From Communal Difference to Communal Holism.
Todd May.
In, Reconsidering Difference: Nancy, Derrida, Levinas, and Deleuze, University Park, PA.: Pennsylvania State University, 1997. Pg. 21–75.

Why is it that we are so concerned with the idea of community these days? We are asking ourselves what constitutes a community, what it means to be in one, how the community or communities in which a person lives play a role in the constitution of that person's identity. These concerns are particularly evident in philosophy, and not only in the debate in political philosophy between communitarians of both the Left ( Charles Taylor) and Right ( Alisdair MacIntyre) and liberal theorists ( John Rawls). Philosophers of language, following either Habermas's or Wittgenstein's lead, have invoked the idea of community for explanations of linguistic use and meaning.1 Philosophers of mind have turned toward the community in seeking explanations of mental content.2 Since the appearance of Thomas Kuhn seminal Structure of Scientific Revolutions,3 philosophers of science have pondered the role of the scientific community in the formation of scientific theories.

The concern with community, however, is not limited to philosophy. In an age of increasing individual isolation, people feel the need to belong to some kind of community with which they can identify. Politicians of both the Right and Left have responded by promoting different visions of community. For the Right, the community is one of traditional values that may be imposed upon those unwilling-and in some cases even unable-to abide by them. For the Left, the community is one of obligation to those who find themselves in a less fortunate situation than others. For both, but for very different reasons, the ideology of rampant individualism that has characterized the traditional liberal approach to politics has shorn us of a resource that is central in sustaining not only our lives but our very identities.

On the Left, however, this concern to reintroduce the notion of community into political thinking has been tempered by the history of Communist experiments in the twentieth century. (It is puzzling that the Right has not been nearly as concerned with the results of its own communal experiments.) In the name of community and community obligation, individual freedoms and lives have been sacrificed without seeming concern and to no appreciable gain. The step from community to totalitarianism has seemed, in practice at least, to be all too short. And so the question arises, If the concept of community is central to any viable approach to understanding what makes people's lives worthwhile, then how shall we conceive community?

It is here that the recent thought of Jean-Luc Nancy becomes relevant. In his most extended reflection on the concept of community, The Inoperative Community,4 Nancy takes it as his task to reconceive community in a way that articulates its place in our lives without its lapsing into totalitarianism. He sees this articulation as one that finds its home on the political Left. "In order to speak of the site that we are dealing with, I might venture the following thought: 'left' means, at the very least, that the political, as such, is receptive to what is at stake in community."5 If the Left is to end its romance with totalitarianism, it must provide a conception of community that combines both the obligation proper to a fully communal spirit and the respect for others that has often been overridden in the name of obligation. Nancy sets it as his task to provide just such a conception.

I should note at the outset that the sense of the term "totalitarianism" as I use it here is often wider than-although related to-the sense of the term when it is used to categorize a type of state governance. Traditionally, "totalitarianism" means something like a government in which the state has a near monopoly not only on the means of violence but also on the means of communication, education, and expression. However, a product of this state totalitarianism-and here Nazism serves as the most striking example-is that people are forced to define themselves and their communities within narrowly defined parameters (e.g., Aryan or Christian or Muslim or Serb or some small combination of these or others). This self-definition within narrowly defined parameters can itself be called, and willed be called here, "totalitarianism."

Totalitarianism, then, in the sense that I am using it, refers to narrow constraints placed upon individual and social identity and behavior rather than just to a type of state. Having said this, I should also note that totalitarianism in that sense need not be a product of state totalitarianism. Although almost all state totalitarianisms, if at all successful, will foster totalitarianism in the sense I mean it, there can be totalitarianisms that are not state fostered. (Before the rise of the Christian Right in politics, the influence of certain forms of Christianity in the Bible Belt could be cited as an instance of nonstate totalitarianism.)

Returning to Nancy, then, among the questions motivating Nancy's approach to community, two emerge as central. At the outset, I want merely to cite them. Later, I discuss a tension between them that has not been fully resolved in Nancy's writings to date. The citing of that tension leads to two alternative readings of the status of Nancy's writings, neither of which is satisfactory. But for the moment, let me just pose the questions that preoccupy him: (1) What is it to be in a community? (2) How can we conceive community in a nontotalitarian manner?)

The first question is a constitutive one. It asks about being-in-community and what that is. I have posed the question that way rather than asking, for instance, What is it to be a member of a community? The latter question, unlike the former, seems to assume (or at least inclines one to assume) that before entering into a community, one is preconstituted, that one enters the community as somehow already constituted.6 Since Nancy denies precisely such preconstitution, the question must be posed in a more neutral manner. Thus the articulation I have given it.

The second question is normative. Rather than ask about how things are, it asks about how things should be. It is a question not about correct and incorrect conceptions of what it is like to be in a community, but about more and less valuable conceptions. In order to see the distinction at work here, the second question could be formulated like this: Regardless of what being in a community is really like, how might we conceive it in ways that avoid the problem of totalitarianism? Nancy does not formulate matters this way, and indeed does not distinguish between the two questions I have posed. Instead, he addresses them indifferently in the course of his discussion, to which I want now to turn.

For Nancy, the being of individuals is, above all, an exposure to what has traditionally been considered outside the provenance of individuality. "'To be exposed' means to be 'posed' in exteriority, according to an exteriority, having to do with an outside in the very intimacy of an inside."7 Rather than view individuals as self-enclosed beings, Cartesian style, to which is added the outside world, Nancy views individuals as always already constituted by what is outside of them. "[W]e are brought into the world, each and every one of us, according to a dimension of 'in-common' that is in no way 'added onto' the dimension of 'being-self,' but that is rather cooriginary and coextensive with it."8 Thus Nancy denies at the outset any conception of individuality that would lend itself to the traditional liberal view of a community as an interaction of preconstituted individuals. A view of that type would be mistaken both about individuals and communities, since not only are individuals not preconstituted, but communities are something other than the sum or the relations among individuals.9

Nancy offers something of a "proof" against the view of an individual selfenclosed, a proof that, by its language, seems directed mostly against Hegelian notions of the absolute as a self-enclosure. It is worth quoting in full.

An inconsequential atomism, individualism tends to forget that the atom is a world. This is why the question of community is so markedly absent from the metaphysics of the subject, that is to say, from the metaphysics of the absolute for-itself... A simple and redoubtable logic will always imply that within its very separation the absolutely separate encloses, if we can say this, more than what is simply separated. Which is to say that the separation itself must be enclosed, that the closure must not only close around a territory (while still remaining exposed, at its outer edge, to another territory, with which it thereby communicates), but also, in order to complete the absoluteness of its separation, around the enclosure itself. The absolute must be the absolute of its own absoluteness, or not be at all. In other words: to be absolutely alone, it is not enough that I be so; I must also be alone being alone-and this of course is contradictory. The logic of the absolute violates the absolute."10

Nancy is arguing here that exposure is a necessarily constitutive aspect of individuals (and also, as he notes, of states, art, science, history, and ideas) because the idea of an individual absolutely separated from others is selfcontradictory. It is self-contradictory because, in order for the separation to be an absolute separation-one that excludes exposure to the other-the closure that performs the separation would have to double itself as a closure, being not only a closure but a closure of a closure. A closure must close around something else; it cannot close around itself without losing its characteristic as a closure. Thus, if a closure must, in order to perform its role as closure, close around itself, it thereby loses its ability to be a closure, and the idea of it as a closure is self-contradictory.

The argument here might seem a bit elusive, but perhaps it will become clearer if we see it against a Hegelian background. One of the lessons of Hegel is that things often harbor within themselves their own opposites, so that in their unfolding they themselves become the opposite of (or at least in some way opposed to) what they were before. This, of course, is the movement of the Hegelian dialectic. Nancy is making a dialectical move here concerning the possibility of self-enclosure. Self-enclosure, in order to be true self-enclosure, would have to immure itself against all contact with the outside. But in order to do that, its own enclosure would itself have to be enclosed. That cannot be, however, because enclosures do not enclose themselves; they enclose that which is within their enclosure. Thus enclosures are, by their very nature, exposed to the outside. But if that which encloses me is exposed to the outside, then I cannot be completely separated from the outside. There is always commerce along the enclosure, and a wholly separate individuality is impossible.

If Nancy is right here, then not only is exposure a necessary aspect of individuality, but the denial that it is, is logically impossible. Although I believe there is much merit in the idea that individuals are exposed to communities (if in a way different from the one Nancy articulates), I am not convinced that the denial of that exposure is self-contradictory. Wrong, but not self-contradictory. In order for Nancy to make his case here, he would have to offer a convincing argument that the closure of individuality must enclose its own enclosure. He states that this is so, but does not give any reason to believe it. Moreover, I think it is at least conceptually possible (and, after all, it is conceptual possibility that is at issue here) to deny it. Hobbesian-inspired forms of social contract theory, for instance, create the scenario of self-enclosed individuals engaging in negotiations in order to realize the maximum of self-interest for each negotiating individual. If Nancy is right, such a scenario must be self-contradictory, because the enclosure of (communicating) individuals must have to enclose itself. But why must that be? Cannot it not be said that the individuals, while communicating, do so only for strategic reasons, and thus retain a strong form of "personal self-enclosure" while doing so? Now, one might argue that this scenario is impossible because in order to communicate one must, for instance, learn a language, and that cannot be done without exposure to others. While this is true, it does not reinforce Nancy's case, because the necessity of learning a language from others is a factual necessity, not a conceptual one. I can imagine beings, preformed in such a way that they are ready to speak, engaging in such communication for the sake of self-interest. If one argues here that that is a bit far-fetched, I would agree. The point is not to defend the idea of self-enclosure-in fact I attack it below-but to resist Nancy's argument that it is conceptually impossible, In resisting that argument, howver, I hope, by the implausible nature of my counterexamples, to have begun to motivate at a more concrete level the idea that there may be something to his positive view. Let us return to pick up the thread of that view.

To be an individual, then, is to be constituted by what is "outside" one as well as what is "inside" one. The terms "outside" and "inside" are misleading, however, since once it is recognized that individual being is always already exposure, then the outside is no longer outside of one, nor the inside inside of one. What is outside of the individual on more traditional views is actually constitutive of him or her, and what is inside is part of the outside that is constitutive of the "inside" of others. When Nancy speaks of the relation of sharing, a concept I return to below, he says that "the relation is not one between human beings, as we might speak of a relation established between two subjects constituted as subjects and as 'securing,' secondarily, this relation. In this relation, 'human beings' are not given-but it is relation alone that can give them 'humanity.'"11

Before deepening the analysis any further, it is worth calling attention to the influence of Heidegger's thought on Nancy here. Although others, most notably Derrida, have developed Heidegger's thought in other directions, Nancy has been particularly concerned to show how a Heideggerian view of personhood can be developed in such a way as to offer a viable conception of community and of the obligations that flow from that conception. Unlike many who have followed Heidegger, however, Nancy's borrowings are from themes that are continuous between the "early" and "late" Heidegger. From the latter, he uses the Derridean theme of presence and absence in a way we will come to see further on. But in his view of individuality as exposure, he relies on earlier Heideggerian writings. For instance, in Being and Time, Heidegger writes, "When Dasein directs itself towards something and grasps it, it does not somehow first get out of an inner sphere in which it has been proximally encapsulated, but its primary kind of Being is such that it is always 'outside' alongside entities which it encounters and which belong to a world already discovered."12 In Nancy's view, however, Heidegger did not fully understand the implications of his own insight, and kept regressing to a concept of a preconstituted individuality.

I will not undertake here the dense and meticulous explication that Heidegger's text would demand. I will be content to propose dryly this double hypothesis: in approaching more closely than we ever have the altered (crossed by the other) constitution of Being in its singularity, Heidegger (1) determined the essence of the Dasein outside of subjectivity (and a fortiori outside of inter-subjectivity) in a beingexposed or in a being-offered to others, of which philosophy (since Plato? despite Plato?) has been, despite everything, the denial, and (2) kept (despite himself?) the assignation of this Dasein in the apparent form of a distinct individuality, as much opposed as exposed to other individualities and thus irremediably kept in a sphere of autonomic, if not subjective, allure.13

Returning to the analysis, if individual being is constituted by what is "outside" one, then it is constituted as well by the community in which one exists. And indeed, for Nancy, the interesting aspect of exposure to the "outside" is the exposure to the community of others, rather than to nonhuman beings or inanimate objects. But if to be is to be already in community, this being-in-community should not be seen-and Nancy is most emphatic about this-as an immersion in some sort of common or communal substance. If I am exposed to others in my being, if who I am consists at least in good measure in my exposure to others, this exposure is not an exposure to a thing, an otherness, which would then come to constitute me. In seeing why this exposure is not exposure to a something, we will come to recognize two important themes in Nancy's view of community: how it is an attempt to avoid the totalitarian temptation I referred to above and how it resists all attempts to give it articulation.

If exposure were exposure to a common substance, if the community in which one is exposed were a thing that everybody had in common, then there would be a single characteristic, or at least a group of them, that everyone possessed. In order, then, to restore community where it is lost, one would only have to retrieve this characteristic or group of characteristics and impose them upon the individuals in the community. Or, if those characteristics were already instantiated in everyone, one would only have to force a recognition of them upon those who refuse to do so. Nancy is leery of the idea that these characteristics exist and particularly of the history of attempts to impose some recognition that a certain characteristic or group of characteristics is definitive of a community.

Nancy's argument against the idea that there are specific characteristics that form the common substance of a community concerns the relationship between the immanence of such characteristics and death. For Nancy, the idea of a common substance that forms the glue of the community is interchangeable with the idea that there is a primal bond between individuals in that community that binds them together in immanence. "Distinct from society (which is a simple association and division of forces and needs) and opposed to emprise (which dissolves community by submitting its peoples to its arms and to its glory), community is not only intimate communication between its members, but also its organic communion with its own essence."14 This is the Rousseauian picture of community-an intimacy or immanence that binds individuals into a common substance and that is lost with the development of modern society.

This picture of community as an organic communion cannot be realized except in death, for it is only in death that full immanence is attained. The reason for this is that individuals, while alive, can never completely close themselves off-either by themselves or as a community-from exposure to what is outside of them. Otherwise put, a community can never become a community of immanence. Recall here Nancy's argument that the logic of complete separation is self-contradictory. If this argument is sound-which I have argued it is not-then the separation from others required for complete immanence is impossible, and thus the idea of an organic community is also impossible. Only with death, when exposure to the outside ends, can immanence be achieved.

