I welcome this occasion to publicly exchange words, ideas, perhaps arguments with Alain Badiou on the topic of « universalism » and « universality ». This is not the first time we do it in a long story of intellectual companionship, and perhaps, in a sense, this was always our common object, therefore also our point of heresy. But each of us keeps working, and the circumstances lead to emphasizing new aspects.
I am firmly convinced that a philosophical discourse on the categories of the universal, universality, universalism - their meaning and their use - has to be a critical one. Not a simple historical one, listing discourses on the universal, some of which claim to be themselves universalistic, and situating them - but certainly also not simply a discourse endorsing any of them, or trying to add to their already very long list. In this matter we (some of us…) have become cautious, even sceptical, because we have learnt that the gap between theory and practice, principles and consequences, cognitive and performative phrases, is intrinsic to the language of universalism, or more generally, as I prefer to say, to any language that endeavours to "speak the universal", as do precisely our own discourses tonight.  This equivocity takes multiple forms, but particularly the form of identical universalistic enunciations receiving opposite meanings and producing opposite effects depending on when, where, by whom and to whom they are spoken, the form of universalistic discourses legitimizing or instituting exclusions, and more disturbing still the form of universalistic discourses whose categories are build on exclusion - that is, on the denial of otherness or alterity -, but sometimes also the inverted form of particularistic or differentialist discourses becoming the paradoxical premises for the invention of new enlarged forms of universalism and determining its content. It would seem, and I am still waiting for a counterexample, that universalism is never simply doing what it says, or saying what it does. As a consequence what I believe is a task for a philosopher (or a philosopher today, in the current moment) with respect to universality is precisely to understand the logic of these contradictions and, in a dialectical way, to investigate their dominant and subordinated aspects, to uncover how they work and how they can be shifted or twisted through the interaction of theory and practice, or if you prefer discourse and politics. What I exclude therefore - already a gesture of exclusion, or perhaps excluding the exclusionary - is a plea for or against universalism as such, or any of its historical names.
I hope, however, that this kind of critical attitude, which I would very much like to push to a form of "negative dialectics", notwithstanding previous uses of this expression, and whose effects I could certainly not completely anticipate myself, will not be misunderstood here. It does not arise from the fact that I would have become shaky or ambiguous in my commitment to certain determinate forms of universalism. To recall some of their names or key notions : secularism, human rights, democracy, egalitarianism, internationalism, social justice, etc. But I would certainly not find it sufficient or even secure to walk out in the streets or enter into a conference room with such declarations as "I am for secularism" (therefore against religious or cultural communitarianism), "I am for internationalism" (therefore against national allegiance, of which I wrote somewhere that it is not really discernible from nationalism, itself not deprived of universalistic aspects), etc. Or at least not without asking immediately such questions as which secularism ? which democracy ? which internationalism and nationalism ? etc., and also : what for ? Under which conditions ? "Tout tient aux conditions" : conditions are always determinant, as my master Althusser, who certainly was no relativist, used to say. And it is because I want to incorporate some of its conditions (including the negative conditions, or the "conditions of impossibility") within the discourse of universalism, or to put it more philosophically, because I want to outline a discourse of universalism that opens the possibility of incorporating within itself its contradictory conditions, the contradictions that always already affect its conditions, that I adopt a critical and dialectical point of view.
And now, after these preliminaries, both too long given the short time we have been allowed, and too quick not to remain superficial, let me indicate the three orientations that seem to me particularly significant from this point of view : one deals with the dilemmas or dichotomised enunciations of universalism in philosophy ; a second one deals with the intrinsic ambivalence of the institution of the universal, or the universal as "truth" ; finally, a third one deals with what, in a quasi-Weberian manner, I would like to call the responsibility (or responsibilities) involved in a "politics of the universal" to which many of us are committed.
