Public Spaces and Democratic Politics

Chantal Mouffe


What interests me in different uses of ‘public’ is a reference to democratic politics in the sense of the common, publicity or openness and it is the aspect that I want to emphasize in the reflections that I will develop, focusing on the idea of the ‘public space’. As for as I am concerned, what is at stake in this debate is the kind of public space that those who want to foster the radical democratic project should try to establish, a space of deliberation and consensus or a space of agonistic confrontation.

To begin, I need to delineate the theoretical framework which will inform my approach. Its main tenets have been developed in several of my previous publications 1 and here I will limit myself to the aspects that are relevant to my argument about ‘the public’. Let’s start with the distinction I have proposed making between ‘politics’ and ‘the political’. In ordinary language it is not very common to speak of ‘the political’ but I think that such a distinction opens important alleys for reflection and a variety of political theorists are making it. The difficulty, though, is that no agreement

exists among them concerning the meaning attributed to these different terms and that may cause some confusion. Commonalities exist, however, and they can provide some points of orientation. For instance, to make this distinction suggests a difference between two types of approach – political science, which deals with the empirical field of ‘politics’, and political theory, which is the domain of philosophers who enquire not about the facts of ‘politics’ but about the essence of ‘the political’. If we wanted to express such a distinction in a philosophical way, we could, borrowing the vocabulary of Heidegger, say that politics refers to the ‘ontic’ level while ‘the political’ has to do with the ‘ontological’ one. This means that the ontic has to do with the manifold practices of conventional politics, while the ontological concerns the very way in which society is symbolically instituted.

But this still leaves the possibility of a lot of disagreement about what constitutes ‘the political’ and that has important consequences for the way ‘the public’ is envisaged. Some theorists like Hannah Arendt envisage the political as a space of freedom and public deliberation, while others see it as a space of power, conflict and antagonism. My understanding of ‘the political’ clearly belongs to the second perspective. More precisely this is how I distinguish between ‘the political’ and ‘politics’. By ‘the political’ I refer to the dimension of antagonism which I take to be constitutive of human societies, while by ‘politics’ I refer to the set of practices and institutions through which an order is created, organizing human coexistence in the context of conflictuality provided by the political.

The Political as Antagonism
I want to take as the point of departure of my reflection about the public space our current incapacity to envisage the problems facing our societies in a political way. Political questions are not mere technical issues to be solved by experts. Properly political questions always involve decisions which require making a choice between conflicting alternatives. This incapacity to think politically is to a great extent due to the uncontested hegemony of liberalism. ‘Liberalism’, in the way I use the term in the present context, refers to a philosophical discourse with many variants, united not by a common essence but by a multiplicity of what Wittgenstein calls ‘family resemblances’. There are to be sure many liberalisms, some more progressive than others but except for a few exceptions (Isaiah Berlin, Joseph Raz, John Gray and Michael Walzer among others) the dominant tendency in liberal thought is characterized by a rationalist and individualist approach which is unable to adequately grasp the pluralistic nature of the social world, with the conflicts that pluralism entails; conflicts for which no rational solution could ever exist, hence the dimension of antagonism that characterizes human societies. The typical liberal understanding of pluralism is that we live in a world in which there are indeed many perspectives and values and that, due to empirical limitations, we will never be able to adopt them all, but that, when put together, they constitute a harmonious and non-conflictual ensemble. This is why this type of liberalism must negate the political in its antagonistic dimension. Indeed, one of the main tenets of this liberalism is the rationalist belief in the availability of a universal consensus based on reason. No wonder that the political constitutes its blind spot. Liberalism has to negate antagonism since, by bringing to the fore the inescapable moment of decision – in the strong sense of having to decide in an undecidable terrain – what antagonism reveals is the very limit of any rational consensus.

When we examine the different perspectives existing within contemporary liberal thought, we can single out two main paradigms. The first one, sometimes called ‘aggregative’, envisages politics as the establishment of a compromise between competing forces in society. Individuals are portrayed as rational beings, driven by the maximization of their own interests and as acting in the political world in a basically instrumental way. It is the idea of the market applied to the domain of politics which is apprehended with concepts borrowed from economics. The other paradigm, the ‘deliberative’ one, developed in reaction against this instrumentalist model, aims at creating a link between morality and politics. Its advocates want to replace instrumental rationality by communicative rationality. They present political debate as a specific field of application of morality and believe that it is possible to create in the realm of politics a rational moral consensus by means of free discussion. In this case politics is not apprehended through economics but through ethics or morality. In both cases what is left aside by this rationalist approach, be it on the mode of instrumental rationality or communicative rationality, is the crucial role played in the field of politics by what I call ‘passions’, the affective dimension which is central to the constitution of collective forms of identification, identifications without which it is impossible to grasp the construction of political identities. Political identities are always collective identities and this is another reason why liberalism with its methodological individualism is unable to grasp the specificity of the political. In politics we are always dealing with a ‘we’ opposed to a ‘them’ and, as I will show in a moment, this is why antagonism cannot be eliminated.