Now, one can raise the question here to Nancy, which I will do when turning from the expository to the critical, whether Nancy is justified in thinking of a common substance solely in terms of immanence. Is it not possible to consider a common substance in terms other than those of a bond that encloses the community in upon itself? If, for instance, the language of a community is part of its common substance, then the community is bound together by something that is itself exposed to the contingencies of history and contact with the outside. (Of course, if one thinks of language in a Derridean way-which Nancy surely does-then language itself cannot be conshy; considered a common substance. That is an issue that will have to wait until the next chapter.) In any case, I want to flag a concern here that I will linger over later. For Nancy, any project that has as its goal the formation of a community of immanence is a project of death. "Immanence, communal fusion, contains no other logic than that of the suicide of the community that is governed by it."15 And with this recognition we can begin to see Nancy's second suspicion regarding the idea of a common substance. Not only is the idea of a common substance illusory, it is also dangerous. It underwrites projects that are suicidal and detrimental to the community it is their seeming goal to construct. For Nancy, the suicidal project of forming an organic community is the character of all totalitarianisms. "[C]ommunity does not consist in the transcendence (nor in the transcendental) of a being supposedly immanent to community. It consists on the contrary in the immanence of a 'transcendence'-of finite existence as such, which is to say, of its 'exposition.'... By inverting the 'principle' stated a moment ago, we get totalitarianism."16 The exemplar of totalitarianism par excellence is Nazism. In a coauthored article, Nancy and his colleague Phillipe Lacoue-Labarthe argue that the logic of Nazism-and indeed of fascism generally-is that of trying to form an identity by means of the contradictory attempt to imitate, and at the same time distinguish oneself from, a model that mythically preexists one. For the Germans, this model was that of the Greeks. Thus the task for Nazism was to imitate the Greeks in such a way as to become themselves-not Greeks but Germans. The logic here is one of what Nancy and Lacoue-Labarthe call a "double-bind." "Why a logic of the double-bind? Because the appropriation of the means of identification must both take place, and not take place, through the imitation of the ancients, especially the Greeks."17 The Nazi resolution to this double-bind was to try to discover beneath the classicism of the Greeks a darker Greece, one of blood and sacrifice. Thus, Nazism created a myth-a fictioning of reality-that both was and was not Greek, so that it could become German. The problem is that this mimetism can only be complete, can only achieve immanence, in death. Thus the Nazis tried to exterminate everyone who was not immanent in the organic community they were fictioning. But since there is no immanence, such an extermination could be applied to everyone: Nazism was suicidal.18

To conceive community, then, in terms of a common substance is both misguided and dangerous: it fails to recognize that such a conception can be realized only in death, and thus it promotes death-including its own death-in trying to realize it. Inasmuch as community cannot be a matter of common substance, however, there is another implication for the attempt to articulate a conception of community, an implication that cannot be avoided: a community's nature cannot be straightforwardly signified. That is to say, one cannot say what the communal nature of the community is. To say what the communal nature of a community is, to signify it, would be to ascribe to it a common substance, which is exactly what Nancy seeks to avoid. Community, then, must lie either beneath, beyond, or in the interstices of signification, but it is not susceptible to a straightforward accounting. "This us of sense, which is sense, this sense that is 'our' being before all anthropology, before all humanism and all antihumanism, requires an ontology that is still to come, that does not mean that it will come, but perhaps that it is in itself, as thought, ordered in the dimension of a 'coming' or of an 'overcoming': that of our compearance [comparution], which is our presentation in the element of sense. This presentation does not itself have signification; it has only place [lieu], ceaselessly, traversing innumerable significations."19

The idea that community cannot be straightforwardly signified may strike readers as odd and self-defeating. After all, is Nancy's project not an addressing of community in order to offer an account of it? Is he then telling us that no account can be given, and that therefore he cannot realize his own aims? Actually, no; he is not saying that. The situation with Nancy regarding community is very close to that with Derrida regarding linguistic meaning. Although the treatment of Derrida's specific view will have to await the next chapter, it can be said here that for Derrida linguistic meaning is not something one can give a straightforward account of, but that does not preclude one from giving an account of why one cannot give a straightforward account of it, and it does not preclude one from denouncing attempts to give a straightforward account of it.

The situation with Nancy closely parallels the Derridean position. The fact that one cannot give a straightforward account of what community is does not prevent one either from giving an account of what is going on in community that resists straightforward signification or from criticizing attempts to give a straightforward signification of community. We have already had a glimpse of the latter in Nancy's treatment of Nazism. I want to turn momentarily, by way of discussing Nancy's concept of "sharing" (partage), to the former. What needs to be emphasized here, however, is that for Nancy the lack of a common substance that characterizes community precludes giving an account-in the traditional sense of an account as that which says what the accounted is, that which signifies the accounted-without precluding discussion of community and while still articulating important aspects of it.

This articulation will involve, as do Derrida's articulations of that which gives rise to linguistic meaning, terms that do not themselves have a straightforward linguistic meaning. Some of Derrida's terms are well-known: differance, trace, supplement. Nancy also employs such multivalenced terms, perhaps the most illustrative among them being that of "sharing." In English as in French, the idea of sharing (partage) can be read as indicating two opposed movements at the same time. First, to share is to divide something up among the participants in the sharing; it is an act of division. Second, and in an important way opposed to this, to share something is for the participants themselves to take part in that something which itself may remain undivided. In the first movement, that which is shared is divided among participants who themselves remain undivided. In the second movement, that which is shared remains undivided and the participants, as it were, divide themselves into it. Taken together, sharing indicates a movement in which division and undivision are in an economic relation, an unstable mutual engendering in which neither shared nor participant retains its boundaries. For Nancy, this idea of sharing indicates the relationship among the members of a community. In a community, individuals are both the participants and the shared of the sharing: they both share of themselves and share in the sharing of others. Nancy claims that Georges Bataille captured this idea of sharing when he stopped thinking of subjectivity as self-enclosed but rather as exposure to others. [ Bataille] gave up thinking the sharing of community and the sovereignty in the sharing or shared sovereignty, shared between Daseins, between singular existences that are not subjects and whose relation-the sharing itself-it not a communion, nor the appropriation of an object, nor a self-recognition, nor even a communication as this is understood to exist between subjects. But these singular beings are themselves constituted by sharing, they are distributed and placed, or rather spaced, by the sharing that makes them others: other for one another, and other, infinitely other for the Subject of their fusion, which is engulfed in the sharing.20 In another context, discussing hermeneutics and the hermeneutic circle, Nancy makes much the same point: "If we are in motion always already 'in the everyday understanding of being,' it is not that we have in an ordinary fashion-nor extraordinary!-the meaning of being, nor a meaning of being, nor the meaning in order to be. It is that we are, we exist, in multiplying voices, and that this sharing is what we are: we give it, we share it, we announce it. 'To be' already in the understanding of being is not to be already in the circulation, not in the circularity of meaning: it is 'to be,' and it is to be abandoned in this sharing, and to its difficult community."21 To be in a community-and to exist (at least as Dasein exists) is to be in a community-is a matter of exposure to others who are similarly exposed, a sharing of exposure in which the borders of the individual are neither clearly drawn nor completely effaced. If the borders were clearly drawn, there would be no exposure, and thus no community; if the borders were completely effaced, there would be a common substance in which all were immersed. Thus the sharing that constitutes a community involves an economy of limits and borders that cannot be fixed-and consequently signified-any more than the economy of differance can be brought to presence. "Community is the community of others, which does not mean that several individuals possess some common nature in spite of their differences, but rather that they partake only of their otherness... All the selves are related through their otherness, which means that they are not 'related'; in any case, not in any determinable sense of relationship. They are together, but togetherness is otherness."22

The term Nancy uses for this communal nature is "being-in-common."23 Being-in-common clearly does not mean having something in particular in common; rather, it means being exposed to the others in a relationship of sharing in which the limits of the individual exposed are neither stable nor destroyed. "Being-in-common means that being is nothing that we would have as common property, even though we are, or even though being is not common to us except in the mode of being shared."24 For Nancy, only such a conception of community can articulate what the "Left" needs without falling into the temptation of subordinating all individuality to real or perceived communal needs. Otherwise put, only this conception of community avoids the twin dangers of liberal individualism and (left- or right-wing) totalitarianism. It avoids the former danger by refusing to countenance the idea of an individual as preconstituted before his or her immersion in the community. (In fact, Nancy even avoids using the term "individual," preferring "singularity" [singularité] instead.) It avoids the latter danger by refusing to countenance the idea of a common substance by which the community is identified and its communal nature articulated.

If community resists articulation in terms of a common substance that can be signified or a closure that can circumscribe it, it resists as well and for much the same reason articulation in terms of what it produces or works on. "Community necessarily takes place in what [ Maurice] Blanchot has called 'unworking,' referring to that which, before or beyond work, withdraws from the work, and which, no longer having to do either with pro-duction or with completion, encounters interruption, fragmentation, suspension."25 This resistance, which Nancy derives (with acknowledgment) from Bataille as well as Blanchot, follows from what has already been said. For if a community is not to be defined by a common substance, surely it cannot be defined by those substances that are the objects made or worked on by the community. Moreover, the situation does not change even if one defines what is produced by the community as "justice" or "a common division of basic necessities." These, too, are common substances that Nancy is at pains to resist as threatening to lead us down the road to totalitarianism. Why? Because, as defining principles for a community's identity, they risk being imposed on everyone in a way that once again subordinates individuals to the perceived needs of the community. If community is to be thought, it must be thought at a level beneath or beyond or in the interstices of what can be signified or produced as a recognizable common identity. With this turn, Nancy diverges profoundly from the mainstream of the Marxist tradition, which defines community in terms of communal labor. I believe that Nancy intends this divergence. For many who have witnessed the abuses of "existing socialisms"-as well as those no longer existing-the question arises, Where did it all go wrong? For Nancy, the place it went wrong was in the thought that there could be a defining communal identity that would subsume all individualities within it. Forced collectivization, the abuses of basic human rights, the marginalization and often institutionalization of those deemed unworthy or threatening-all this occurred in the name of' the communal identity of productivity to satisfy the needs of all. Once productivity is deemed to be the defining characteristic of a community, the delicate balance of interpersonal relationships and recognition is already threatened. "[I]t was the very basis of the communist ideal that ended up appearing most problematic: namely, human beings defined as producers (one might even add: human beings defined at all), and fundamentally as the producers of their own essence in the form of their labor or their work."26 We can see here also why Nancy believes that the nature of community resists signification altogether. "We are the community of sense, and this community does not have signification... This means very precisely that it is the community of not having signification and because it does not have signification."27 Just as productivity or the possession of a common substance threatens to lead to totalitarianism, so does any signification that the community holds over itself as its defining identity. Communal significations are, in the end, significations of common substances that, in defining who "we" are, also define who "we" are not and cannot be on pain of marginalization, exclusion, or even extermination. It is in the nature of signification-at least the nonindexical signification with which Nancy is concerned-to create generalities out of specificities. Linguistic meaning cannot be bound to a time and place: a point Derrida has exploited in his deconstructions. Therefore, insofar as a community gives itself a signification-a linguistic account of itself-it runs the totalitarian risk that Nancy discovers in Nazism and the experiments in "existing socialism." Nancy's term "sharing," in being an economic concept that cannot be reduced to a single meaning, resists the signification that would make it a candidate for the totalitarian temptation he seeks to avoid in addressing communal "structure."28 In more recent writings, Nancy has also used the term "freedom" (liberté) to articulate this communal structure. By "freedom," Nancy does not mean either freedom from determinism or political freedom as it is traditionally understood. Rather, he means "ungrounded" or "lacking a determining principle." Thus, a free being is a pure gratuity, a gift that exists, "is there," without a reason for being there. "To be sure, here there is no longer even 'freedom,' as a defined substance. There is, so to speak, only the 'freely' or the 'generously' with which things in general are given and give themselves to be thought about."29 This freedom is, in turn, intimately connected to the being-in-common of a community. "The community shares freedom's excess. Because this excess consists in nothing other than the fact or gesture of measuring itself against nothing, against the nothing, the community's sharing is itself the common excessive measure of freedom. Thus, it has a common measure, but not in the sense of a given measure to which everything is referred: it is common in the sense that it is the excess of the sharing of existence."30 The link between freedom, thus understood, and community runs (unsurprising for a student of Derrida's) precisely through signifi-cation. Freedom is not a definable characteristic that people possess; instead, it is the inability to reduce people to a defining (set of) characteristic(s), to a signification. The community, then, is the sharing of that inability to be reduced to a signification, which is the nature (shall I put this word under erasure?) of freedom. In community, an individual's-or, in Nancy's terms, a "singularity's"-freedom is exposed to the freedom of other individuals, and in that mutual exposure exists the relationship that, resisting signification, forms the communal bond. This does not mean that freedom is then the common substance that defines the nature of the community. Freedom is instead the opposite of a common substance; it is precisely the resistance to the kinds of identities of which common substances are made.31 Before turning to a critical appraisal of Nancy's conception of community, I want to deal with an objection, mentioned above, that many may want to raise to him. For Nancy, community cannot be signified; the being of community-being-in-common-cannot be given a straightforward articulation. But has Nancy, in his entire treatment of community, not sought to give us nothing other than such an articulation? Is not Nancy's approach to community another in the series of offerings of communal identities that he sees as leading down the road to totalitarianism? I do not believe that it is. In order to understand why not, we must understand what Nancy means by signification. Although a full treatment of the Derridean view of linguistic meaning is discussed in the next chapter, I can here give enough of a sketch of that view to shed light on how Nancy would defend himself against this charge. For Nancy, signification is a matter of stable identities. It is bound up with the general philosophical project of foundationalism, of giving answers to philosophical questions that are either unsurpassable or indubitable. If, for instance, we can signify who we are as a community, then we can, at least in principle, answer the question, Who are we? in such a way that we have exhausted the phenomenon of our communal being, that we have captured it completely and therefore left nothing more to be said about it. Whatever else Nancy has done, he has surely not given an account of some sort of stable identity that constitutes the nature of community. If he is giving something like what we might want to call an "account," it is only partially an account of community-a community that itself admits of only partial accounting. In addition to the partial account of community, however, Nancy offers another account, an account of what it is about community that resists giving an account of it. That both the partial account of community and the account of why any such account is necessarily partial rely on a term-namely, "sharing"-that does not have a stable or determinable meaning is exactly to the point of what Nancy is saying. There is something that happens in the being of community, in being-in-common, that, while constitutive of both community and of individuals, resists articulation in a straightforward manner, articulation by means of a determinable linguistic meaning or an ascription of a stable identity. Therefore, it is false to say that Nancy is offering more of the same, since the status of his "account" is not that of wholly accounting for the phenomena but only partially that and partially accounting for a particular nonaccountability that inhabits the phenomena. Given this, there are at least two further challenges one may want to raise to Nancy here. First, one might query whether language has to be signifying in the sense of forming stable identities. For Nancy, as for Derrida, it does not and indeed cannot. People may use language in an attempt to signify in this way-and, as we will see, for Derrida this attempt is precisely what characterizes the philosophical project. However, language, resting as it does on an unsignifiable "ground" of differance, cannot do so. Signification (i.e., full signification), then, is not even one aspect of language; one would be most accurate in calling it one of language's false temptations. For Nancy, however, it is the temptation that is dangerous, because it is what motivates the totalitarian project.32 Thus, even if signification is an illusory temptation, we must avoid engaging in it, especially in our reflections upon community, because of the damage subscribing to this illusion leads to. The second challenge one might raise is in the connection between signification and foundationalism. Is signifying-again, in the sense of giving a stable identity-necessarily bound to foundationalism? Is giving an account (of something) that one can defend against all comers necessarily the same thing as giving an account that is unsurpassable, an account that one can be assured (presumably through some sort of transcendental guarantee) exhausts the phenomena it attempts to account for? Otherwise put, is the ascription of a common substance to a community necessarily a project of signification?33 This is a question I return to both in this chapter and in the next one, since it assumes a connection, between the giving of accounts and foundationalism, that is spurious. Before turning to that second challenge, however, we must first return to a question, posed at the outset of this discussion, regarding the ambiguity of the question Nancy is attempting to answer in his treatment of community. There I noted that Nancy attempts to answer two questions that he does not distinguish, the question of what it is to be in community and the question of how we can conceive a community in a nontotalitarian fashion. The first question is a constitutive one, the second a normative one. It is time now to confront this ambiguity, because on it depends the status of Nancy's own discourse and, consequently, the issue of how he is to be evaluated. Is Nancy attempting to answer the constitutive question, the normative one, or both at a stroke? In order to approach this issue, we need to see how separable these questions are. And in order to do that, we must recognize that a satisfactory answer to one question may not be a satisfactory answer to the other. Let us suppose, for instance, contra Nancy, that there were a constitutive answer to the question of community that involved a common substance. And let us further suppose that recognition of that common substance would indeed lead to some form of totalitarianism. Let us suppose, for instance, that the United States really is a Christian country and that what forms the essence of a community here is its adherence to Christian principles of the kind advocated by one of our current televangelical luminaries. Or, if that is too much to stomach, suppose that we were really Hobbesian agents that formed and stayed in communities in order to realize our own self-interests. Even if this were constitutive of community, discovering it to be so would not give us a satisfying answer to the normative question. In fact, what it would mean is that community possesses a totalitarian nature. A satisfying answer to the constitutive question, then, is not necessarily a satisfying answer to the normative one. The same thing happens if we go the other way around. It may be a normatively good idea to believe that what constitutes community is a sharing of the kind Nancy proposes. And this may be good to believe even if that is not what constitutes community. If community were really constituted by some sort of implicit social contract among Hobbesian agents, we would be wrong in thinking it to be constituted by sharing. But nevertheless, it still might be a normatively good thing to think so. Why? next one, since it assumes a connection, between the giving of accounts and foundationalism, that is spurious. Before turning to that second challenge, however, we must first return to a question, posed at the outset of this discussion, regarding the ambiguity of the question Nancy is attempting to answer in his treatment of community. There I noted that Nancy attempts to answer two questions that he does not distinguish, the question of what it is to be in community and the question of how we can conceive a community in a nontotalitarian fashion. The first question is a constitutive one, the second a normative one. It is time now to confront this ambiguity, because on it depends the status of Nancy's own discourse and, consequently, the issue of how he is to be evaluated. Is Nancy attempting to answer the constitutive question, the normative one, or both at a stroke? In order to approach this issue, we need to see how separable these questions are. And in order to do that, we must recognize that a satisfactory answer to one question may not be a satisfactory answer to the other. Let us suppose, for instance, contra Nancy, that there were a constitutive answer to the question of community that involved a common substance. And let us further suppose that recognition of that common substance would indeed lead to some form of totalitarianism. Let us suppose, for instance, that the United States really is a Christian country and that what forms the essence of a community here is its adherence to Christian principles of the kind advocated by one of our current televangelical luminaries. Or, if that is too much to stomach, suppose that we were really Hobbesian agents that formed and stayed in communities in order to realize our own self-interests. Even if this were constitutive of community, discovering it to be so would not give us a satisfying answer to the normative question. In fact, what it would mean is that community possesses a totalitarian nature. A satisfying answer to the constitutive question, then, is not necessarily a satisfying answer to the normative one. The same thing happens if we go the other way around. It may be a normatively good idea to believe that what constitutes community is a sharing of the kind Nancy proposes. And this may be good to believe even if that is not what constitutes community. If community were really constituted by some sort of implicit social contract among Hobbesian agents, we would be wrong in thinking it to be constituted by sharing. But nevertheless, it still might be a normatively good thing to think so. Why? Because it might dull some of the sharper edges of our nature if we possessed certain mistaken beliefs about who we were. Of course, if we really were Hobbesian, this dulling could only go so far; but normatively, that would be better than the cynicism that a recognition of our true natures would, in all likelihood, promote.