Let me begin with a few words on the dilemmas and dichotomies that, right from the beginning, characterize our disputes over universalism. It is indeed intriguing, but also revealing, that most arguments about universalism, combining logical distinctions with ethical or political choices, take the form of building symmetries, couples or dilemmas of opposite notions, or conceptions, or realizations of universalism. One could suggest that, in fact, the content of the opposition is always the same, at least in modern times, only rephrased to adapt to different contexts, but this is not completely satisfactory because it precisely leaves the "conditions" outside. A dialectical approach, following the example of Hegel in his phenomenology of conflicting universalities , will try to describe these dilemmas in their own terms, taking them seriously in order to discover what is at stake, each time, in their opposition. It will also, on their example, explain why debates about the opposition of the universal and the particular, or a fortiori universalism versus particularism, are far less interesting and determinant than debates opposing different conceptions of the universal, or different universalities, or in fact only cover a strategic defence of one conception of the universal as a "negation" of its opposite, which it presents as the particular.
I am particularly sensitive to this first dialectical issue, because some years ago, I coined myself a distinction between intensive and extensive universalism.  I was particularly interested in the figure of the citizen and the history of the institution of citizenship, with its exclusionary and inclusionary effects. In the modern era citizenship had been closely associated, almost identified with nationality. I would explain that nationalism, but also other forms of universalism in the sense of the suppression or the neutralization of natural and social differences, such as the great religious discourses of redemption, had a double orientation, one of establishing equality or suppressing distinctions, be it really or symbolically, within a certain community based precisely on that suppression, which could be either small or large, depending on the circumstances, and the other one of removing every preestablished limit or borderline for the recognition and the implementation of its principles, ultimately aiming at the creation of a cosmopolitical order, which could be implemented either in a revolutionary manner, from below so to speak, or in an imperialistic manner, from above. And I argued that albeit radically opposed and in fact incompatible, they could both claim to illustrate the logic of universality, or perhaps better said "universalization". Towards the same period, Michael Walzer gave his 1989 Tanner Lectures on Nation and Universe, whose first part was titled "Two Kinds of Universalism", in which he confronted - with a clear preference for the second - a "covering-law universalism", which includes all claims of rights in the same justice, all experiences of emancipation in the same narrative, and what he termed a "reiterative universalism", whose immanent principle would be differenciation, or rather the virtual capacity of moral values and definitions of right to emulate and communicate, in a process of mutual recognition.  Between these two dichotomies, my own intensive vs. extensive and Walzer’s covering vs. reiterative dilemma, there were at the same time obvious affinities and striking discrepancies, which would become more interesting if I tried here to set up a debate and especially feed it with concrete issues, such as the issue of nationalism. But we have no time for that now, and therefore let me simply indicate, in a rather formal manner, that as soon as one really enters the debates on universalism such dichotomies, both symmetric and dissymmetric, or if you prefer both descriptive and normative, become inescapable. They are a good sign of the fact that every speaker (and speech) of the universal is located within, not without the field of discourses and ideologies that he/she/it wants to map.
It cannot be by chance that many, perhaps most discourses on universalism and the universal itself take a refutative form, what the Greeks called an elenchus, saying not so much what the universal is, than what it is not, or not only. There is indeed no meta-language of universality, or the surest way to destroy the universality of a universalistic discourse is to claim that it provides the meta-language of universality, as Hegel already knew. But there are possibilities of shift, and strategic choice, among the categories that grant a specific explanatory or injunctive value to the distinction of antithetic forms of universalism. To classify these categories, and also to show how they can be at the same time very old and periodically renewed, would be to sketch a speculative history of universality and universalities, into which it is tempting to embark, because it could shed more light onto some contemporary controversies.