I contend that it is only when we acknowledge ‘the political’ in its antagonistic dimension that can we pose the central question for democratic politics. This question, pace liberal theorists, is not how to negotiate a compromise among competing interests, nor is it how to reach a ‘rational’, i.e. fully inclusive consensus, a consensus without any exclusion. Despite what many liberals want us to believe, the specificity of democratic politics is not the overcoming of the we/them opposition, but the different way in which it is established. What democracy requires is drawing the we/them discrimination in a way compatible with the recognition of the pluralism which is constitutive of modern democracy.

In developing this point I have found the notion of the ‘constitutive outside’ particularly useful because it unveils what is at stake in the constitution of identity. This term has been proposed by Henry Staten2 to refer to a number of themes developed by Jacques Derrida around notions like ‘supplement’, ‘trace’ and ‘difference’. The aim is to highlight the fact that the creation of an identity always implies the establishment of a difference, which is often constructed on the basis of a hierarchy, for example between form and matter, black and white, man and women etc. Once we have understood that every identity is relational and that the affirmation of a difference is a precondition for the existence of any identity, i.e. the perception of something ‘other’ which constitutes its ‘exterior’, we can understand why politics is concerned with the constitution of a ‘we’ which can only exist by the demarcation of a ‘them’. This does not mean of course that such a relation is necessarily one of friend/enemy, i.e. an antagonistic one. But we should realize that, in certain conditions, there is always the possibility that this we/them relation can become antagonistic. This happens when the ‘them’ is perceived as putting into question the identity of the ‘we’ and as threatening its existence. From that moment on, as the case of the disintegration of Yugoslavia testifies, any form of we/them relation, be it religious, ethnic, economic or other becomes the locus of an antagonism.

Let’s draw a first theoretical conclusion from the previous reflections. What we can assert at this stage is that the we/them distinction, which is the condition of possibility of formation of political identities, can always become the locus of an antagonism. Since all forms of political identities entail a we/them distinction, this means that the possibility of emergence of antagonism can never be eliminated. It is therefore an illusion to believe in the advent of a society from which antagonism would have eradicated. Antagonism is an ever present possibility, the political belongs to our ontological condition and when we envisage the public space this is something that needs to be taken into account.

Politics as Hegemony
Next to antagonism, the concept of hegemony is, in my approach, the other key notion for addressing the question of ‘the political’. To acknowledge the dimension of ‘the political’ as the ever present possibility of antagonism requires coming to terms with the lack of a final ground and the undecidability which pervades every order. It requires in other words recognizing the hegemonic nature of every kind of social order and the fact that every society is the product of a series of practices attempting to establish order in a context of contingency. As Ernesto Laclau indicates, ‘the two central features of a hegemonic intervention are, in this sense, the “contingent” character of the hegemonic articulations and their “constitutive” character, in the sense that they institute social relations in a primary sense, not depending on any a priori social rationality.’3 The political is linked to the acts of hegemonic institution. It is in this sense that one has to differentiate the social from the political. The social is the realm of sedimented practices, that is practices that conceal the originary acts of their contingent political institution and which are taken for granted, as if they were self-grounded. Sedimented social practices are a constitutive part of any possible society; not all social bonds are put into question at the same time. The social and the political have thus the status of what Heidegger called existentials, i.e. necessary dimensions of any societal life. If the political – understood in its hegemonic sense – involves the visibility of the acts of a social institution, it is impossible to determine a priori what is social and what is political independently of any contextual reference. Society is not to be seen as the unfolding of a logic exterior to itself, whatever the source of this logic could be: forces of production, development of the Spirit, laws of history, etc. Every order is the temporary and precarious articulation of contingent practices. The frontier between the social and the political is essentially unstable and requires constant displacements and renegotiations between social agents. Things could always be otherwise and therefore every order is predicated on the exclusion of other possibilities. It is in that sense that it can be called ‘political’ since it is the expression of a particular structure of power relations. Power is constitutive of the social because the social could not exist without the power relations through which it is given shape. What is at a given moment considered as the ‘natural’ order – jointly with the ‘common sense’ which accompanies it – is the result of sedimented hegemonic practices. It is never the manifestation of a deeper objectivity exterior to the practices that bring it into being.