Thus the constitutive and normative questions are separable. They might have answers that diverge, and even diverge profoundly. The answer to the question of what it is to be in community is not necessarily the answer to the question of how to conceive community in a nontotalitarian fashion. This does not mean that the answers to the two questions cannot converge. They very well might. But it does mean that we must ask the two questions separately, and cannot presume that, having answered one, we have answered the other. Although it is clear that the two questions are separable, it is equally clear that Nancy did not separate them. At moments, his approach to community seems constitutive; he seems to be telling us what it is to be in community, and in doing so tells us what it is to be an individual or "singularity." A particularly important moment in this orientation is his argument that complete separation is inconceivable. Alternatively, at other moments he seems much preoccupied with avoiding the totalitarian implications of certain conceptions of community. Among these moments, the argument against a common substance is prominent. Now, it may be that he believes that we are in the fortunate situation in which what constitutes community is also what can save us from totalitarian conceptions of community, if only we could come to realize it. However, since he does not separate the questions, we cannot tell whether he believes this. In offering a critical evaluation, then, the most effective way to move would be to approach each question separately, asking whether he has convincingly made his case, and then, should the answer to each be in the affirmative, ask about the relation between the two. As it turns out, we will not get to the issue about the relation. I believe that Nancy does not answer either the constitutive or the normative question adequately, and thus does not present us with the possibility of a happy convergence. IInstead, Nancy presents us with a self-defeating answer to the normative question and an unconvincing answer to the constitutive one. I want to turn first to Nancy's normative answer, or, more precisely, to the status of Nancy's conception of community as a response to the question of how to conceive community in a nontotalitarian fashion. I turn here first because the criticism of his conception construed normatively is straightforward, while the criticism of his conception of community construed constitutively is more involved. Has Nancy, then, offered us a conception of community adequate to the task of avoiding totalitarianism? The answer to this question is in the negative, but not because his answer still permits totalitarianism a hold on community. Rather, it is because his conception of community, if it is to be understood as an answer to a normative question, presumes a prior commitment to a type of community bond that the conception itself excludes as totalitarian. Otherwise put, his conception of community proves too much against totalitarianism, eliminating in the same gesture not only totalitarianism but any grounds one could have to oppose it. To see how this is so, we must bear in mind that, on the normative interpretation, the question Nancy is answering is that of how to conceive community in a nontotalitarian fashion. One must assume that the answer he offers is one that a community that opposes totalitarianism might consider embracing. Otherwise put, the conception he promotes cannot just hang out there as a way to conceive community nontotalitarianly (pardon the expression), but, in order to have normative force, must be a conception that a community can consider as a reasonable possibility for itself to adopt. Unfortunately, no community can coherently embrace Nancy's conception of community as an antidote to totalitarianism, because in order to do so such a community would have to deny its own ratification of the value for which it would embrace the conception. How so? Whatever else a community that rejects totalitarianism would hold to (and, as I argue in Chapter 3, it must hold to much else), it would have to hold to a value something like "Totalitarianism is morally bad." Or, if we prefer to put the matter as a principle for action rather than as a value, "Communities should not choose totalitarian modes of self-constitution." In either case, the community is defining itself in terms of who it is or wants to be.

And it is precisely that kind of self-definition that Nancy's conception of community precludes. Let us recall that, for Nancy, community cannot be signified: "it is the community of not having a signification and because it does not have signification." But to hold a value or a principle like the ones discussed in the last paragraph is to have, if not full signification, at least a common substance, that substance being a value or principle by which the community defines itself. And it is precisely that which Nancy precludes as a possibility for a conception of community, because self-definition in terms of a common substance is a project of signification. To be precise, he precludes it not because actually giving such a signification is totalitarian or at least runs the risk of totalitarianism. In fact, he thinks that such an attempt is impossible to realize, because the project of signification is, for Derridean reasons, impossible to realize. Rather, he precludes it because the very project of such a signification, even though it always results ultimately in failure-that is, death-is precisely the totalitarian project, since the totalitarian project is a project of death. Thus, embracing Nancy's nontotalitarian conception of community cannot be done for the reason that it avoids totalitarianism, because a community cannot at the same time give itself that reason and hold to Nancy's conception.

The argument I have made does not entail that Nancy has not given a coherent answer to the question of how to conceive a community in an nontotalitarian fashion. I take it that he has done so. Rather, the argument is that it is a conception that no community can hold without denying the basis upon which it would hold it. Having seen this, it is important also to recognize that a community can hold to Nancy's conception if it decides to do so for no reason whatsoever. Otherwise put, if a community decides to conceive itself in a nontotalitarian fashion, it can do so according to Nancy's conception; but it cannot do so for the reason that it is a nontotalitarian conception or, for that matter, for any other reason. This is because any reason that one might adduce for embracing Nancy's conception would have to have reference to a value or a principle, which is precisely what Nancy's conception precludes. Why must a reason in this neighborhood have reference to a value or a principle? Because at issue is why a community would want to conceive itself one way rather than another, and reasons to do that refer to what is or should be valued or done.34

If all this is right, another conclusion follows as well. For Nancy, any common substance by which the community identifies itself runs the risk of totalitarianism. But it seems that we have isolated a common substance-a communal self-identification-that may be crucial in avoiding running this risk. That communal self-identification is in terms of a value or principle of nontotalitarianism. The situation is a bit complex here, so we need to move carefully in understanding this. I have argued that if a community were coherently to embrace Nancy's conception of community, it would have to do so for no reason at all. Such an embrace would be irrational, in the mundane sense of being done for no reason. But if the embrace of a communal nature is done for no reason, then any communal nature is as good as any other. There is no more reason to embrace Nancy's conception of communnity than to embrace a totalitarian one. And if this is true, then it is the requirements that Nancy's conception of community places upon a community that might embrace it that run the risk of totalitarianism. The risk is run because no reasons are offered in favor of Nancy's conception, and thus no justification in favor of it as opposed to any other communal conception. Now, this does not entail that the alternative of communal self-identification in terms of a value or principle of nontotalitarianism guarantees that a community will not become totalitarian. One can imagine a community that, in the name of nontotalitarianism, begins prohibiting all sorts of practices that it feels might pose a risk to full acceptance of other people. This would be a paradoxical situation, to be sure, but stranger things have happened. In the end, the idea that there may be a conceptual or philosophical guarantee against totalitarianism is suspect. Rather than claim that embracing a value or principle of nontotalitarianism guarantees-as opposed to Nancy's conception-that the risk of totalitarianism is avoided, I want to claim that it is a crucial condition for self-consciously avoiding that risk. A community cannot coherently embrace Nancy's conception of community as a means of avoiding totalitarianism except by accident. It would have to do so for no reason. But a community that wants to avoid totalitarianism-and not by accident-would seem forced to subscribe to a value or principle of nontotalitarianism. That route (or one like it) may be the only way to go to avoid totalitarianism by more than chance. I want to emphasize here that the argumentative strand I have been following here does not claim that Nancy's conception of community is mistaken. Nor does it claim that Nancy's conception of community is self-contradictory. The question of the plausibility of his conception of community will be investigated shortly. Regarding self-contradiction, the claim is that a community cannot embrace his conception of community without either engaging in self-contradiction or irrationality. Thus it is not the conception itself that issues out in contradiction (or irrationality), but the attempt to embrace it that does so, for the reasons I have detailed. So far, I have been investigating Nancy's conception of community as a normative conception, a conception intended to answer the question of how to think of a community in a nontotalitarian way. Some who either have read the previous summary of his thought or are familiar with Nancy may feel a bit uncomfortable, however, with the terms in which this discussion has been taking place. I have talked about a community's "embracing" Nancy's conception as though it were something a community had a choice to do. But Nancy often speaks of community as something that happens without a choice, as something any individual always already is. That way of speaking, to which I now want to turn, is the constitutive side of Nancy's conception of community. It tries to answer the question, What is it to be in a community? In the critical discussion so far, I have been reading Nancy as engaged in the normative project of offering a community a way to conceive itself in a nontotalitarian fashion. Another way to read Nancy, however, is as offering a constitutive account of community. This account is not, as mentioned above, a signification of what a community is, but instead an account of what it is in community that avoids the project of signification. Nancy does not separate the two issues, although I have argued that they must be considered separately. Let us turn, then, to a consideration of the merits of Nancy's constitutive account of community. In a constitutive reading of Nancy, no clear self-defeating problem haunts it. I want to discuss four weaknesses to Nancy's position considered constitutively, but, unlike the case with the normative reading, these weaknesses do not rule Nancy's conception out of court. There is no knockdown argument here. Taken together, however, and moreover in combination with an alternative constitutive conception of community that I will outline, I believe these weaknesses render Nancy's conception of community an unattractive one to pursue. Although there may be lines of development that would allow one to come to terms with these weaknesses, those lines, if my arguments are right, are going to be long ones to reel in. The first weakness concerns Nancy's justification for the constitutive case he offers. In fact, he offers no argument for the analysis of community in terms of sharing, and the one argument he offers for the necessary exposure of individuals is more limited in scope than Nancy believes. That argument, which we saw earlier, contends that to believe in complete enclosure or immanence is self-contradictory. I, on the other hand, have argued that while it may be that the idea of complete enclosure is wrong, it is not selfcontradictory. Moreover, having failed to establish that enclosure is selfcontradictory, Nancy's argument-since it was an argument concerned solely with self-contradiction-fails to show anything at all. In a bit, I will try to show that there are indeed arguments that complete enclosure is impossible (but not self-contradictory); these arguments, however, lead in a very different direction from the conception of sharing Nancy wants to promote. Moreover, even if Nancy's argument were right, it would not go very far in buttressing his case for community as sharing. All that argument purports to show is that complete immanence, complete self-enclosure, is a conceptual contradiction. But that tells us nothing about what the nature of the opening to community must be. Even if there is a constitutive exposure to others-and I believe that there is-what is its nature? And if it does not have a nature, why not? The argument against enclosure, which is the only argument Nancy provides for a constitutive reading of his conception of community, does not tell us. And thus we have a controversial conception of community, one that claims that individuals are at least partially constituted by community, but does not offer any justification for itself as an account of that partial constitution. Part of the reason for this lack of justification lies, I believe, in Nancy's failure to distinguish the normative from the constitutive interpretations of his conception of community. Much of the force of the argument for his conception carries normative, rather than constitutive, force. It is the urgency of avoiding totalitarian conceptions of community that preoccupies him: "In order to speak of the site that we are dealing with, I might venture the following thought: 'left' means, at the very least, that the political, as such, is receptive to what is stake in community." Rethinking the political approach of the Left, worrying about Stalinist and Nazi totalitarianism, trying to preserve difference without reducing it to identity: these are the concerns that drive Nancy's analysis. Thus the justifications he gives for his analysis, as we have seen, are largely normative. If the normative reading fails, however, then much of the justificatory force behind his approach is lost. Thus it is not entirely surprising that he offers little in the way of a constitutive interpretation of community-an interpretation of sharing as an answer to the question of what it is to be in a community-but it leaves this aspect of his view undeveloped. This first weakness is related to a second one. Not only does Nancy not offer much in the way of defense for his conception of community considered constitutively; it is not clear that he is in a position to assert that much more could be said in defense of it. This is because of his Derridean view of language and signification, and of the role the term "sharing" plays vis-à-vis this view. Now, there is a separate problem that arises for Nancy inasmuch as he depends on Derrida: to the extent that he depends upon Derrida his own position stands or falls with Derrida. I argue in the next chapter that Derrida does fall-or at least that he fails to convince. I do not want to treat that issue here, however, since it may be possible to hold Nancy's position independent of a commitment to the specificities of Derridean views about language. Nancy's case is parallel to, but perhaps not founded upon, Derrida's own. However, there is a problem that haunts Nancy concerning signification and sharing, and it is to that question I want now to turn.