For instance there is the opposition of true and false universality. A good recent example is provided by Alain Badiou, who at the outset of his essay on Saint-Paul (1997) opposes a true universalism of equality, removing or deposing genealogical, anthropological, social differences such as Jew and Greek, Man and Woman, Master and Slave, whose principle was transmitted by Christianity and later secularized by modern republicanism, and a false universalism, or a "simulacrum" of universalism (but problems could come from the fact that this simulacrum is in a sense much more real, or effective, than the "true"), namely the universalism of the liberal world-market (or perhaps the liberal representation of the world market), which relies not on equality but equivalence and therefore allows for a permanent reproduction of rival identities within its formal homogeneity. This second term pushes the notion of "extensive universalism" towards the extreme : the idea that extensive universalism is an ontological product of extension as such, or territorialization and deterritorialization as such. It has many philosophical ancestors, among which I would emphasize the Rousseauist distinction of the "General Will" and the "Will of All". It would certainly have been strongly objected to by Marx, who spent a good deal of his intellectual life showing, not only that the universality of the market is "real", but also that it is "true", i.e. provides an ontological basis for the juridical, moral and political representation of equality. Interestingly, another influential contribution to current debates on universalism, I am thinking of Dipesh Chakrabarty’s Provincializing Europe. Postcolonial Thought and Historical Difference (2000), also describes what he calls "equivalence", or "commensurability", associated with the "meta-narratives" of value (or labor-value) and progress, as a dominant form of universalism whose results, in fact, contradict its egalitarian claims. But he draws from there opposite consequences. In Chakrabarty’s terminology, "translation" is a generic name for universality, so he would confront "Two Models of Translation". In a strong reliance on a certain romantic representation of the singularity of languages and cultures, he would picture the antithesis of equivalence or translation, which is also a form of universalism, although based on the recognition of the "untranslatable", as heterogeneous, "nonmodern" (rather than postmodern) and "antisociological".
Rather than the antithesis of true and false, what becomes relevant here are the old categories of the One and the Multiple, so that we could speak of a universalism of the One (or unity), and a universalism of the Multiple (or multiplicity), where the essential characteristic of multiplicity is to exceed every possibility of subsumption, therefore of common denomination, or only in the form of "negative denomination". This is a long story that traces back to the conflicts between monotheistic and polytheistic religions in the Ancient helleno-semitic world, but also entirely dominates the oppositions of modern Enlightenment, as illustrated by the "war of universals" between follower’s of Kant’s strongly univocal and indeed monotheistic concept of the universality of the categorical imperative, and Herder’s not only historicist but polytheistic concept of world history, in which unity only exists as the absent cause of the harmonic multiplicity of cultures. Now such antitheses can become shifted theoretically and practically, as I said before, and there is a possibility to show it here, even very schematically. Both Kant and Herder were indeed typical cosmopolitans ; they incarnated the two models of cosmopolitanism that, until today, have dominated uses of this notion. But take a discussion like the one between Derrida and Habermas, for example.  In a deep sense, they are both Kantians, and they both refer to the Kantian definition of the "Weltbürgerrrecht", although we could say that their dispute retrospectively emphasizes a rift within Kant’s discourse itself, as illustrated by the distance between his Religion within the bounds of Reason alone, and his Doctrine of Law. Habermas would define cosmopolitanism as the limit or the horizon of a line of progress (not without its obstacles or resistances) that tends to substitute international relations with a "world domestic policy" (Weltinnenpolitik), that is, not so much a global institutional integration than an institutional exclusion of exclusion. And Derrida would condone the cosmopolitan motto on the condition that it became associated, through such names as "hospitality" or "justice" (or rather "unconditional" hospitality and justice) with a radical critique of the legal foundations of politics. But this did not prevent them from joining efforts after 9/11, not only against a certain form of sovereign unilateralism and generalization of the warlike model of politics, but for a certain construction of the global, transnational and transcultural public sphere, in what I would dare to call a certain "politics of the universal". Old Spinoza would perhaps have seen there an illustration of his idea, as exposed in the Theologico-Political Treatise, that in certain circumstances, or within certain conditions, opposite theoretical premises, or conflicting concepts of the universal, can in practice lead to the same consequences. The reverse is true, indeed.
I want now to allude, and this will have to be telegraphic, to another aspect of the dialectics of universality to which I devoted some attention in the past and more recently. This concerns the institution of the universal, or even the institution of the universal as truth, therefore involving the additional difficulty that it cannot become contradicted from inside, that is, on the basis of its own logic or premises. Not because it would become imposed by some external authority or power that would prohibit the contradiction or the refutation, but because the contradiction is already included in the definition of the universal itself. As we will see, this is closely related to the fact that certain forms of universality at least derive their institutional strength not from the fact that the institutions in which they are embodied are absolute themselves, but rather from the fact that they are the site of endless contestations on the basis of their own principles, or discourse.