Summarizing this point, every order is political and based on some form of exclusion. There are always other possibilities that have been repressed and that can be reactivated. The articulatory practices through which a certain order is established and the meaning of social institutions is fixed are ‘hegemonic practices’. Every hegemonic order is susceptible of being challenged by counter-hegemonic practices, i.e. practices which will attempt to disarticulate the existing order so as to install another form of hegemony.

As for as collective identities are concerned, we find ourselves in a similar situation. Identities are the result of processes of identifications and they can never be completely fixed. We are never confronted with ‘we/them’ oppositions expressing essentialist identities preexisting the process of identification. Moreover since, as I have stressed, the ‘them’ represents the condition of possibility of the ‘we’, its ‘constitutive outside’, this means that the constitution of a specific ‘we’ always depends of the type of ‘them’ from which it is differentiated. This is a crucial point because it allows us to envisage the possibility of different types of we/them relation according to the way the ‘them’ is constructed.

Which We/Them for Democratic Politics?
Once the ever present possibility of antagonism is acknowledged, one can understand why one of the main tasks for democratic politics consists of defusing the potential antagonism that exists in social relations. If we accept that this cannot be done by transcending the we/them relation, but only by constructing it in a different way, then the following question arises: what could constitute a ‘tamed’ relation of antagonism, what form of we/them would it imply? How could conflict be accepted as legitimate and take a form that does not destroy the political association? This requires some kind of common bond to exist between the parties in conflict, so that they will not treat their opponents as enemies to be eradicated, seeing their demands as illegitimate, which is precisely what happens with the antagonistic friend/enemy relation. However, the opponents cannot be seen simply as competitors whose interests can be dealt with through mere negotiation, or reconciled through deliberation, because in that case the antagonistic element would simply have been eliminated. If we want to acknowledge on the one hand the permanence of the antagonistic dimension of the conflict, while on the other hand allowing for the possibility of its ‘taming’, we need to envisage a third type of relation. This is the type of relation which I have proposed calling ‘agonism’.4 While antagonism is a we/them relation in which the two sides are enemies who do not share any common ground, agonism is a we/them relation where the conflicting parties, although acknowledging that there is no rational solution to their conflict, nevertheless recognize the legitimacy of their opponents. They are ‘adversaries’ not enemies. This means that, while in conflict, they see themselves as belonging to the same political association, as sharing a common symbolic space within which the conflict takes place.

What is at stake in the agonistic struggle is the very configuration of power relations around which a given society is structured, it is a struggle between opposing hegemonic projects which can never be reconciled rationally. The antagonistic dimension is always present. It is a real confrontation but one which is played out under conditions regulated by a set of democratic procedures accepted by the adversaries. An agonistic conception of democracy acknowledges the contingent character of the hegemonic politico-economic articulations which determine the specific configuration of a society at a given moment. They are precarious and pragmatic constructions which can be disarticulated and transformed as a result of the agonistic struggle among the adversaries. Contrary to the various liberal models, the agonistic approach that I am advocating recognizes that society is always politically instituted and never forgets that the terrain in which hegemonic interventions take place is always the outcome of previous hegemonic practices and that it is never a neutral one. This is why it denies the possibility of a non-adversarial democratic politics and criticizes those who, by ignoring the dimension of ‘the political’, reduce politics to a set of supposedly technical moves and neutral procedures.

The Public Space
It is time now to examine the consequences of the agonistic model of democratic politics that I have just delineated for envisaging the public space. The most important consequence is that it challenges the widespread conception that, albeit in different ways, informs most visions of the public space conceived as the terrain where consensus can emerge. For the agonistic model, on the contrary, the public space is the battleground where different hegemonic projects are confronted without any possibility of final reconciliation. I have spoken so far of the public space, but I need to specify straight away that we are not dealing here with one single space. According to the agonistic approach, public spaces are always plural and the agonistic confrontation takes place on a multiplicity of discursive surfaces. I also want to insist on a second important point. While there is no underlying principle of unity, no predetermined centre to this diversity of spaces, diverse forms of articulation always exist among them and we are not faced with the kind of dispersion envisaged by some postmodernist thinkers. Nor are we dealing with the kind of ‘smooth’ space found in Deleuze and his followers. Public spaces are always striated and hegemonically structured. A given hegemony results from a specific articulation of a diversity of spaces and this means that the hegemonic struggle also consists of the attempt to create a different form of articulation among public spaces.

