as the "expressivist" view of language that Taylor (1985b) defends or the neopragmatist account of language as a contingent achievement offered by Rorty (1989). 57. Levinson (1983,112-3). Indeed, the ubiquitous possibilities for exploitation have induced some to "conclude that there is no such thing as language" if, by this, we mean that "we communicate by appeal to conventions" (Davidson 1986,446). 58. Postmodem theorists, in particular, might consider my suggestion here rather preposterous. They should recall, however, that Lyotard (1984,60) quite explicitly classifies game theory as a "postmodern science" precisely insofar as it reveals and helps analyze indeterminacy in social and political life. I endorse a different interpretation of game theory (Johnson 1993, 1994b). And I dispute the supposition, which Lyotard seems to hold, that recognition of indeterminacy somehow entails the abandonment of "metanarratives" of justice, emancipation, and so on. Nevertheless, the results which I sketch here might, if they took them seriously, afford postmodem theorists a powerful analytical resource. In a remarkable, provocative essay, JeanPierre Dupuy . . (1989) traces a number of connections between the concerns of game theorists and those of postmodemism. Kreps (1990) is an excellent, accessible, brief survey of game theory. The idea that game theoretic models are counterfactual constructs is due to McCloskey (1987). 60. This finding is reinforced by the social psychological research that Orbell, Dawes, and van de Kragt (1988) report. 61. Crawford (1990) and Farrell and Rabin (1996) offer brief, accessible surveys of the research described in this paragraph. Farrell (1993), published after a prolonged career as a working paper, is a foundational contribution. 62. On this one might usefully compare Schelling (1960) and Foucault (1983,216-26). REFERENCES Benhabib, Seyla. 1989. On contemporary feminist theory. Dissent 36 (summer): 366-70. Bemstein, Richard. 1992. The new constellation. Cambridge, MA: MIT Press. Coles, Romand. 1992. Communicative action and dialogical ethics. Polity 25:71-94. Connolly, William. 1984. Legitimacy and modernity. In Legitimacy and the state, ed. W. Connolly. New York: New York University Press. . 1985. Taylor, Foucault and otherness. Political Theory 13:362-76. . 1987. Politics and ambiguity. Madison: University of Wisconsin Press. . 1991. Identity/difference. Ithaca, NY: Cornell University Press. 582 POLITICAL THEORY I August 1997 , 1993. The ethical sensibility of Michel Foucault. Political Theory 21:365-89 Crawford, Vincent. 1990. Explicit communication and bargaining outcomes. American Economic Review 80:213-9. Davidson, Donald. 1986. A nice derangement of epitaphs. In Truth and interpretation, ed. E. Lepore. New York: Blackwell. Dewey, John. 1958. Experience and nature. New York: Dover. Dews, Peter. 1987. Logics of disintegration. London: Verso. Docherty, Thomas, ed. 1993. Postmodernism: A reader. New York: Columbia. Dreyfus, Hubert, and Paul Rabinow. 1983. Michel Foucault, 2d ed. Chicago: University of Chicago Press. Dumm, Thomas. 1988. The politics of post-modem aesthetics. Political Theory 16:209-28. Dupuy, Jean-Pierre. 1989. Common knowledge, common sense. Theory and Decision 27:37-62. Farrell, Joseph. 1993. Meaning and credibility in cheap-talk games. Games and Economic Behavior 5:514-31. Farrell, Joseph, and Matthew Rabin. 1996. Cheap talk. Journal of Economic Perspectives 10:103-18. Foucault, Michel. 1978. The history of sexuality. Vol. 1. New York: Vintage. . 1979. Discipline andpunish. New York: Vintage. . 1980. Powerknowledge. New York: Pantheon. . 1983. Afterword: The subject and power. In Michel Foucault, 2d ed., ed. H. Dreyfus and P. Rabinow. Chicago: University of Chicago Press. . 1984a. On the genealogy of ethics. In The Foucault reader, ed. P. Rabinow. New York: Pantheon. , 1984b. Polemics, politics and problematizations. In The Foucault reader, ed. P. Rabinow. New York: Pantheon. . 1984c. Politics and ethics: An interview. In The Foucault reader, ed. P. Rabinow. New York: Pantheon. . 1984d. What is enlightenment? In The Foucault reader, ed. P. Rabinow. New York: Pantheon. . 1988a. The ethic of care for the self as a practice of freedom. In Thefinal Foucault, ed. J. Bemauer and D. Rasmussen. Cambridge, MA: MIT Press. . 