The problem is that, if Nancy's Derridean reading of signification is right, then it is unclear how far his conception of community actually admits of defense. Recall that the project of signification, the project of articulating linguistic entities with more or less stable meanings, is necessarily a failure. It is a failure because beneath (or within) signification is "sense," which is an undeterminable that generates (always partial) determinability. For Nancy, the term "sharing" is one of these undeterminables, as is differance for Derrida. It points to (one wants to say, although the terminology under consideration forbids it, "signifies") a constitutive feature of community, that not only does community resist signification or full determinability, but in addition it is this very resistance itself. "Sharing" is an economic concept; it designates the movement of double exposure that is constitutive of community (and of individuality). As such, it does not even admit of the partial signification that other, more pedestrian terms admit of. This, by the same stroke, also severely limits what can be said in justifying its use in a constitutive "account" of community.

The problem Nancy faces here is analogous to the one faced by Derrida. Many people are familiar with the fact that the practice of deconstruction occurs by means of operations on a given text. Deconstructive claims, on the other hand (as opposed to deconstructive practice), are of wider scope than those texts, which is what makes them philosophically interesting. Since, however, differance (arch-writing, trace, etc.) is discovered in a single text, the task of generalization is a difficult one. This is because the internal limits that deconstruction discovers are text-bound, and in that way deconstruction is always parasitic upon specific texts. Its findings do not readily generalize. Derrida attempted to get around this problem by providing deconstructive readings of many texts and finding analogous internal limits within them. But, as I argue in the next chapter, the limits he found are not as generalizable as he believes they are.

This issue of generalization is even more urgent for Nancy than it is for Derrida, because Nancy does not even offer any treatments of specific communities from which he can launch the term "sharing" in the first place. While Derrida can point to a number of specific texts and say, "Look, the same kind of internal limits are approached in each; something like differance is happening in each," Nancy has neglected his homework in this area. He moves directly to the construction of a perspective, without offering evidence for it. But, and this is the crucial problem, even if he did offer evidence in the form of treatments of specific communities, he would still face the problem Derrida does: How does one generalize a term that is used pre-cisely to demarcate an undeterminability? How does one move from the communities upon which a term or analysis is parasitical to those communities that have not been treated? It is precisely in the nature of such parasitical terms that they do not readily generalize. How, then, would we even begin to understand the generalization of a term or a perspective from analyses of specific communities, assuming Nancy had given us any? Thus the approach to signification that Nancy relies upon-and specifically the role he sees the term "sharing" playing in that approach-makes it difficult for him to construct an adequate defense of his conception of community as an answer to the general question, What is it to be in a community? Having said this much, though, I should say just a bit more in order to allay any fears that I have just argued that without a foundationalist philosophical approach one cannot give an adequate defense of a conception of community. My argument is not that there has to be something like an absolute, foundational signification in order for a concept to be useful in accounting for some phenomenon. In the positive rearticulations of each of the substantive chapters of this book, I sketch accounts of things like community, language, morality, and ontology that (I hope) do not require such signification. Moreover, as mentioned, the immediate connection that Nancy assumes between foundationalism and some sort of stability in identity is spurious. The argument here, rather, is that the specific linguistic role that the term "sharing" plays in this Derridean-oriented view of language renders the generalization of that term as an approach to community a difficult one to defend, for the same reasons that other terms occupying that role, for example, differance, are difficult to defend.35 The third weakness of the constitutive reading is that, in characterizing community in terms of a bond that does not occur by means of any shared normative principles, it fails to explain what we might call the phenomenology of community. People experience themselves as bound to others at least in part through their commonly held commitments, which are instances of Nancy's common substances. Communities of political activists are bound by the ideals they believe need to be instantiated or by revulsion at the oppression that others face. Familial communities are bound by the projects of contributing to one another's lives or by common passions or interests.