My approach is therefore clearly very different from the one defended by Jürgen Habermas, who when he envisages the political public space, which he calls the ‘public sphere’, presents it as the place where deliberation aiming at a rational consensus takes place. To be sure Habermas now accepts that it is improbable, given the limitations of social life, that such a consensus could effectively be reached and he sees his ideal situation of communication as a ‘regulative idea’. However, according to the perspective that I am advocating, the impediments to the Habermasian ideal speech situation are not empirical but ontological and the rational consensus that he presents as a regulative idea is in fact a conceptual impossibility. Indeed it would require the availability of a consensus without exclusion, of a we without a them, which is precisely what I have shown is impossible.

To support my argument about the impossibility of such a rational consensus I have referred to Derrida and the constitutive outside, but I could also have reached the same conclusion with the help of different thinkers. We can for instance also use Wittgenstein’s insights to undermine Habermas’s conception of procedures and the very idea of a neutral or rational dialogue. In order to have agreements about opinions, there must first be agreement on the language used, and this, as Wittgenstein points out, implies agreement about forms of life. According to him procedures only exist as a complex ensemble of practices. Those practices constitute specific forms of individuality and identity that make possible the allegiance to the procedures. It is because they are inscribed in shared forms of life and agreements on judgments that procedures can be accepted and followed. They cannot be seen as rules, created on the basis of principles, and then applied to specific cases. Rules for Wittgenstein are always abridgments of practices. They are inseparable from specific forms of life. Therefore distinctions between ‘procedural’ and ‘substantial’ or between ‘moral’ and ‘ethical’, which are central to the Habermasian approach cannot be maintained.

Still another way to problematize the very possibility of the notion of the ‘ideal speech situation’ conceived as the asymptotic ideal of intersubjective communication, free of constraints and where the participants arrive at consensus by means of rational argumentation, is to follow the lead of Slavoj Zizek, through Lacon. Indeed a Lacanian approach reveals how discourse itself, in its fundamental structure, is authoritarian since, out of the freefloating dispersion of signifiers, it is only through the intervention of a master signifier that a consistent field of meaning can emerge. For Lacon the status of the master signifier, the signifier of symbolic authority, founded only on itself, is strictly transcendental: the gesture that ‘distorts’ a symbolic field, that ‘curves’ its space by introducing a non-founded violence is in the stricter sense correlative to its very establishment. This means that, if we were to substract from a discursive field its distortion, the field would disintegrate, it would ‘de-quilt’ to speak in Lacanese. This clearly undermines the very basis of the Habermasian view, according to which the inherent pragmatic presuppositions of discourse are non-authoritarian since they imply the idea of a communication free of constraints where only rational argumentation counts.

I also want indicate that, despite the similar terminology, my conception of the agonistic public space also differs from the one of Hannah Arendt which has become so popular recently. In my view the main problem with the Arendtian understanding of ‘agonism’ is that, to put it in a nutshell, it is an ‘agonism without antagonism’. What I mean is that while Arendt puts great emphasis on human plurality and insists that politics deals with the community and reciprocity of human beings who are dif-
ferent, she never acknowledges that this plurality is at the origin of antagonistic conflicts. According to her, to think politically is to develop the ability to see things from a multiplicity of perspectives. As her reference to Kant and his idea of ‘enlarged thought’ testifies, her pluralism is not fundamentally different from the liberal one because it is inscribed in the horizon of an intersubjective agreement. Indeed what she looks for in Kant’s doctrine of the aesthetic judgment is a procedure for ascertaining intersubjective agreement in the public space. Despite significant differences between their respective approaches, Arendt ends up like Habermas envisaging the public space in a consensual way. To be sure, as Linda Zerilli has pointed out, in her case the consensus results from the exchange of voices and opinions (in the Greek sense of doxa) not from a rational ‘Diskurs’ like in Habermas. While for Habermas consensus emerges through what Kant calls ‘disputieren’, an exchange of arguments constrained by logical rules, for Arendt it is a question of ‘streiten’, where agreement is produced through persuasion, not irrefutable proofs. However neither of them is able to acknowledge the hegemonic nature of every form of consensus and the ineradicability of antagonism, the moment of ‘Wiederstreit’, what Lyotard refers to as ‘the differend’. It is symptomatic that, despite finding their inspiration in different aspects of Kant’s philosophy, both Arendt and Habermas have in common that in Kant’s aesthetic they privilege the aspect of the beautiful and ignore his reflection on the sublime. This is no doubt related to their avoidance of ‘the differend’.

