Contesting representation

Introduction

In recent philosophical debates about political representation, the scope of this concept has significantly broadened. Representation used to be (and in certain parts of the literature still is) understood in terms of a technical device for decision-making processes in groups that cannot possibly meet face to face and which therefore require a form of delegation between clearly defined persons or groups. Hanna Pitkin’s classic text on the subject is but one example of understanding representation in terms of such a relation between constituted entities (Pitkin 1972). A great part of the recent literature, in contrast, underlines the constitutive and ubiquitous nature of representation. Stressing the aesthetic and symbolic dimensions of representation, it pays more attention to the role of representation in constituting political communities or identities (Ankersmit 1997). Rather than being a relation of delegation or trusteeship between existing entities, representation is now increasingly seen as a construction of the represented, as a form through which the invisible is made visible. In his pathbreaking book on the ‘representative claim’, Saward stresses, for example, that representation constructs the represented, and that the politician therefore becomes an artist, a portrayer of the represented (Saward 2010: 16).

This has opened up a new way of approaching political representation, moving away from the old conception of representation as the alienation of a people that pre-exists representation (e.g. Rousseau 1966: 134). According to this new approach, representation ‘does not reproduce the visible but makes visible’ (Saward 2010: 174). As a result of this, representation can no longer be restricted to a particular type of relation between constituted entities, but in a certain way becomes ubiquitous (Saward 2010: 79). That is not merely because representative claims are increasingly being made outside the sphere of conventional representative democracy, but also because of the conceptual changes that the notion of representation has undergone itself and that make it possible to see representation or representative claims where they were formerly perhaps not visible.

This shift in understanding representation has effects on how to conceive of democracy. To the extent that representation somehow becomes co-extensive with the use of language, the notion of representation loses its exclusive link with the problematic of the relation between state and society, in terms of which it used to be and sometimes still is understood (e.g. Weymans 2005). As a result, the concept of representative democracy has somehow become a tautology (Näsström 2006). If representation is ubiquitous, there can be no democracy or politics without representation. This also entails that it no longer makes sense to oppose parliament as a form of representative democracy to institutional forms of so-called ‘direct democracy’, such as councils. Both are inevitably representative. It also means that members of parliament no longer have a monopoly on representation as representative claims can be made everywhere. Furthermore, to the extent that representation is increasingly understood as a claim or a construction, this also broadens the space for contestation and counterclaims (Saward 2010).

This triggers an important question: how can we reconceive of democratic representation when the concept of representation has undergone such shifts? When representation is constitutive, democratic representation can no longer be exclusively understood in terms of its representativity with regard to a given constituency. A representation can no longer be contested because it would inaccurately portray a pre-existing reality. Even notions such as authorisation and accountability have to be rethought.

A whole field of conceptual elaboration is thus opened. In this paper, I will try to show the tremendous importance of Jacques Rancière’s contribution to this debate. Rancière never systematically elaborated upon the concept of representation. When he used the notion, it was often in the context of a critique of representative government as an oligarchic form. However, he also rejects traditional understandings of ‘direct’ democracy and dismisses attempts to dissolve representations into underlying ‘realities’ and to understand political subjects as ‘real’ (Rancière 2004a). Moreover, he considers politics to be essentially of a theatrical nature. Here it will be argued that the problematic of representation is absolutely central to his political philosophy. Some of his most important concepts, such as the distribution of the sensible, the political, the police, and disagreement, circle around this problematic.

Rancière’s philosophy also sheds new light on questions about what constitutes the specificity of democratic representation, how representations are contested, or what it means to radically take the aesthetic dimension of representation seriously. As will be demonstrated, Rancière shifts the focus from the relation between representative and represented to the relation between the distribution of the sensible as a space of representability and its disruption or contestation. As a result of this theoretical shift, Rancière also provides new ways to criticize or contest the present state of what is called ‘representative democracy’. This contestation has lost its evidence of late. As Plotke has stated, ‘the opposite of representation is not participation. The opposite of representation is exclusion’ (1997: 19). As a consequence, the single most important way to contest representative government is by pointing to forms of exclusion. However, as liberal consensual politics is supposedly all-inclusive (at least in principle), the resources of critique seem to be quickly evaporating. If we are all in favour of inclusion, is it still possible to formulate a principled critique of representative government that goes beyond pragmatic considerations about how it is put into practice? As I will argue, Rancière’s philosophy provides valuable elements to reformulate such a critique.