Friendships are also bound by common interests, and often by an articu- lable common past that allows them to know one another better than others know them, and-in the case of deep friendships-in some ways better than they know themselves. These experiences, experiences of community in the pedestrian sense of "experience," not only do not find their way into Nancy's conception of community; they seem to be betrayed by it. My claim here is not that an adequate constitutive conception of community must preserve the phenomenology of community intact. That would be an absurd requirement to place on any conception of community, or of anything else for that matter. Rather, my claim is that Nancy's conception fails to explain, and may be precluded from explaining, this experience as having anything to do with community. Before seeing why that is so, let me offer an analogy so that the claim is clear. In explaining my experience of a chair as brown, physics does not preserve the phenomenology of my experience intact. On the contrary, there are certain aspects of my experience that are called illusory, for instance, that there is something that is brown out there in the world, independent of my sensory apparatus. An explanation must, however-if it is to be an adequate explanation-explain why it is that I experience that chair as brown. What is it that gives rise to brown-experiencing where in reality there are only atoms in the void? Merely to say the experience is illusory, that the chair is not really brown, is not enough, because it does not address how this illusion can arise and yet seem real enough. Nancy does not do this for the phenomenology of community, and it is not clear that he can do it without betraying that experience in a way that appears to me to be far too costly to his account. If he were to explain the phenomenology of community on the basis of sharing, he might offer an account of how common normative commitments arise from the bedrock of sharing-or, if we want to view sharing as woven into, rather than beneath, those commitments, how sharing is woven into them. Nancy does not do that. Now, the mere fact of his not doing it is hardly enough to constitute a weakness in the account. Philosophers do what they can; they do not do everything. After all, we philosophers have lives to lead. We cannot always be thinking of how to extend our analyses into whatever directions catch someone's fancy. The problem for Nancy lies not in the fact that he does not draw the explanatory connection between the phenomenology of community and sharing, but in the fact that he may not be able to even if he so desired it. Recall that for Nancy the idea of a common substance is, rather than an extension of or a relation to sharing, an other to it-specifically, a totalitarian other. On the one hand, there is sharing, and on the other, totalitarianism. Given this alternative, there seems little doubt on which side common normative commitments would fall. Thus, rather than relating common normative commitments to sharing, they would seem to be betraying those commitments. Otherwise put, a conception of community in terms of sharing would require that the phenomenology of community be read as totalitarian. This seems to me to be a betrayal of that phenomenology, rendering evil not only many of the bonds through which people see themselves as related to one another, but any common normative commitment that might act as such a bond. That seems to me to be too high a price to pay for any conception of community, particularly if there are equally good constitutive conceptions of community available to us, as I argue momentarily that there are. One might want to object at this point that I have misread Nancy here. Nancy does not see the idea of a common normative commitment as a straightforward betrayal of community, because he does not see in straightforward terms the sharing/totalitarian dichotomy I have outlined. We can no more avoid common normative projects than we can avoid the project of signification in speaking. What creates totalitarianism is not the mere fact of having common normative commitments, but failing to recognize the sharing that must subtend or be woven into them. Therefore, the objection runs, there is no strict division between sharing and common normative commitments, but rather a weaving in which sharing is primary but not exclusive.36 In dealing with this objection, we need to distinguish it from any normative reading we might be tempted to offer. What is at issue is not whether a conception of community as sharing ought to allow a place for common normative commitments, but whether it does. Once that distinction is made, it seems clear that it does not. Sharing may be woven into common normative commitments, but it is woven in as two different pieces of cloth may be woven together. These two different pieces of cloth may not be wholly different in kind. After all, the relationship between signification and sense is not that between stable identities and what resists them, but the more subtle relationship between the always only partially realized project of forming stable identities and the resistance that internally destabilizes that project. So it is with sharing and common normative commitments. The latter seek the common substance to which the former is the resistance. The latter cannot achieve it, because of the former. It may be the case that the latter require the former, but even if they do, they do so in the form of a totalitarian risk that the former acts to defuse in constituting what is really community. Thus the latter, the common normative commitments that appear in the phenomenology of community, can only be viewed, inasmuch as they are common normative commitments, as a threat to community rather than a constitutive part of it. Those commitments may be constitutive of community in the everyday sense, but not in the sense of what really binds us to one another. Now, an objection may be raised from another quarter here, complaining that with this objection I have just betrayed my own distinction between the normative and the constitutive. Is this third alleged weakness not a normative complaint against a constitutive analysis? Am I not saying that we ought not to think of community this way, because of the normative price imposed by such thinking-that is, its cost in terms of denying the constitutive value of common normative commitments? That is in good part what I am saying. My claim can be seen as having two parts: first, a part that says it is unclear how Nancy could account for the phenomenology of community at all by appeal to the notion of sharing; and second, a part that suggests that any avenue open to offer such an account would skate on normatively thin ice. The first part of the claim is constitutive, the second normative. Now the second part of the claim cannot, it is true, be raised by itself as a reason to reject the constitutive reading of Nancy. It cannot be offered as a claim that Nancy is mistaken in his answer to the question of what it is to be in a community. However, it does have a place. If there is another conception of community that is at least as good as the one Nancy offers, and, in addition to its power as a constitutive view of community, has the additional advantage that it can explain what I have been calling "the phenomenology of community" without making it look like a totalitarian moment in the communal structure, then we have a reason to embrace that alternative conception. A normative reason, I grant, but a reason nonetheless. Otherwise put, its standing as a reason against Nancy is parasitical upon its being in favor of an account that is at least constitutively as strong as Nancy's. Which brings me to the fourth weakness. There is at least one competitor analysis, which I shall sketch, that has the strengths but not the weaknesses of Nancy's account, and that does not make normative hash of our common normative commitments. To put the claim broadly, the fourth weakness is that Nancy's conception just is not as good as another conception of community. We are not stuck with the alternatives of Nancy's conception or totalitarianism, but can articulate another view, a view I have called "contingent holism," that can capture the insights that Nancy's approach yields without at the same time having to swallow its undue side effects. The account I want to promote has roots in both contemporary Continental and Anglo-American traditions. On the Continental side, Michel Foucault can be seen to have articulated this view. It also has roots in the thought of Gilles Deleuze and Jean-François Lyotard, although their embrace of it is more ambivalent. In the fourth chapter, I criticize the side of Deleuze's ambivalence that tells against this account, while in this chapter I use him to bolster my arguments. At a more distant remove, one can cite this conception's affinities with Jürgen Habermas, although in order for those affinities to become identities, Habermas would have to jettison many of the transcendental or "quasi-transcendental" moves he makes. On the Anglo-American side, the clearest forerunner is Ludwig Wittgenstein. The debt of this conception to Wittgenstein's thought will become clear immediately. Also of note on the Anglo-American side are Wilfrid Sellars and Robert Brandom. In fact, when I turn in the next chapter to a positive rearticulation of the Derridean approach to language and linguistic meaning, I rely heavily on their own views. Before turning directly to the account itself-or at least to a sketch of what such an account would look like-I want to list what I see as the requirements such an account must meet in order to be a viable competitor analysis to the one Nancy has offered. The best candidate for competitor would be able to embrace both the normative and constitutive claims to which Nancy lays claim. Fundamentally, the requirements upon such an account are two: that community be analyzed in a way that avoids totalitarian conceptualizations and that it do so while still recognizing that individuality is in good part constituted by community. The first requirement corresponds to Nancy's normative task, the second to his constitutive one. In approaching this account, we must bear in mind, as I have argued above, that these requirements are separable and answer to two separate questions. In order to keep that idea in view, I approach this sketch in three parts. First, I offer an answer to the constitutive question, What is it to be in a community? Second, I indicate how it is that being in a community means that community is partially constitutive of individuality. My view of that partial constitution diverges radically from Nancy's, although it shares with Nancy the idea that a community is not merely partially causally responsible for who one is, but is in addition conceptually a part of who one is. Finally, I show how the view of being in a community as I have sketched it is nontotalitarian in its implications. The central claim about which this sketch revolves is this: that a community is defined by the practices that constitute it. In order to understand what a community is, we must understand both what practices are and how a community comprises them. From there, it is only a short step to answering the question of what it is to be in a community. I want to define a practice as a regularity or regularities of behavior, usually goal directed, that are socially normatively governed.37 Let me expand a bit on the three aspects of that definition. Most practices have some aim in view. Teaching, for instance, is a practice that has as its goal the imparting of knowledge to students. Cleaning is a practice that has as its goal the removal of dirt. Children playing baseball are engaged in a practice, although children running aimlessly around a playground are not. However, there can be instances of a practice that do not have an end in view. While staring off into space is not a practice, sitting Zen, which (at least according to its proponents) has no aim in view, is a practice. Such goal-less practices, I believe, are the exception rather than the rule. They do exist, however. A practice must, in order to be a practice, be socially and normatively governed. I hesitate to say "rule governed," since there need not be explicit rules governing all the behaviors engaged in. However, although there may not be explicit rules, there will at least be "know-how" about how the practice is engaged in. There are right ways and wrong ways of engaging in the practice; and though not all the participants may be able to articulate exactly what makes the right way right, they know it when they see it. (The people who can articulate what is right and wrong about various aspects of a practice are often thought of as the experts in that practice.) When teaching one's kid how to ride a bike, for instance, one may find oneself saying something like, "When the bike goes this way, do this," where this is some sort of twist of the body. The normative governance of practices must also be social. That is to say, just as there is no such thing as a private language, there is no such thing as a private practice. Now, the socially normative governance of a practice does not entail that the practice itself must be social. Diary writing, for instance, is a solitary activity. It is both socially and normatively governed, however. There are ways one writes diaries, types of topics that are considered, potential readers (if even only oneself) that are kept in mind, and so forth. These ways are socially recognized as constitutive of the practice of diary writing. If one does not conform to these norms, one cannot be said to be engaged in an instance of the practice of diary writing.38 Some may worry about the social aspect of the normative governance of practices. Must what is to count as a practice be socially recognized as being one (not necessarily by everyone, of course, but by a significant portion of that part of the population that can be expected to recognize such practices)? Does this not lead to some kind of totalitarianism of the social? I do not believe it does. The point in isolating the concept of practices is not to tell people how to act. There is no implicit commitment here to an idea that practices are good and behaviors that are not engagements in instances of practices are bad. As will be discussed below, some practices are good and others are bad. What I am getting at here is the proper level of analysis for understanding what it is to be in a community. Introducing the concept of practice allows us to see both what constitutes a community and what it is about communities that is constitutive of individuals. The concept of a practice, then, lies at the intersection of individuality and community, as significantly constitutive of the former and perhaps fully constitutive of the latter (although I am not sure about that). As such, it will naturally have a social aspect. This social aspect-the social nature of a practice's normative governance-does not require us to take any normative stand pro or contra practices, and it does not give the members of a community any special power in delegating to them the say-so over what are and are not practices. It merely allows us to discover the social aspect of practices at the right point. If that is right, then there need be no worry that the requirement that practices be socially recognized as such to be such carries any insidious dangers in its wake.39 The third characteristic of a practice is that it involves a regularity of behavior. In order to be a practice, the various people engaged in it must be said to be "doing the same thing" under some reasonable description of their behavior. What constitutes a reasonable description in a given situation can be some matter of debate; and we can begin to see the pitfalls that might befall attempts at description when we recognize that, under a suitably abstract description, people we would want to say are engaged in very different practices might be said to be doing the same thing. For instance, a person handing a letter to a postal clerk and a person holding up that same postal clerk can be said to be doing the same thing under the description "interacting with a postal clerk." But we would surely want to say that mailing letters and robbing post offices are two different practices. I will not address that conundrum here, because for a couple of reasons I do not think it affects the general thrust of the account. First, I do not want to deny that practices can be composed of other practices, so the general issue of level of abstractness does not introduce any particular problems. Second, I see no reason to deny that practices have fuzzy borders and that it is difficult determine when someone marginally involved in a practice passes over into not being involved in it. This fuzziness is not a defect of the account but a fact about practices. As regularities, they allow for variation in the kinds of behavior that can be considered part of the regularity; and those variations, in turn, need to be assessed in order to say whether someone is involved in a particular practice. A person who approaches a church by walking slowly up the church steps can be said to be engaged in the same practice, attending church, as the person who approaches a church by bounding up the steps. But how about the person who stands in the doorway and half listens to the sermon? This is a matter for deliberation, not because practices are not regularities, but because what variations on behavior count as being within the range of a regularity is not always clear. That said, however, there is a clarification here that needs to be made. In practices, the regularities that different people are engaged in may not be identical but instead complementary. For instance, in baseball there are often at least ten people on the field at the same time (not including managers, the people in the dugout, etc.). They are all engaged in the practice of playing baseball, and indeed are all engaged in the same instance of that practice. However, they are not all "doing the same thing" in the sense of displaying the same regularity. (Note that this issue of doing the same thing is different from the one just discussed. There it was a question of different practices that may have similar descriptions; here it is a question of different regularities in the same instance of a practice.) Each of the roles taken by the players is socially and normatively governed, and each involves its own regularities. Taken together, those regularities are complementary; they constitute that instance of the practice of playing baseball. Thus, we need to understand the regularities of a practice as being either regularities of identity or regularities of complementarity. In addition to the three characteristics cited in the definition, there is a fourth characteristic that practices possess, one that does not appear in the definition but is entailed by it. Practices are discursive, by which I mean that they involve the use of language. It is easy to see why practices must be discursive. Since a practice is socially normatively governed, it must involve some sort of communication between participants in order that they may either learn or coordinate the activities that the practice involves. (This does not imply that all the norms of a practice must be articulable by each participant in the practice, but only that some of those norms must.) Moreover, this communication must be potentially accessible to nonparticipants, since without such accessibility the practice would cease to exist when its current participants dropped out. The communication required by a practice, then, must be linguistic. The idea of linguistic communication can be broadly construed here, needing only a set of public signs with assignable meanings. But practices do require language. This idea of practices as discursive-in later chapters I use the term "discursive practice" to emphasize the linguistic component of practices-is akin to Wittgenstein's idea that language games are central components of forms of life. Although my term "practice" can be read either as "language game" or as "form of life" ( Wittgenstein was notoriously obscure about these notions), believe that the approach to community that I am advocating is of a piece with Wittgenstein's own perspective. The discursive nature of practices, however, requires something to be said about language in order for the concept of practices to be more deeply understood. At this point, let me offer a promissory note on that score. Although I address the issue of the discursivity of practices a bit further on in this chapter, the positive rearticulation presented in the next chapter outlines in more detail how to think about language and about semantic terms from a perspective that takes practices to be linguistic in nature and language to be fundamentally "practical." Before turning to the question of how practices constitute communities, it is important to recognize that there can be crucial divergences between the goals a practice has in view and the results it actually achieves, even when it does meet its goals. Otherwise put, the effects of a practice (or of an instance of one)40 can be other (or more) than the ends promoted by those engaging in that practice. This can happen in at least a couple of ways. First, practices in a society are not isolated. They occur in the presence of-and often in interaction with-other practices. In combination with those other practices, a practice may have unintended effects in areas that are not its normal purview. As we will see, Michel Foucault describes some of these unintended effects in showing how oppressive power arrangements can emerge in a particular practice through the course of its interaction with other practices. These unintended effects are often opaque to the actor who has helped produce them, but not because he or she cannot in principle become aware of them. In a world of intersecting practices, it is difficult (perhaps impossible) to reflect upon the effects of one's practices upon others who are engaged in practices remote from one's own. This is because the complexity of the effects of interactions of practices that occur within a given society presents formidable obstacles to such a reflective project. This does not entail, however, that one cannot make some assessment of effects, or that assessment, inasmuch as it can reasonably be made, should not be attempted. Rather, the implication is that there are likely to be effects of one's engagement in practices that one is not in a position to recognize. There is another, related way in which the effects of one's practices can be other than the goals one has in view in participating in those practices-a way to which thinkers such as Foucault, Gilles Deleuze, and Jean-François Lyotard have called our attention. We often think of power as a negative, repressive force, serving to limit our actions and our thoughts but not to determine them in the first place. This way of thinking about power, which Foucault calls "juridico-discursive,"41 masks other and at times more important ways in which power operates. For instance, power can be a factor not only in the limitation, but in the constitution, of a practice, in its terms and in the justifications it gives itself. In turning to Foucault's analysis below, we will see this idea of power as a constitutive force as well as a negative, restricting one. If power is constitutive as well as restrictive, this implies that there can be effects-effects of power-that emerge from practices and that are opaque to those who engage in those practices, at least in their role as participants in the practice. If, for instance, the terms in which psychological practice is carried out, terms such as "normal" and "deviant," help create the objects of psychological study (normal and deviant people), then it is impossible within the parameters of psychological discourse to become aware of the effects of power that engaging in this discourse has. For Foucault, Deleuze, and Lyotard, the goal of a genealogical approach to certain practices is precisely to locate the unrecognized creative aspects of power within the practices that are the objects of a genealogy. We are now prepared to turn to the question of the relationship between practices and communities. This question can be straightforwardly answered: a community comprises its practices or intersections of practices. Since practices are socially normatively governed, a given practice may define a community. Often, however, a community is defined by the intersection of several related practices. An instance of a single-practice community would be people working in a particular political campaign. They are engaged in a common task, recognize their compatriots as being so engaged, and are bound by this engagement, this recognition, and the norms of their practice. Everyday talk reflects the use of the term "community" in this way: we speak of political, religious, and even economic communities in referring to communities comprising specific practices. In many cases, however, it is an engagement in several overlapping practices that forms a community. Within the world of political organizing, for instance, communities are formed not just among the people working on specific campaigns, but among those within those campaigns who also work on other campaigns and who-often as a consequence of this-socialize with one another away from political work. As another example, people who work at the same institution may form a community, but in addition communities are formed among those who are engaged with one another outside the workplace, whether in socializing, volunteer work, raising children, churchgoing, or whatever. The members of these types of communities are in general probably more self-consciously aware of themselves as members of a community than those who share only a single practice, although there are exceptions here (as, for instance, in the case of a community of actors who define themselves by their abandonment to their art).

Now, in the former case, we might want to refer to these communities as communities of communities, communities composed of single-practice communities. There is no bar to doing this, provided that we refrain from ascribing any ontological or metaphysical significance to such a reference. Such a community of communities is not (or at least not necessarily) a "metacommunity." It is just another type of community.

Much of what I have said so far in my positive rearticulation of community may sound pretty pedestrian to those who have followed the convolutions of my discussion of Nancy. Good. I believe that there are complications involved in thinking through community, and I treat a couple of them below. But I also believe much that is useful can be said about community by calling our attention to some noncontroversial facts about the ways in which we live our lives.

It might be objected here, however, that my treatment of community renders community vague. It may be difficult to tell sometimes whether or not there is a community at hand, since communities depend on practices and, as noted, the borders of practices are themselves vague. As in the case of practices, however, the borders of communities just are vague. Where one community leaves off and another begins, and whether there is a community in the first place, can be difficult questions to answer. And however they are answered, it cannot be by recourse to a set of finely honed specifications to which the object of investigation can be submitted. It should be noted, moreover, that the vagueness that characterizes practices and communities has at least some passing similarity to the idea, promoted by Nancy, that community cannot be signified. Although I think there is much more to be said about the nature of community than Nancy does, we agree that what can be said is exhausted before the point of absolute clarity. Otherwise put, there can be no foundationalism when it comes to articulating community, for the reason (among others) that the borders of community cannot be precisely fixed in a way that a foundationalism would require.

Having discussed what a community is, I have not yet answered the question of what it is to be in community. I want to turn to that question-but not just yet. First, I want to call attention to the fact that as the view I am constructing has proceeded, it has, in addition to explicitly addressing practices and communities, at the same time implicitly offered a conception of the social. This conception sees the social as a network of intersecting practices (and thus communities) with no single binding principle (e.g., the economic substructure) by which it is to be explained. Although there may be nodes or practices in the network that are particularly important to study in order to see what the network is all about-nodes where many or particularly important practices intersect-there is no privileged point of view from which the whole may be surveyed from "above" or "outside" the specificities of the various component practices. This conception of the social is central to the writings of Foucault, Deleuze, and Lyotard. Deleuze, in collaboration with Felix Guattari, has called attention to it with the striking image of the rhizome. Deleuze and Guattari contrast rhizomatic conceptions with "arboreal" ones, conceptions that look for a single root from which structures arise and which, once understood, yields the principle for grasping the entire structure. In depicting the rhizomatic approach, Deleuze and Guattari state: Let us summarize the principal characteristics of a rhizome: unlike trees or their roots, the rhizome connects any point to any other point, and its traits are not necessarily linked to traits of the same nature; it brings into play very different regimes of signs, and even nonsign states. The rhizome is reducible neither to the One nor the multiple. It is not the One that becomes Two or even directly three, four, five, etc. It is not a multiple derived from the One, or to which One is added (n + 1). It is composed not of units but of dimensions, or rather directions in motion. It has neither beginning nor end, but always a middle (milieu) from which it grows and which it overspills.42 This image of the rhizome, of practices (which involve both discursive and nondiscursive aspects-"different regimes of signs, and even nonsign states") that are always evolving and possess no defining or constraining center, is the best image I am aware of to picture what is meant by the term "contingent holism." Contingent holism sees the social world as composed of practices that intersect with and affect one another (although not every practice intersects with every other practice), that change over time, that form the parameters within which we understand ourselves and our world, but that do not offer a foundation from which the world can be exhaustively or indubitably understood. Individuals, moreover, living as they do in social worlds, are largely constituted by the practices of those worlds. This constitution is not only a causal one, in which engaging in certain practices causes one to be a certain kind of individual. To think that practices are only causally related to who one is assumes that one can separate the entirety of who one is from the practices one is engaged in. That is an untenable assumption. I, for instance, am engaged in a number of practices: philosophy teaching and research, political organizing, the raising of children. Ask me who I am outside of these practices I engage in, and I begin to stumble. It is not that I cannot come up with anything about myself that is not part of a practice; rather, it is that most of the important aspects of who I am are tied up with the practices I am engaged in. Otherwise put, I am in part conceptually as well as causally constituted by the practices in which I am engaged.43 To be so constituted is to be constituted by the communities of which one is a member, since communities are themselves constituted by practices. In this way, then, communities are partly constitutive of individuality. Now, one might be tempted to object at this point that this form of community constitution is too voluntarist. After all, do we not choose our practices, and thus our communities? And does this not allow the possibility of a preformed individuality that only later chooses practices, in contrast to the deeper form of constitution that Nancy was after? We will see momentarily another form of individual constitution by communities, one that is closer to the kind of constitution Nancy has in mind. But even here it would be conceding too much to admit a deep voluntarism into this type of community constitution. Individuals are not brought into this world with preformed personalities that only later get involved in some of the practices of their society. Individuals are plugged into practices from the moment of birth. At first, they are more passive than active participants in those practices. But they are being constituted nonetheless by practices they have not chosen and about which they have no say. Those practices are partly, but not solely, familial-a point Deleuze and Guattari have emphasized.44 And later, when one chooses other practices, the process of choice does not occur by means of removing oneself completely from the practices that heretofore have had a hold on one. Choosing practices always comes from within practices. And in that way, practices are constitutive of oneself causally as well as conceptually.