The Public as Publikum
To end I would like to share with you some thoughts concerning the relation between the Public as ‘Öffentlichkeit’ and the Public as ‘Publikum’. It is clear that we are not dealing with two preconstituted entities facing each other but that there exists a relation of mutual implication. The very identity of a given public space is a function of its public and reciprocally the identity of the public is at stake in the way the public space is constructed. Since I am focusing here on the political aspect of this relation, the question I would like to address concerns the implications of this discursive construction for the political role that progressive critical arts practices could play.

I want to stress at the outset that when I think about the relation between art and politics, I do not see it in terms of two separately constituted fields, art on one side and politics on the other, between which a relation would need to be established. There is an aesthetic dimension in the political and there is a political dimension in art. This is why I never speak of political art because I consider that one cannot make a distinction between political and non-political art. From the point of view of the theory of hegemony, artistic practices play a role in the constitution and maintenance of a given symbolic order or to its challenging and this is why they necessarily have a political dimension. The political, for its part, concerns the symbolic ordering of social relations, what Claude Lefort calls ‘the mise en scene’, ‘the mise en forme’ of human coexistence and this is where its aesthetic dimension lies.

The real issue concerns the possible forms of critical art, the different ways in which artistic practices can contribute to the questioning of the dominant hegemony. Once we accept that identities are never pre-given but that they are always the result of processes of identification, that they are discursively constructed, the question that arises is the type of identity that critical artistic practices should aim at fostering. Clearly those who advocate the creation of agonistic public spaces, where the objective is to unveil all that is repressed by the dominant consensus, are going to envisage the relation between artistic practices and their public in a very different way from those whose objective is the creation of consensus, even if this consensus is seen as a critical one. I am for this reason very suspicious of the current tendency to promote ‘commemorative’ art because, even when the intention is a critical one, it tends to impose one accepted way of seeing things instead of opening up the debate and facilitating an agonistic confrontation. According to the agonistic approach, critical art is art that foments dissensus, that makes visible what the dominant consensus tends to obscure and obliterate. I do not think, however, that critical art only consists of manifestations of refusal, that it should be the expression of an absolute negation, a testimony of the ‘intractable’ and ‘unrepresentable’. We witness today a certain infatuation with the ‘sublime’ which leads to dismissing the importance of proposing new modes of coexistence, of contributing to the construction of new forms of collective identities. There is too much emphasis on ‘dis-identification’ at the expense of ‘re- identification’. This perspective, while claiming to be very radical, remains trapped within a very deterministic framework according to which the negative gesture is in itself enough to bring about the emergence of a new form of subjectivity. As if this subjectivity was already latent, ready to emerge as soon as the weight of the dominant ideology would have been lifted. Such a conception is in my view completely anti-political. It fails to come to terms with the nature of the hegemonic struggle and the manifold of practices, discourses and language games through which identities are constituted. I am convinced that it is only by recognizing the need for a plurality of forms of interventions, taking place in a variety of public spaces, that critical artistic practices can contribute to the constitution of a variety of agonistic spaces where a radical and plural conception of democracy could be fostered.


1. Ernesto Laclau and Chontal Mouffe, Hegemony and Socialist Strategy. Towards a Radical Democratic Politics (London: Verso, 1985); Chantal Mouffe, The Return of the Political (London: Verso, 1993); Chantal Mouffe, The Democratic Pradox (London: Verso, 2000)

2. Henry Staten, Wittgenstein and Derrida (Oxford: Basil Blackwell, 1985).

3. Ernesto Laclau, Emancipation(s) (London: Verso, 1996), p.90.

4. This idea of ‘agonism’ is developed in my book The Democratic Paradox, op. cit., chapter 4 (see note 1). To be sure, I am not the only one to use that term and there are currently a variety of ‘agonistic’ theorists. However, they generally envisage the political as a space of freedom and deliberation, while for me it is a space of conflict and antagonism. This is what differentiates my agonistic perspective from the one defended by William Connolly, Bonnig Honig or James Tully.

Chantal Mouffe is Professor of Political Theory at the Centre for the Study of Democracy at the University of Westminster in London.