Such discussions are meaningless and incomprehensible unless one refers - at least allusively - to some case, and I would not deny that the case I have in mind is both ideologically determined and politically oriented, and perhaps what I say on this basis is only valid for this case. This would mean that the history of universality is in fact only made of singularities. The singular universality on which I reflect is not the Pauline enunciation of the equality of the faithful, later transferred upon the Humans, but it is the somewhat different civic principle, or proposition of "equal liberty" (which I suggested to read as a single name : equaliberty). In English the formula emerges in some "tracts" of the 17th century English Levellers, which is an indication of its close relationship to the ideals of the so-called "bourgeois revolutions". But it has roots in a much older tradition, in Roman Law and moral philosophy, and also, perhaps more significantly (although this involves problems of translation), in the democratic ideals and discourses of the Greek polis. And it produces continuous effects, it becomes reiterated (therefore iterated) until our time within democratic institutions and social movements, both on the liberal and the socialist side. I leave this aside because it would be indeed a very long story. Suffice it to recall the twin formulations of the American and French declarations from 1776 and 1789, which already represent an interesting iteration within the "originary" event, or which inscribe the constitutive reciprocity of equality and liberty (or freedom, or independence) within partially converging, partially diverging contexts. Although I derive much of my understanding of the action of this proposition from Arendt’s discussion of its meaning for the institution of the political, I would not say with her that we have on side a "revolution (or constitution) of liberty" and on the other side a revolution of equality (and "happiness"), but on the contrary, that we have in both cases a strong, and absolute, enunciation of the necessary link between the two concepts, albeit with a permanent tension revealing something like an "impossible" equilibrium.
From the discussions that I devoted to this enunciation , I will recall three ideas :
1) The first is the idea of the refutative structure of the proposition or if you prefer its embodying an elenchus, a "negation of the negation". In constitutional texts the proposition appears as a positive one, asserting that "Men were born free and equal", or were such by nature, birthright, etc. Meaning : only institutional violence can deprive them of these rights. But these formulations arise from revolutions or "insurrections", in the broad sense, and summarize the effect of the insurrection. They are based on the theoretical critique and the practical rejection of vested inequalities or privileges, and relationships of subjection. More precisely they are based on the conviction - in my opinion completely vindicated by history - that you cannot have discriminations without having also subjection (or, in the language of the tradition : "tyranny"), and conversely you cannot have subjection or tyranny without also discriminations and inequalities. As a consequence the political institution, citizenship if you like, has to be grounded on a double rejection, not a single one, more profoundly it embodies the negative link between the two "core values" of citizenship. This was reiterated many times in the history of emancipatory movements, particularly the labor movement, the feminist movement, and the anticolonialist struggles. I want to immediately connect this logical negation with a crucial political fact concerning the power and effectiveness of this form of universalism : far from its many failures and practical limitations, i.e. the fact that in practice states or societies, including so-called "democratic" states and societies, are full of inequalities and authoritarian relationships, destroying the principle itself, it is this very practical contradiction that accounts for its immortality. Discriminated and subjected individuals and groups rebel in the name and for the sake of the principles that are officially valid and practically denied. It is the possibility of the rebellion inherent in the principle, provided it "seizes the masses", as Marx would say, that accounts for the capacity of democracies to survive, at the risk of conflicts or civil wars.