Here one might be tempted to raise a larger question, which I intend to duck. The question is how much influence practices have on one's choices. To what extent is someone able to choose one's practices divorced from the hold one's past and current practices have on one? Alternatively, are one's choices wholly determined by the effects of those practices? These are large and important questions that relate to the large and important question of freedom and determinism. I have no light to shed on this matter, and fortunately do not require it. My point here is the more modest one that who one is and how one gets to be that way are in good part a matter of the practices in which one engages, and that the practices in which one engages are influential in the choices one makes about what further practices to engage in.

Having seen, then, how community is in good part constitutive of individuality, we can recognize as well that individuals often belong to different communities at the same time, because they often belong to widely divergent practices. This seemingly obvious and pedestrian point has two ramifications that have often been overlooked. First, it allows us to recognize that individual uniqueness need not be conceived as a matter of who one is outside of one's practices, as though uniqueness were a matter of what is left over after the common social (or individual, but socially normativized) projects in which one is engaged are removed. This is not to claim that uniqueness is reducible to the practices in which a person participates, but that uniqueness can derive in part from-rather than apart from-the unique array of one's practices.

Second, and more important, this conception of individuals in community points the way toward a resolution of one conundrum that has haunted political theorists regarding the proper way to conceive of minority identities. One example of this conundrum is the following: There seems to be something shared by many, if not all, of those in U.S. society who are of African descent-something more than the fact of having darker skin than some other members of that society. But how to conceive what is shared is a bit difficult. Oppression, surely. But if that is all, then African-Americans are definable in their uniqueness only negatively, only by means of something bad that has happened to them. One of the responses to this problem has been to say that there is something about being an African-American that is positive and to be valued above and beyond the common experience of oppression. This response runs the danger, however, of saying that there is something to the darkness of the skin or the fact of descent that provides the basis for a positive valuation. The former is problematic because it plays into the hands of racist ideologies; the latter is problematic because the connection many African-Americans have with Africa is very tenuous. The idea of descent by itself is empty, and the history of African-Americans is, aside from the commonality of oppression, very different from that of Africans over the past two hundred years. Thus the question arises, how can AfricanAmericans (or other traditionally marginalized groups) conceptualize their communities positively without falling prey to various ideologies that are better off avoided?

The answer, I believe, lies in turning to the concept of practices. AfricanAmericans have shared a set of practices in which non-African-Americans have had only a marginal presence. One of those practices, for instance, is jazz. African-American identity is bound up with jazz in a way that nonAfrican-American identity is not. This does not mean that jazz develops from some deep essence of blackness, or that non-African-Americans cannot make a contribution to jazz. Rather, it means that this is one of the practices that has been defining for much of African-American identity: one of the practices that have constituted the individuality of many African-Americans. The value, then, in studying topics like African-American history lies in coming to understand what it is to be of African descent in this society, coming to understand the practices in which people of African descent have been engaged. Looking at matters this way helps get past the bouncing back and forth between essentialist affirmations and denials of all positivity that has characterized discussions of the identities of traditionally marginalized communities.45

So far, we have seen how community is constitutive of individuality inasmuch as it is one of the things that is defining for identity. There is another way individuality is constituted by community. This has to do with the discursivity of practices. Earlier I claimed that practices must be linguistic, or else the social nature of a practice could not be had. Tyler Burge has argued that this discursivity is constitutive of individuality in the sense that language is determinative not only for identity but also for a person's mental content. Again, this determination is not merely causal but also conceptual. What a person is thinking is conceptually dependent on what is going on outside a person. In terms closer to Burge's own idiom, individuation of mental content cannot be conceived individualistically. It must be conceived as well by reference to the language of one's communities and even to the external physical world. I turn now to his arguments in order to show how we can conceive of community as deeply constitutive of individuality at a depth that competes with Nancy, while avoiding the untoward consequences of embracing Nancy's own conception.

Before turning to the specific arguments, let me situate Burge briefly, since readers more steeped in the Continental tradition may be unfamiliar with his work. Burge's work is primarily in the philosophy of mind, and specifically in that area of intersection between philosophy of mind and philosophy of language that goes by the name of mental content. He has been concerned to show that mental content is not simply a matter of what goes on within a person's mind or brain, but also a matter of what is going on outside, specifically in that person's environment and language. The interest of this for us is that, to the extent that he can show this, he has developed a position similar to-although perhaps better in its development than-Nancy's position that there is an economy of "inside" and "outside" that resists any strict delineation of borders. Otherwise put, individuality is constituted at least in part by community.

Burge's argument is that mental content-what a person is thinking about, feeling, believing, and so forth-cannot be accounted for solely by an examination of the state of that person's "internal qualitative experiences, his physiological states and events, his behaviorally described stimuli and responses, his dispositions to behave and whatever consequences of states (non-intentionally described) mediated his input and output."46 Burge offers an array of arguments, mostly in the form of thought experiments, to make his case; however, these arguments can be divided broadly into three types: social, physical, and dialectical.

Burge's social argument is offered most forcefully in his paper "Individualism and the Mental." Burge asks us to consider a person who has a small set of true beliefs about arthritis, including the belief that he has arthritis. This person, however, also believes-falsely, of course-that he has arthritis in his thigh. Arthritis is an ailment that only affects the joints. Now, suppose the social situation of this person were different, such that in his society the term "arthritis" were used to refer to a variety of rheumatic ailments, and that therefore the person in question were using the term correctly. In that case, the person would not have thoughts, beliefs, or feelings about arthritis-that is, arthritis as we use the term. There is, in the counterfactual situation, neither an individuation of his mental content that corresponds to our notion of arthritis nor a linguistic individuation of our notion of arthritis. When he says, "I think I have arthritis," he will not mean arthritis, but instead a general rheumatic ailment.47

What is crucial to this thought experiment is not simply that someone raised in a different society would have concepts different from ours, but more deeply that one has them even though one would have the same physiological, behavioral, and mental experience. Nothing has changed in this thought experiment except the outside social environment; that change, however, has changed the mental content of the "arthritic" individual. He no longer has thoughts, beliefs, and feelings about arthritis, some of which were right and some wrong. He now has a mental relationship to something else-albeit a something else that is called "arthritis."

Burge offers an analogous argument regarding the composition of the physical environment. He relies on Hilary Putnam's Twin-Earth experiment in "The Meaning of 'Meaning'" to argue, in an extension of Putnam's own analysis, that not only would the term "water" not mean water on the planet of the doppelgänger (where its chemical composition is XYZ instead of H 2 O), but the doppelgänger would in fact not have propositional attitudes toward water at all. "[I]t is hard to see how Adam(te) [the doppelgänger] could have acquired thoughts involving the concept of water... There is no water on Twin-Earth, so he has never had any contact with water... Further, no one on Twin-Earth so much as uses a word which means water."48 Putnam, therefore, though correct on the analysis of the meaning of natural-kind terms, is wrong when he assumes in his thought experiment that the earthling and his doppelgdnger have the same thoughts.49 Different physical environments determine different thoughts. Burge's last argument against individualism in the individuation of mental content relies not upon the physical or social environment but upon how we learn language and what the status of that learning is.50 He claims that the learning of many words, if not most, consists in a back-and-forth movement between "normative characterizations," which are guidelines dictating what something must have or be in order to be counted as a member of the set covered by a certain term, and "archetypal applications," which are exemplary members of that set.51 The primary thrust of his argument is that since this dialectic exists, it is often not incoherent or self-contradictory to doubt statements that a community considers true by virtue of meaning. The dialectic precludes in many cases a tightness of fit between accepted synonyms, so that room for doubt or correction remains. What this room for doubt implies, however, is that two individuals, identical except for their social environments, can have two different thoughts. The thought experiment Burge relies upon to show this is to imagine two people who think their respective communities are wrong in their beliefs about the meaning of a term. However, one of the skeptics is correct in what he or she holds the accepted meaning of that word to be, and the other is not. In the latter case the meaning that the skeptic thinks is the correct one is, by coincidence, in fact the one his community agrees is the correct one. Now, both skeptics are in the same physiological position, but both have different thoughts. The skeptic who is correct about what he or she thinks is the accepted meaning has doubts about the meaning of that very term. The other skeptic has no thoughts about that meaning, since it nowhere appears in his or her community's language. There is a negative conclusion to be drawn from Burge's thought experiments: mental content is not purely an "internal" affair. There is a positive conclusion to be drawn as well: mental content relies upon a person's social and physical surroundings. Together, these conclusions point to a fact about community that runs close to the perspective Nancy wants to defend, but diverges in crucial ways. What is going on inside an individual, the mental content of that individual, is constituted not only by the individual's physiological states but also by the discursive practices-and thus the communities-in which that individual is engaged. To put the claim in Nancy's terminology, the individual (or singularity) is exposed to the community, and thus is not a self-enclosed whole. Earlier, in assessing Nancy's argument against self-enclosure, I said that although the idea of self-enclosure is not self-contradictory, it is false. Now we can see why. The idea of selfenclosure can be conceived without contradiction, but so to conceive it would miss the important distinction between the existence and the content of an individual's mental states. The existence of an individual's mental state may indeed be self-enclosed, at least conceptually (although not causally, as we saw earlier); but the content of an individual's mental state cannot even be conceptually self-enclosed. This way of arguing for exposure has several advantages over Nancy's approach. First, it allows for a much finer grained approach to exposure than Nancy's approach does. In fact, one of the implications of Nancy's approach is precisely that anything more than a very partial account of community cannot be given. In contrast, Burge's approach, if embedded in a larger account of community, allows us to say much more about what a specific community is, what binds it together, and what its limits are. All this would occur by way of accounting for its particular discursive practices. Second, this approach to exposure does not face the problems that Nancy's approach, constitutively interpreted, faces. Unlike Nancy's approach, it does offer arguments for its position; those are the arguments here rehearsed. In part, it is able to offer such arguments because it does not endorse the Derridean view of language that Nancy endorses, and thus does not face Derridean problems of generalization. (The next chapter will offer a view of language that fits neatly with this view of community.) Finally, it gets the phenomenology of community right. Although Burge's argument that mental content is partially determined by a community's discursive practices may initially be counterintuitive, it converges with the larger idea that who someone is, is in good measure a matter of specific recognizable bonds that he or she shares with others in his or her community. In addition to the two conclusions I have cited so far, there is another positive conclusion as well that can be drawn from the third thought experiment. This last conclusion concerns the nature of meaning, and will become of deeper interest to us both in the final part of this chapter and in the next chapter. If the dialectic Burge describes is an accurate account of the learning of meaning, then linguistic meaning is normative in the sense that it tells us how, among various possibilities, we should use our words. An account of meaning that precluded this dialectic (or something analogous to it) and considered meaning to be fixed by community usage would not be normative in this rich sense. On this latter account, there could be no debate over correct word usage, because the fact of community usage would be the final arbiter of any disagreement. For Burge, however, the field of linguistic meaning is more open than that: community usage is a recommendation rather than a legislation, and thus the normativity of linguistic meaning can be seen as a field of struggle rather than merely a binary opposition between correct and incorrect use.

It is the failure to understand the normative underpinnings of Burge's conception of language that has motivated much of the criticism of his work. Andrew Woodfield, among others, raises the question whether Burge has elided a distinction between the ascription of mental content and that content itself, and argues that the preservation of that content would subvert the power of Burge's thought experiments.52 "[T]he same practices reveal that our ascriptions of content are sensitive to all sorts of background information, including information about the social environment of the subject, yet we do not, or should not, treat such variability as proof that the content in itself varies."53 Woodfield argues that mental content should be considered as analogous to linguistic content in the sense that a concept is defined by its inferential role, and concludes from this that the first patient in Burge's arthritis example-the patient who shares our linguistic situation-was no more thinking about arthritis than the person in the counterfactual situation. This is because the inferences that the person drew from what he thought was arthritis were different from the inferences compatible with the concept of arthritis. Thus, not only did the patient misuse the term "arthritis" in the self-ascription of mental content; the content of his intentional life did not contain arthritis, but something else. Moreover, that something else was the same content that the person in the counterfactual situation intended. Neither of the two people in Burge's thought experiment were thinking about arthritis.