At the same time, I will argue that reading Rancière through the lens of the problem of representation can clarify a number of difficult issues in his work. In the first place, it enables us to go beyond his focus on pedagogy, and to underline his importance for contemporary debates in political philosophy. Although Rancière never systematically deals with the concept of representation, it nevertheless forms a kind of red thread throughout his oeuvre, from his early work about the impossibility of the intellectuals speaking for the masses and the unrepresentability of the working class (Rancière 1974, 1983, 1989), until his recent work on the distribution of the sensible (2007b) and his theatrical understanding of political action (1999, 2006, 2007a).

In the following, I will first shortly discuss Rancière’s early criticisms of representation, before engaging with his recent political-philosophical work on politics and democracy that elaborates upon some of his early insights on the unrepresentability of the working class. The central concepts of this latter work are those of the ‘distribution of the sensible’ and of the ‘part of those who have no part’. I will try to show that these provide the building blocks of a new approach to representation, and even make it possible to think anew of what democratic representation can mean.

Representation and its discontents

Rancière’s early work already contains a critique of the logic of representation. It is present in the distance he took from Althusser’s conception of the role of intellectuals in guiding the proletariat (Rancière 1974), and from the Maoists’ claim to represent the ‘true’ identity of the proletariat against the intellectuals (Rancière 1989: xviii). Rancière quickly understood that the Marxist claim to represent the proletariat could function as a repressive device: it turns the ‘proletariat’ into a normative category which tends to ‘purify’ and discipline the class. Rancière’s search for an alternative through his archive studies on the early workers’ movement only confirmed this diagnosis. The logic of representation is such that it represses the difference between the would-be representatives and the workers themselves. Rancière soon became keenly aware of the fact that there was no ‘proletarian’ standpoint from where the logic of representation could be contested as such. Rancière’s only alternative was to deconstruct workers’ representations by playing them off against each other: ‘there is no voice of the people. There are broken, polemical voices, which each time divide the identity they are staging’ (2003: 11). Similar to what Laclau would later write about the Lumpenproletariat as a figure of excess that is not merely marginal but says something of the essence of political subjects (Laclau 2002, 2005: 144), Rancière states: ‘this aleatory population [of Marx’s Lumpenproletariat for example] represents less the army of the marginal or declassed than the proletariat in its very essence’ (1989: 147). As a result of the heterogeneity of its voices, the proletariat somehow deconstructs its own representations. It becomes an impossible object of representation in the then prevailing meaning of the term (Rancière 1983).

This ‘unrepresentability’ has important implications. It puts Rancière before the task of developing a totally new way of conceiving workers’ emancipation, which can no longer be conceived in terms of a correct political representation of a social bond. His struggle with the insidious logic of representation is comparable to Claude Lefort’s trajectory. Both dismiss the idea that one can adopt a pre-representational ‘standpoint of the proletariat’, as this only leads to a most dangerous type of representation. However, while Lefort (1963) acknowledges the inevitability of representation and the importance of its contestation for democracy, Rancière takes a more radical turn. Although he acknowledges that dissolving representation into the real is impossible, he refuses to give in to the logic of representation. Its contestation must entail more than the juxtaposition of one representation to another. For Rancière, politics is essentially about what makes representation possible in the first place, about the (im)possibility to make certain things visible. Central to his later approach to the problem of representation is what he calls the ‘distribution of the sensible’, a concept that has a similar function to notions such as ideology and hegemony in Marxism or discourse in Foucault, in that it delineates a configuration that sets limits to what is sayable or visible. This concept allows him to fundamentally recast his critique of representation.

Representative government and the distribution of the sensible

While Lefort concludes his reflections on the problem of representation with a strong argument in favour of representative democracy, the older Rancière remains much more critical of representative democracy (as it has traditionally been understood, i.e. as parliamentary government). According to him, ‘(d)emocracy first of all means this: anarchic “government,” one based on nothing other than the absence of every title to govern’ (2006: 41). Contemporary representative government, he argues, reintroduces a system of titles to govern through mechanisms of elite selection (2006: 42). Therefore, what is today called democracy should better be called the ‘parliamentary system’ or the ‘pluralist constitutional system’, following Raymond Aron (2006: 54). More generally, he thinks representation is not a practical device invented because growing populations made it impossible to gather all citizens. Instead, it is, ‘by rights, an oligarchic form, a representation of minorities who are entitled to take charge of public affairs’ (2006: 53). From its very origin, therefore, representation (understood in the narrow sense of the principle of representative government) was ‘the exact contrary of democracy’ (2006: 53).