1988b. Politics, philosophy, culture. New York: Routledge. Fraser, Nancy. 1989. Unruly practices. Minneapolis: University of Minnesota Press. Gutting, Gary. 1994. Michel Foucault: A user's manual. In The Cambridge companion to Foucault. Cambridge: Cambridge University Press. Haber, Honi Fem. 1994. Beyond postmodern politics. New York: Routledge. Habermas, Jiirgen. 1987. Thephilosophical discourse ufmodernity. Cambridge, MA: MIT Press. Hiley, David. 1984. Foucault and the analytics of power. Praxis International 4:192-207. Hoy, David Couzens. 1986. Introduction. In Foucault: A critical reader, ed. D. C. Hoy. Oxford: Basil Blackwell. . 1988. Foucault: Modem or postmodem? In After Foucault, ed. J. Arac. New Bmnswick, NJ: Rutgers University Press. Hoy, David Couzens, and Thomas McCarthy. 1994. Critical theory. Oxford: Blackwell. Johnson, James. 1991. Habermas on strategic and communicative action. Political Theory 19:181-201. . 1993. Is talk really cheap? American Political Science Review 87:74-86. . 1994a. Public sphere, postmodemism and polernic: Comment. American Political Science Review 88:427-30, 432-3. Johnson I FOUCAULT AND POSTMODERNISM 583 . 1994b. Review of James Bohman New philosophy of social science. Philosophy of the Social Sciences 24:385-90. Keamey, Richard. 1984. Dialogue with Jacques Denida. In Dialogues with contemporay continental thinkers, ed. R. Keamey. Manchester: Manchester University Press. Keenan, Tom. 1987. Reading Foucault on the bias: The "paradox" of knowledge and power, Political Theory 15:5-37. Kelly, Michael, ed. 1994. Critique andpower. Cambridge, MA: MIT Press. Kreps, David. 1990. Game theory and economic modelling. Oxford: Clarendon. Levinson, Stephen. 1983. Pragmatics. Cambridge: Cambridge University Press. Lyotard, Jean-Francois. 1984. Thepostmodern condition. Minneapolis: University of Minnesota Press. MacIntyre, Alasdair. 1983. The indispensability of political theory. In The nature of political theory, ed. D. Miller and L. Seidentop. Oxford: Clarendon. McCarthy, Thomas. 1991. Ideals and Illusions. Cambridge, MA: MIT Press. McCloskey, Donald. 1987. Counterfactuals. In The new Palgrave: A dictionary of economics, vol. 1, ed. M. Milgate et al. London: Macmillan. Megill, Allan. 1985. Prophets of extremity. Berkeley: University of California Press. Orbell, John, Robyn Dawes, and Alphons van de Kragt. 1988. Explaining discussion induced cooperation. Journal of Personality and Social Psychology 54:811-9. Rorty, Richard. 1989. Contingenq irony and solidarity. Cambridge: Cambridge University Press. . 1991. Essays on Heidegger and others. Cambridge: Cambridge University Press. Rouse, Joseph. 1994. Powerhowledge. In The Cambridge companion to Foucault. Cambridge: Cambridge University Press. Schelling, Thomas. 1960. The strategy of conflict. Cambridge, MA: Harvard University Press. Shapiro, Michael. 1988. Thepolitics of representation. Madison: University of Wisconsin Press. Simons, Jon. 1995. Foucault and the political. New York: Routledge. Taylor, Charles. 1984. Foucault on freedom and truth. Political Theory 12:152-83. . 1985a. Connolly, Foucault and truth. Political Theory 13:377-85. . 1985b. Human agency and language. Cambridge, MA: Cambridge University Press. , 1991. The ethics of authenticiry. Cambridge: H m a r d University Press. Villa, Dana. 1992. Postmodemism and the public sphere. American Political Science Review 86:7 12-21. . 1994. Public sphere, postmodemism and polemic: Response. American Political Science Review 88:430-2. Walzer, Michael. 1986. The politics of Michel Foucault. In Foucault: A critical reader, ed. D. C. Hoy. New York: Blackwell. Weberman, David. 1995. Foucault's reconception of power. The Philosophical Forum 26: 189-217. White, Stephen. 1991. Political theory andpostmodernism. Cambridge: Cambridge University Press. Wolin, Sheldon. 1988. On the theory and practice of power. In After Foucault, ed. J. Arac. New Bmnswick, NJ: Rutgers University Press. James Johnson teaches social and political theory at the University of Rochester: His published work hai appeared in Political Theory, Philosophy of the Social Sciences, American Political Science Review, and NOMOS. He currently is working on a book manuscript titled Symbol and Strategy: On the Politics of Possibility.