Burge's conception of language, however, raises doubts about the analysis Woodfield offers. Woodfield assumes that in deciding upon mental content one can reinterpret one's having a mistaken assumption about a term as actually thinking about something else.54 If the meaning of linguistic terms is conceived normatively, however, this is not the case. Meaning is not merely a matter of describing, it is also a matter of commitment. To say of oneself or someone else that one is thinking about arthritis is to say that one is committed to a certain inferential pattern. If one draws an inference from the term that flies in the face of community usage, then one can either defend one's use of the term or abandon it. If one abandons previous usage, then one admits that one was thinking about that term mistakenly. But it makes no sense to say that one was actually thinking about something else. While it may be true that at the physiological level certain events were occurring that would not occur if one had the correct idea of what arthritis is about, this is irrelevant to Burge's argument. The force of his thought experiments relies on the fact that physiology can be distinguished from mental content. What cannot be distinguished, if Burge is right, are linguistic ascriptions of content from content itself. One might want to object here that the normativity of meaning as I have interpreted it undercuts the idea that a community's discursive practices are partially constitutive for an individual's mental content. After all, if the dialectic Burge cites is right, then what a community thinks the meaning of a term is does not necessarily determine what goes on in a person's head. And if that is so, then a person's mental content is not necessarily determined by the practices of a person's linguistic community. Although the first conclusion of this objection is true, the second one does not follow. It is true that meaning cannot be reduced to community beliefs. If meaning could be so reduced, it would not be normative in the sense the dialectic implies. However, this fact does not mean that the practices of a community play no role in determining mental content. When a person engages in questioning the meaning of a term as used by a community, he or she must do so on the basis of other terms the community uses that he or she accepts (even if those terms arise within the context of practices of another community). Questioning meaning, then, can only be local; it cannot be global. Thus, even though a community may be wrong in a specific instance about the meaning of a term-which implies that its beliefs are not constitutive of mental content-nevertheless its discursive practices remain generally determinative, since that error can only be conceived on the basis of a background of correct meaning use. Moreover, if a community is vindicated against a challenge of meaning by a member of that community, then it can be said in that specific instance that what is believed by the community does partially determine that person's mental content. We must be clear, however, that this determination does not arise because the community believes it, but rather-and this is the normative point-because the community is correct in its belief.

An alternative objection one might want to raise here concerns the claim I made earlier that there is a separation between the constitutive and normative questions regarding community. By introducing normativity into the constitutive account, have I not collapsed the two questions in my own approach, just as I accused Nancy of doing in his? In order to see why I have not, we need to recall first that the normativity at issue affects not so much the question of what a community is, but the question of how it is constitutive of individuality. The next step is to see that what it is to be in a community is a matter, in part, of how one's mental content is determined by a community. (It is also a matter, in part, of one's being what one's practices are, which involves no normative issue.) Recognizing that, we can further recognize that although what that constitution is in a given case is a normative matter-inasmuch as meaning is normative-the fact of such constitution is not. Thus the answer to the question of what it is to be in a community is not normative, although there is normativity in the answer to the question of what it is to be in this community, regardless of which community this community is. Therefore, while there is a point of contact between the normative and constitutive questions, there remains a strict distinction between asking what it is to be in a community and how community should be conceived.

The positive rearticulation of community I have been sketching here has, so far, addressed the questions of what it is to be in a community and how individuality is constituted by community. That is, I have only been addressing the constitutive issues. Lingering in the wings, however, is the normative question; and it is a question of no less urgency than the constitutive one. Recalling that I have argued that the constitutive and normative questions are separable, it is entirely coherent to argue that although I have gotten the constitutive approach right, to conceive things aright constitutively would bring all sorts of normative problems in its wake. Put in Nancy's terms, the constitutive approach I have offered might run the risk of totalitarianism. In closing the positive rearticulation, I would like to turn to this question. I do not want to be interpreted as holding that whatever practices a community embraces either dictate who we are and how we must act or are immune from critique. My argument here is that there is nothing that leans toward totalitarianism in this approach; in fact, it allows us to conceive a progressive politics that avoids totalitarianism. This allowance is exemplified by applying a specific Foucaultian analysis to the considerations to which Burge has drawn attention. All of this does not imply, however, as I pointed out before, that this approach necessarily resists all totalitarianism. Rather, it implies that the alleged bent toward totalitarianism possessed by every approach toward community that views community in terms of common substances is missing in this case.55 The entry into some of the normative possibilities of this approach is via the normativity of meaning, the third conclusion of Burge's thought experiments. This normativity invites not only questions of the rational coherence of decisions about meaning but also questions of historical and political emergence. Otherwise put, the issue of normativity can be viewed as a matter of getting it right not only according to the epistemic standards of a community but also according to its moral and political standards. Although Burge recognizes a social factor in determining what "goes on in our heads" and derives part of that recognition from an analysis of the social openness of semantic decisions, he neglects at least two political questions that can impose themselves in the wake of that recognition and that analysis. First, what political and historical factors need to be taken into account in understanding why our linguistic meaning structure is the way it is? Although Burge opts to discuss pragmatic features of most rational usage in showing how the dialectic works, his thought experiment hints at another direction that his work opens up. He refers to someone who believes that the use of the term "sofa" to indicate a specific type of furnishing serves to "conceal, or represent a delusion about, an entirely different practice."56 Here it is not a pragmatic consideration that is at issue in deciding what meaning a term should have, but rather a political one. What Burge recognizes in this passage, although that recognition remains undeveloped in his work, is that if linguistic meaning resists reducibility to community beliefs, then the question of why certain meanings, rather than others, have been adopted by a community may admit of answers of a more politically controversial nature than the idea of pragmatic features of usage would lead one to believe. The second question is related to the first. How do we come to understand what "goes on in our heads" as a purely internal affair? Here the issue is not the meanings of specific terms but a misconception about the social (and thus political and historical) dimension of meaning (and thus mental content, individuality, and community) itself. In our society, certain perspectives on meaning and mental content have been privileged at the expense of others. In particular, an approach to mental content that conceives it-mistakenly, as we have seen-as a purely internal affair (and, concomitantly, an approach to linguistic meaning that has ignored or overlooked its normative dimension) has privileged inquiry into individuals, rather than into social surroundings, as the arena of explanation of individual experience.57 The social constitution of experience and the political dimension of discursive practices have been ignored in favor of individualizing epistemic approaches. Inseparably, what "goes on in our heads" has come to be seen as the product of a fixed an unchanging nature rather than as the product-at least in part-of conflict and struggle. These two questions converge on the status of the discipline of psychology,58 asking first what its relationship to the surrounding disciplines is and second why that relationship has not received the scrutiny it might deserve. Psychology, in other words, rather than be accepted solely as a paradigm by means of which discovery takes place, must as well be investigated as the product of historical contingencies and the source of social and political effects. That Burge accepts the psychological paradigm without political or historical question is evident in his article "Individualism and Psychology".59 There, in arguing from perception that individuation of mental content is nonindividualistic, he admits that his analysis assumes both that there are such things as psychological states that represent the world and that "there is a scientific account to be given that presupposes certain successes in our interaction with the world... and that explains specific successes and failures by reference to these states."60 Broadly, Burge's anti-individualistic thought experiment in this last text hinges on the fact that what is perceived is socially determined in the way that what is thought is socially determined in the articles cited above. Here, however, he makes explicit the idea that it is successful perception that provides a model for what is perceived. He writes: "Theories of vision, of belief formation, of memory, learning, decision-making, categorization, and perhaps even reasoning all attribute states that are subject to practical and semantical evaluation by reference to standards partly set by a wider environment." The political implications of such standards are hinted at by Burge when he claims that while psychological theories are not of themselves evaluative, "they often individuate phenomena so as to make evaluation readily accessible because they are partly motivated by such judgement."61 In arguing that mental content is nonindividualistically individuated, that social and linguistic standards inform that individuation, and that the dialectical process by which those standards are arrived at precludes in many cases the possibility of their closure, Burge has opened the door to questions of the historical emergence and the political orientations of those standards, of what "goes on in our heads," and thus of the ways in which individuals are communally constituted. That he has not engaged those questions, and indeed accepts the psychological framework within which he asks them, is probably due to the fact that the story he tells is conceptual rather than causal. However, the very conceptual story he tells leads naturally to causal questions concerning the status of discourse about the mental, questions that have been addressed by Michel Foucault's genealogy of psychological discourse. Foucault's project regarding psychology is one that shows how it arose, what other discourses and practices it intersected with, and what political effects it has had. In his most sustained discussion of the genealogy of psychology, Discipline and Punish, he writes of the modern soul that "it exists, it has a reality, it is produced permanently around, on, within the body by the functioning of a power that is exercised on those punished-and, in a more general way, on those one supervises, trains and corrects, over madmen, children at home and at school... On this reality-reference, various concepts have been constructed and domains of analysis carved out: psyche, subjectivity, personality, consciousness, etc."62 A genealogy of psychology, then, will reveal at least some of the social, historical, and political conditions whose convergence determined the idea of the soul, or the mind, as we currently conceive it. Thus, like Burge, Foucault attempts to remove the mind from its moorings in the state of the individual; and like Burge, that removal focuses upon the concepts we use in order to determine what is going on inside a person's head. However, unlike Burge, Foucault's target is not mental content as a conceptual matter but the mind as a political concept. The details of Foucault's genealogy of psychology cannot be recounted in detail here.63 The part of the story offered in Discipline and Punish, however, can be given a brief summary. Torture, as the preferred method of punishment before the nineteenth century, often had effects unintended by the punishing authority and the king in whose name punishment was performed. Its gruesome details served not only as a deterrent against further crime but also as a source of resentment against regal authority and as a symbolic martyrdom of the criminal. Reformers emerged who called for gentler forms of punishment, both as a more humane practice and as more conducive to social order. At the time of these reformers' writings, however, a set of disciplinary practices that had been spread about in different regions of the social field-for example, monasteries and the military-came to the attention of punishing authorities. What distinguished discipline from torture was, among other things, the change from destroying the body to manipulating it down to its last detail. This manipulation allowed it to be made both maximally conforming and maximally productive. These changes, moreover, coincided with the rise of industrial capitalism and its need for disciplined labor. With this newfound focus upon manipulation, which was adopted in the prisons and in the factories (and was refined in the military), the goal of punishment was no longer that of eliminating what was criminal but that of reforming what was out of order. Practices of normalization came to replace practices of torture; observation and manipulation replaced random spectacle and execution; psychological intervention replaced regal authority. As Foucault writes, "These two great 'discoveries' of the eighteenth century-the progress of societies and the genesis of individuals-were perhaps correlative with the new techniques of power, and more specifically, with a new way of administering time and making it useful, by segmentation, seriation, synthesis and totalization."64

Thus the concept of the mind as a discrete substance belonging to an individual whose nature it is the project of psychology to discover is inseparable from an intersecting set of politically charged practices within which it emerges. To the first of the two questions raised earlier regarding the political and social factors that determine the emergence of linguistic structures, Foucault offers the structure of an answer with his genealogical method.65 To the second question, why we understand the mind as a purely internal affair, Foucault responds with a specific genealogy that isolates the practices and motivations that are intertwined in the emergence of psychology as a therapeutic discipline.

Foucault's approach to psychology does not, of course, tell the whole story about how to conceive community. In fact, if the view I have been articulating is correct, that question is best answered-aside from some of the generalities I have offered in this chapter-by means of a lot of local analyses. If community is a matter of practices, then the urgent normative question facing most communities is not, How should we conceive community? but rather, How should we conceive our community? It is those specific conceptions that will be determinative not only for how the community will look, but also, because of their discursivity, for the landscape of the mental content of the individuals engaged in that community.

Although Foucault works along a different register from that of Burge-the latter dealing in mental content and the former in the concept of the mind-they converge in their articulation of how the mental defies individualistic approaches. Foucault's causal story expands and deepens, and at points questions, the conceptual story offered by Burge; Burge's story offers a picture of language that renders Foucault's genealogy more plausible. Taken in their complementarity, and in combination with the general picture of community with which I opened the positive rearticulation, these stories offer a picture of community that has both the normative and constitutive virtues that Nancy seeks but does not find, without confusing the normative and constitutive dimensions of conceptions of community. Moreover, in contrast to Nancy, for whom specific political recommendations seem to be impossible, since they would imply an appeal to principles of common substance, which he rejects, the view of community that I am pressing here allows-as the Foucault example illustrates-specific analyses with specific recommendations based upon specific principles. (We will see in more depth the application of a specific principle in Chapter 3.) Therefore, not only does the view I am proposing here contain the normative virtues of nontotalitarianism that Nancy desires, it does so without exacting the cost of precluding specific political analyses and recommendations. Finally, the proposed view, although saying much more about community than Nancy thought either possible or safe, avoids the foundationalism that Nancy's account eschews. It does so by distinguishing between the idea that a community defines itself by means of a common substance or, more often, common substances and the idea that those common substances must be signified in the full sense both Nancy and Derrida reject. Common substances arise in the context of practices, which are both historically contingent and evolving. Communal identities are not the product of a transcendental operation, and they receive no transcendental guarantee. Instead, they emerge in the unfolding history of practices that form communities, and vanish as those practices change. Some of those identities are beneficial, others insidious. Since, however, they are anchored in no deep foundations, they remain as distant from the totalitarian projects of signification as the concept of community Nancy develops. Jean-Luc Nancy's work has done much to call the attention of Continental philosophers to the importance of a conception of community in our philosophical work. It has also done much to raise the issues of the community's role in constituting individuality and of the totalitarian dangers of communal self-perceptions. The argument of this chapter is that, however important the issues he raises, his approach to them is inadequate. What one needs in order to address these issues is not a view of community as unsignifiable exposure, but a view of community as practice or practices. In addition, one needs to embed this view of practices in a contingent holism, an approach that rejects both foundationalism and reductionism and stresses instead the historically changing, politically invested, and socially interactive characteristics of practices. It is this view of community that underlies the positive rearticulations in the following three chapters, and that must be kept in mind so that the positions articulated in those chapters are not taken to be a return to the traditional philosophical project of providing unsurpassable foundations for thought.

Endnotes

1 I, too, following Wittgenstein, invoke the idea in my positive reinterpretation of Derridean concerns in the next chapter.

2 Tyler Burge, whom I discuss below, is an important figure in this turn.

3 Chicago: University of Chicago Press, 1962.

4 Jean-Luc Nancy, The Inoperative Community, ed. Peter Connor, trans. Peter Connor, Lisa Garbus , Michael Holland, and Simona Sawhney ( Minneapolis: University of Minnesota Press, 1991; or. pub. 1986).

5 Ibid., xxxvi. The alternative to the Left that Nancy sees seems to be not the Right that I have alluded to but a more traditional liberal laissez-faire Right. He follows the quote I have just cited with this parenthetical remark: "(On the other hand, 'right' means, at least, that the political is merely in charge of order and administration)." This conception of the Right, however, particularly with its ambiguous word "order," can be interpreted in both communal and individualist ways.

6 This is an idea that the communitarian Michael Sandel attributes to-and criticizes in-John Rawls. Cf. Liberalism and the Limits of Justice ( Cambridge: Cambridge University Press, 1982), esp. chap. 4.