At the same time, however, he does not advocate what is often called direct democracy as an alternative either: ‘It is just as false to identify democracy with representation as it is to make the one the refutation of the other’ (2006: 54). He stresses, rather, that so-called democratic states do not rest upon one singular logic but are inevitably composed of a conflictual mix of oligarchic and democratic tendencies. What we usually call democracy was initially founded on the privilege of ‘natural’ elites, but has gradually been transformed by democratic struggles. Universal suffrage, for example, is

a mixed form, born of oligarchy, redirected by democratic combats and perpetually reconquered by oligarchy, which puts its candidates, and sometimes its decisions, to the vote of the electoral body, without ever being able to rule out the possibility that the electoral body will behave like a population that draws lots. (Rancière 2006: 54)

This critique of representation is grounded in a deeper philosophical reflection on its very condition, namely what he calls the distribution of the sensible [le partage du sensible]. He defines this as

the system of self-evident facts of sense perception that simultaneously discloses the existence of something in common and the delimitations that define the respective parts and positions within it. A distribution of the sensible therefore establishes at one and the same time something common that is shared and exclusive parts. This apportionment of parts and positions is based on a distribution of spaces, times, and forms of activity that determines the very manner in which something in common lends itself to participation and in what way various individuals have a part in this distribution. (Rancière 2007b: 12)

Rancière elaborates the concepts of ‘having a part’ and of ‘partage’ through a discussion with Plato and Aristotle who, each in his own way, argue that certain parts of the population have no part in the act of governing and being governed: slaves, who do not possess the language of the rulers, in the case of Aristotle, and artisans, who have no time to be involved in governing, in Plato. From the outset, the distribution of the sensible and the way it configures the parts and positions has political relevance. It ‘reveals who can have a share in what is common to the community based on what they do and on the time and space in which this activity is performed’ (2007b: 12).

At the same time, it functions as a configuration that limits what is sayable and visible. Therefore, it is a kind of historical a priori (Ruby 2009: 53), in the Kantian sense of the forms determining what presents itself to sense experience. It is ‘the most basic system of categorization through which we perceive and intuitively classify the data provided to our senses’ (Citton 2009: 120). This is of central importance to politics, according to Rancière, as politics ‘revolves around what is seen and what can be said about it, around who has the ability to see and the talent to speak, around the properties of spaces and the possibilities of time’ (Rancière 2007b: 13).

Rancière thus sheds a new light on the question of representability. Given the effectiveness of the distribution of the sensible, the question of the ‘correct’ representation of the social is a deadlock. There is no ‘objective’ condition of social realities, but only a ‘partial and partisan, provisional and polemical tying of fragments of experience and of forms of symbolisation’ (2003: 13). However, this provisional and polemical nature of what is sayable and visible is organized and stabilized by what Rancière calls the ‘police’. The latter should not be understood as a repressive force, according to him, but as ‘a form of intervention that prescribes the visible and the invisible, the sayable and the unsayable’ (Rancière 1998: 211). This concept of the police is partly inspired by the work of Foucault on the origin of the police in the 17th century. Rancière’s concept of the police differs from Foucault’s, but shares its rejection of a purely ‘repressive’ understanding of it. Rancière does not understand the police in terms of a logic of governing and the disciplining of bodies, as Foucault does, but sees it as the force that governs their appearances and the distribution of their occupations (1999: 29). In other words, it organizes and reproduces a particular distribution of the sensible. The police are:

an order of the visible and the sayable that sees that a particular activity is visible and another is not, that this speech is understood as discourse and another as noise. It is police law, for example, that traditionally turns the workplace into a private place not regulated by the ways of seeing and saying proper to what is called the public domain, where the worker’s having a part is strictly defined by the remuneration of his work. (Rancière 1999: 29)

The police governs this distribution in a particular way: it turns it into a configuration where each and everyone has her place. The police order is not only an order of inequality, as all social orders inevitably are as a result of the necessity of a certain partage. It is also and especially an order whose basic principle is the absence of a void or a supplement. It is a configuration that appears as full, well-founded and just. In contrast, politics and democracy are about the disruption of this distribution of the sensible by the manifestation of a ‘part of those who have no part’ that demonstrates the equality of anyone with everyone, and thus reconfigures what is visible and sayable.