2) Second idea I want to recall : although it has to be instituted (again and again), "Equaliberty" is not an institution like any other. We might say that it is, in modern democracies, the arch-institution, or the institution that precedes and conditions every other institution. It is in this context that Arendt’s profound reflexion on the "right to have rights", not by chance developed in the context of an analysis of the most extreme forms of destruction of human life and their roots in the concept of individual rights instituted by the universalistic nation-states, acquire their full meaning.  "Equaliberty" is a name for the "right to have rights" par excellence, which emphasizes the active side of this notion : meaning in practice, their can be a right to have rights only where individuals and groups do not receive them from an external sovereign power or from a transcendent revelation, but confer this right upon themselves, or grant themselves rights reciprocally. It would be important to develop this idea of a limit-institution, or an institution of the institution itself to discuss its progressive transfer from a "naturalistic" form of the discourse on human rights (men, or humans, are free and equal by nature) to a historical form, in which the universality appears to be grounded not in the essence, but in the contingency of the insurrection itself, or if you prefer the struggle. And it would be important to also relate this limit-situation, essentially manifested in the form and the circumstances of the negation, to the subsequent contradictions affecting a positive institution of equaliberty, or if you prefer of democracy. The entire modern history of democratic regimes and struggles testifies for the difficulty, in fact the internal obstacle that prevents actual institutions, or political regimes, from equally or evenly advancing in the direction of equality and the direction of liberty, or evenly protecting them. The simultaneous destruction is only too frequent. The simultaneous realization is hardly to be observed or only as tendency, as exigency. From which I conclude, not that civic universality is an absurd myth, but precisely that it exists as a tendency, or an effort, a conatus. The driving force within this tendency remains the force of the negative, as beautifully expressed by some philosophical formulas : la part des sans-part (the share of the shareless), in Jacques Rancière, and also what is perhaps his model : le pouvoir des sans-pouvoir in Merleau-Ponty ("Note sur Machiavel"). 
3) Finally, I want to recall a third idea, perhaps the most embarrassing of all, but one without which every discourse on universalism is futile in my opinion : this is the violent side inherent in the institution of the universal. Again I insist on the fact that this violence is intrinsic, not additional, not something that we could blame on the bad will, or the weaknesses or the constraints weighing upon the bearers of the universalistic institution, because they are made its bearers by the institution itself, or its historical movement. I said in the beginning that we had learned that the gaps between theory and practice, all the more unstable when a realization of theory in history and politics is at stake, and above all the perverse effects of exclusion arising from the principle of inclusion themselves, were not accidents. Not something which could lead us to say : try again, and this time we will avoid this dark side of universality. But the intrinsic violence of the universal which belongs to its conditions of possibility also belongs to its conditions of impossibility, or self-destruction, it is a "quasi-transcendental", as Derrida would say. The dark side therefore belongs to the dialectic itself, therefore it belongs to the politics of the universal (an expression that, distancing myself from some contemporary authors, like Charles Taylor, I do not identify with a politics of universality as opposed to the idea of a "politics of difference", because a "politics of difference is also a politics of the universal). Now the violent exclusion inherent in the institution or realization of the universal can take many different forms, which are not equivalent and do not call for the same politics. A sociological and anthropological point of view will insist on the fact that setting up civic universality against discriminations and modes of subjection in legal, educational, moral forms involves the definition of models of the human, or norms of the social. Foucault and others have drawn our attention to the fact that the Human excludes the "non-Human", the Social excludes the "a-social". These are forms of internal exclusion, which affect what I would call "intensive universalism" even more than "extensive universalism". They are not linked with the territory, the imperium, they are linked with the fact that the universality of the citizen, or the human citizen, is referred to a community. But a political and ethical point of view, which we can associate with the idea or formula of a "community without a community", or without an already existing community, has to face still another form of violence intrinsically linked with universality : this is the violence waged by its bearers and activists against its adversaries, and above all against its internal adversaries, i.e. potentially any "heretic" within the revolutionary movement. Many philosophers - be they themselves adversaries or fervent advocates of universalistic programs and discourses, such as Hegel in his chapter on "Terror" in the Phenomenology or Sartre in the Critique of Dialectical Reason, have insisted on this relationship, which is clearly linked to the fact that certain forms of universalism embody the logical characteristic of "truth", i.e. suffer of no exception. If we had time, or perhaps in the discussion, it should be our task now to examine the political consequences that we draw from this fact. I spoke of a quasi-weberian notion of "responsibility".  Responsibility here would not become opposed simply to "conviction" (Gesinnung), but more generally to the ideals themselves, or the ideologies which involve a universalistic principle and goal. A politics of Human Rights in this respect is typically a politics which concerns the institutionalization of a universalistic ideology, and before that a becoming ideological of the very principle that disturbs and challenges existing ideologies. Universalistic ideologies are not the only ideologies that can become absolutes, but they certainly are those whose realization involves a possibility of radical intolerance or internal violence. This is not the risk that one should avoid running, because in fact it is inevitable, but this is the risk that has to be known, and that imposes unlimited responsibility upon the bearers, speakers and agents of universalism.