7 Inoperative Community, xxxvii.

8 Ibid., xxxvi.

9 There is, of course, a disagreement within liberal circles about whether to consider the idea of a preconstituted individuality an ontological commitment or a methodological starting point. The former view is more constitutive and the latter more normative. Since Nancy does not make the constitutive normative distinction, it is difficult to make out which of the two interpretations he would set himself against.

10 Inoperative Community, 4. Because of the importance of this argument to Nancy's overall position, it is also worth citing the French original (including the part of the quote I deleted from the translation, which I put in brackets): "L'individualisme est un atomisme inconséquent, qui oublie que l'enjeu de l'atome est celui d'un monde. C'est bien pourquoi la question de la communauté est la grande absente de la métaphysique du suject, c'est-à-dire-individu ou Etat total-de la métaphysique du pour-soi absolu [: ce qui veut dire ausst bien la métaphysique de l'absolu en général, de l'être comme ab-solu, parfaitement détaché, distinct et clos, sans rapport. Cet ab-solu peut se présenter sous les espèces de l'Idée, de l'Histoire, de l'Individu, de l'Etat, de la Science, de l'Oeuvre d'art, etc. Sa logique sera toujours la même, pour autant qu'il est sans rapport.] Elle sera cette logique simple et redoubtable qui implie que ce qui est absolument séparé renferme, si on peut dire, dans sa séparation plus que le simple separe. C'est-à-dire que la séparation elle-même doit etre enfermée, que la clôture ne doit pas seulement se clore sur un territoire (tout en restant exposée, par son bord externe, à l'autre territoire, avec lequel elle communique ainsi), mais sur la clôture elle-même, pour accomplir l'absoluité de la séparation. L'absolut doit être l'absolu de sa propre absoluité, sous peine de n'être pas. Ou bien: pour être absolument seul, il ne suffit pas que je le sois, il faut encore que je sois seul à être seul. Ce qui précisément est contradictoire. La logique de l'absolu fait violence à l'absolu" ( La communauté désoeuvrée [ Paris: Bourgois, 1986], 17-18).

11 Jean-Luc Nancy, The Experience of Freedom, trans. Bridget McDonald ( Stanford: Stanford University Press, 1993; or. pub. 1988), 73.

12 Martin Heidegger, Being and Time, trans. John Macquarrie and Edward Robinson ( New York: Harper & Row, 1962; or. pub. 1927), 89. Cf. idem, Basic Problems of Phenomenology, trans. Albert Hofstadter ( Bloomington: Indiana University Press, 1982; or. pub. 1975), 65 : "Transcendence, transcending, belongs to the essential nature of the being that exists (on the basis of transcendence) as intentional, that is, exists in the manner of dwelling among the extant. Intentionality is the ratio cognoscendi of transcendence. Transcendence is the ratio essendi of intentionality in its diverse modes."

13 Inoperative Community, 104.

14 Ibid., 9.

15 Ibid., 12.

16 Ibid., xxxix.

17 Jean-Luc Nancy and Phillipe Lacoue-Labarthe, "The Nazi Myth", trans. Brian Holmes, Critical Inquiry 16, no. 2 ( 1990): 300.

18 For more on mimetism and its relationship to Nazism, see Lacoue-Labarthe Heidegger, Art, and Politics, trans. Chris Turner ( Oxford: Basil Blackwell, 1990), esp. chap. 8. It is worth noting that this melding of mimetism and community is not, in Nancy's and Lacoue-Labarthe's eyes, confined to the Nazi project. "We wish only to underline just how much this logic, with its double trait of the mimetic will-to-identity and the self-fulfillment of form, belongs profoundly to the mood or character of the West in general, and more precisely, to the fundamental tendency of the subject, in the metaphysical sense of the word" (" Nazi Myth," 312.) Here we can get a good glimpse of the normative character of Nancy's project: to try to articulate a conception of community that is no longer totalitarian. More on this below.

19 Jean-Luc Nancy, L'oubli de la philosophie ( Paris: Editions Galilee, 1986), 95. Translations from this work are my own. I use the term "compearance" to translate comparution in conformity with existing translations of Nancy's work. Although a full discussion of this would be beyond the scope of the present chapter, it is worth noting the internal coherence Nancy sees between what he calls "sense" (which is not to be confused with linguistic signification, but rather is, like differance, the "ground" of its possibility), being, and community. In his introductory essay to Une pensée finie, Nancy writes, speaking of being, that "[i]t is a matter of a diaresis or a dissection of the 'self' that precedes all relationship with the other, as well as all identity of the self. In this diaresis, the other is already the same, but this 'being' is not a confusion, and still less a fusion: it is the being-other of the self insofar as neither 'self' nor 'other,' nor some relationship of the two can be given to it as an origin. It is less and more than an origin: the to-self [a-soi] as appropriation of the inappropriable of to-being [a-être]-of its sense" ( "Une pensée finie", in Une pensée finie [ Paris: Editions Galilee, 19901, 17; my translation).

20 Inoperative Community, 25.

21 Jean-Luc Nancy, "Sharing Voices", in Transforming the Hermeneutic Context: From Nietzsche to Nancy, ed. Gayle Ormiston and Alan Schrift ( Albany: State University of New York Press, 1990; essay or. pub. 1982), 244.

22 Jean-Luc Nancy, "Finite History", in The Birth to Presence, trans. Brian Holmes et al. ( Stanford: Stanford University Press, 1993; essay or. pub. 1990), 155 .

23 Cf., e.g., Jean-Luc Nancy, "Of Being-in-Common", trans. James Creech, in Community at Loose Ends, ed. the Miami Theory Collective ( Minneapolis: University of Minnesota Press, 1991).

24 Experience of Freedom, 69.

25 Inoperative Community, 31. The original French title for The Inoperative Community is La communauté désoeuvrée, which might also be translated as "the unworked community."

26 Ibid., 2.

27 L'oubli de la philosophie, 101.

28 For that reason, I think we should read as among the totalitarians Nancy wants to struggle against those who Nancy cites (without naming them) at the beginning of L'oubli de la philosophie, decry the current "crisis" in philosophy, and want to return to sense, "sense intended as signification... [which] consists in the establishment or in the assignation of presence" (30-31).

29 Experience of Freedom, 55.

30 Ibid., 72.

31 We should also be clear here that it is not individuals who, in their freedom, resist signification. Rather, it is freedom as constitutive of an individual that is this resistance itself.

32 We might recall here that the impossibility of realizing the totalitarian project parallels the impossibility of realizing the project of signification. 33 It is clear that, for Nancy, the idea of a common substance and the project of signification must be internally linked. Otherwise, the critique of the idea of a common substance as totalitarian would be unmotivated. For Nancy, there is an either-or here: either a totalitarianism of common substance (a project of immanence) or sharing. This either-or is structurally the same as Derrida's either-or I discuss in the next chapter: either foundationalism or differance.

33 It is clear that, for Nancy, the idea of a common substance and the project of signification must be internally linked. Otherwise, the critique of the idea of a common substance as totalitarian would be unmotivated. For Nancy, there is an either-or here: either a totalitarianism of common substance (a project of immanence) or sharing. This either-or is structurally the same as Derrida's either-or I discuss in the next chapter: either foundationalism or differance.

34 It might be objected that I am begging the question against Nancy. After all, has he not offered a conception of community that does not refer to values and principles? And do I not, by stipulating that community must refer to values and principles, just define Nancy out rather than argue against him? In fact, I do not stipulate that community must refer to values and principles. If I stipulate anything, it is the more uncontroversial idea that reasons for conceiving community one way or another must refer to values or principles. Thus, I am arguing, not against Nancy's conception of community, but against a community's holding his conception for a reason.

35 In the next chapter, I do not say much about this particular difficulty in generalizing Derrida's analyses of specific texts, in part because I have just said it, and in part because the kinds of texts he does treat (at least the way he reads them) have a constricted enough approach to linguistic meaning that Derrida runs into deep problems before one might even be tempted to worry about generalization.

36 This reading sees Nancy's view of community as directly modeled on the Derridean view of differance. One might want to ask, at this point, why I do not consider this interweaving between sharing and common normative commitments in my consideration of the normative reading of Nancy's conception of community. There, I put the matter starkly as a choice between sharing and common substances. The reason I do not raise it there is that what is at issue at that point is not whether there is in fact a relationship between the two, but which way of thinking about community we should appeal to in order to avoid totalitarianism. For Nancy, this is clearly the normative issue, and his opting for sharing is clearly the only viable answer to be given to that issue. To read Nancy as saying that we ought to give pride of place to both sharing and common substances is, I think, not to be reading Nancy. Nevertheless, one might want to ask, independent of the interpretation of Nancy, whether that is a viable position. To that query, I think my positive rearticulation of community in terms of contingent holism provides a response: namely, that we can appeal to something like common substances-but nonfoundationally conceived-drop sharing, and get a good nontotalitarian conception of community. Otherwise put, sharing, as Nancy articulates it, need not even be introduced to achieve a nontotalitarian view of community. (This does not mean that it cannot be introduced or that it should not be introduced. Rather, it means something like, why bother?)

37 There is another account of the concept of practices that, while developed independently, is similar to-and more detailed than-the one presented here. It appears in chapter 6 of Joseph Rouse Engaging Science ( Ithaca: Cornell University Press, 1995). (As this book was going to press, a fuller treatment of the concept of practices appeared, Theodore Schatzki Social Practices: A Wittgensteinian Approach to Human Activity and the Social [ Cambridge: Cambridge University Press, 1996]. The account offered there parallels closely, although in more depth, the account of practices offered here.)

38 I am grateful to an anonymous reader for clarifying this point, which in turn has led me to revise an earlier definition of a practice.

39 One could worry here that there is another problem attaching to social recognition. Social recognition of a practice may not be recognition of it explicitly as a practice. There is a difference between, say, a recognition, "They're playing chess," and a recognition, "They're engaging in an instance of the practice of chess playing." I do not believe that this difference causes any great difficulties for the account I am trying to give, however, since there seems no reason to bar implicit recognition of practices as practices from being a form of such recognition.

40 For ease of exposition, I will henceforth use the term "practice" to cover both the general practice and its specific instances. I do not think this will occasion any problematic confusion in what follows.

41 Michel Foucault, The History of Sexuality, vol. 1, An Introduction, trans. Robert Hurley ( New York: Random House, 1978; or. pub. 1976), 82

42 Gilles Deleuze and Felix Guattari, A Thousand Plateaus: Capitalism and Schizophrenia, trans. Brian Massumi ( Minneapolis: University of Minnesota Press, 1987; or. pub. 1980), 21 .

43 This conception of an individual's relationship to community has affinities with Honi Fern Haber's concept of "subject-in-community," outlined in the fourth chapter of Beyond Postmodern Politics: Lyotard, Rorty, Foucault ( New York: Routledge, 1994).

44 Especially in "Anti-Oedipus: Capitalism and Schizophrenia", trans. Mark Seem, Robert Hurley , and Helen Lane ( New York: Viking Press, 1977; or. pub. 1972).

45 This view of culture as characterizable in terms of practices seems to inform Kwame Anthony Appiah's book on African identity, In My Father's House: Africa in the Philosophy of Culture ( Oxford: Oxford University Press, 1992).

46 Tyler Burge, "Individualism and the Mental", in Studies in Metaphysics, Midwest Studies in Philosophy, vol. 4, ed. Peter French, Theodore Uehling Jr., and H. Wettsten ( Minneapolis: University of Minnesota Press, 1979), 79.

47 This thought experiment is given and analyzed on pages 77 - 79 of "Individualism and the Mental".

48 Tyler Burge, "Other Bodies", in Thought and Object: Essays on Intentionality, ed. Andrew Woodfield ( Oxford: Clarendon Press, 1982), 109.

49 In setting up his thought experiment, Putnam asks us to assume that the earthling and his doppelgänger "were exact duplicates in appearance, feelings, thoughts, interior monologue, etc." ( The Meaning of 'Meaning,' in Mind, Language, and Reality, vol. 2 of Philosophical Papers [ New York: Cambridge University Press, 1986; or. pub. 1975], 224).

50 The fullest articulation of this argument is given in Tyler Burge, "Intellectual Norms and the Foundations of Mind", Journal of Philosophy 83, no. 12 ( 1986): 697-720, esp. 703-10. See also idem, "Wherein Is Language Social?" in Reflections on Chomsky, ed. A. George ( London: Basil Blackwell, 1989), esp. 181-84.

51 "Intellectual Norms", 703.

52 Andrew Woodfield, "Thought and the Social Community", Inquiry 25, no. 4 ( 1982): 435-50. See also Kent Bach, Thought and Reference ( Oxford: Clarendon Press, 1987), esp. 262-80.

53 Thought and the Social Community, 447.

54 This assumption is also central to Bach's critique. "At any rate, if [the patient in the thought experiment who shares our linguistic situation] misunderstands the word 'arthritis' and does not associate it with the concept of arthritis, there seems to be no reason to suppose that the concept of arthritis figures in his belief" ( Thought and Reference, 267).

55 Were it to imply the former, then there would not be the strict separation between the constitutive and normative questions that I have argued there is. Thus, what follows must be read not as an entailment of the view of community I have outlined, but as one direction in which normative questions about community can be worked out without leading toward totalitarianism.

56 Intellectual Norms, 707.

57 Oddly, Derrida, whose project is a radical questioning of accepted approaches to linguistic meaning, also neglects this social dimension, as we will see in the next chapter.

58 At least, psychology in its normative dimensions, which includes not only psychopathology but perceptual and cognitive psychology, personality theory, child psychology, and so forth.

59 Tyler Burge, "Individualism and Psychology", Philosophical Review 95, no. 1 ( 1986): 3-45.

60 Ibid., 44.

61 Ibid., 25.

62 Michel Foucault, Discipline and Punish, trans. Alan Sheridan ( New York: Random House, 1977; or. pub. 1975), 29.

63 These details occur not only in Discipline and Punish but also in the treatment of psychiatry and psychoanalysis in the first volume of his History of Sexuality and in interviews and discussions during the last years of his life.

64 Discipline and Punish, 160.

65 Although I have only referred to Discipline and Punish here, Foucault offers another, only slightly less direct historical analysis of our perceptual categories in his archaeology of medical perception entitled Birth of the Clinic, trans. A. M. Sheridan Smith ( New York: Random House, 1973). In that text, he articulates some of the ways in which social relations help to constitute the emergence of the perceptual orientation-the medical gaze-which characterizes current-day diagnostic practice. I am indebted to Thomas Flynn for calling my attention to this point.