I will come back to these issues below. What is of interest here is that Rancière’s conceptions of the distribution of the sensible and of the police enable a more profound critique of the concept of representation. The distribution of the sensible is nothing but the space within which representation, as it was traditionally understood, can take place. The Rancierian critique, therefore, particularly focuses on what I would call the ‘police conception of representation’, which entirely takes place within the existing order of the visible, composed of fully present and visible parts and identities. Only within this configuration can the classical questions be asked about authorization and accountability (which assume there is a ‘part’ that authorizes or to whom the representatives should be accountable) or about the ‘representativity’ of representatives (i.e. the extent to which they express a broad range of interests and opinions present in a population). Only what can appear within the configuration of the visible can be represented in this way. In other words, the classical criteria for distinguishing democratic from non-democratic representation are fully situated on the side of the police order in Rancière’s philosophy.

If that is the case, and democracy is what disrupts the police order, the notion of democratic representation that we think can be reconstructed on the basis of this philosophy will require totally different criteria. Although the police establishes a particular space of representability, it is possible, according to Rancière, to disrupt it and change its ‘ways of doing, ways of seeing and ways of speaking’, As he states,

the factory, street or theatre are forms of this distribution in which the economic, the political and the ideological incessantly change roles, in order to define a certain conflictual relation of space and time, of ways of being and ways of doing, of the visible and the sayable. That is also why there are always several kinds of worker in the factory, several kinds of passage in the streets, several peoples in a theatre. (Rancière 2003: 13)

One can thus reoccupy a space and resymbolize it. This is what Rancière is interested in: changes in the distribution of the sensible as a result of workers taking the word into the factory, the street becoming a theatre, the theatre becoming a tribune. The crucial question then becomes how one can withdraw from and subvert the existing distribution of the sensible.

The part of those who have no part

One basic operation of representation that constitutes a distribution of the sensible is ‘counting’ the ‘parts of the population’. This always entails a miscount, according to Rancière, and that gives rise to politics, which is the manifestation of ‘a part of those who have no part’. ‘There is politics’, he states, ‘when there is a wrong count of the parts of the whole’, and thus, ‘when there is a part of those who have no part, a part or party of the poor’ (1999: 10–11).

This notion of ‘the part of those who have no part’ is one of the most crucial but at the same time most enigmatic in Rancière’s philosophy. It is about more than mere exclusion. If the miscount would only mean that someone or some group has not been counted, the solution would be easy: count again, and include the excluded. If democratization would be about such a type of inclusion, its endpoint is always already known in advance: it is a question of counting correctly what is already there (Bingham and Biesta 2010: 82). Moreover, in its contemporary form, the police, whose function it is to maintain a particular count of the parts and parties, already functions in an ‘inclusive’ way. Its hallmark is consensus, which is nothing but a particular distribution of the sensible: a world of which it is thought that everything is visible, and all are included, a world which supposedly counts all its parts without there being a surplus. This police regime of consensus did not exist yet in the 19th century, when there was still a clear demarcation line between the world of logos and that of pure noise. This line could be symbolized by the excluded and become the object of struggle. Today, however, everything and everybody is thought to be already included.

According to the law of consensus, who is a subject (of rights for instance) is always already determined. For example, ‘it is a representation of the community proper to the police that identifies citizenship as a property of individuals’ (Rancière 1999: 31), although for individuals to have become such subjects, a political event has been necessary that made the individual into an object of litigation and thus shifted the order of the visible. Consensus is a way of getting rid of this dissensual stage and replacing it with a configuration of ‘real partners, social groups, identity groups, and so on’ (Rancière 2004b: 306). The contemporary discourse about governance is a case in point. It limits what exists to what is counted, without a surplus. Everybody is included, is a part, has its place, and is turned into a partner: ‘there are only parts of society – social majorities and minorities, socio-professional categories, interest groups, communities, and so on. There are only parts that must be converted into partners’ (Rancière 1999: 14).

Can there then still be a part of those who have no part, and how should we conceive of it? Of course, it can be a part that has been rendered invisible by the partition or distribution of the sensible itself. However, it seems to be possible that even a part that is a part of this distribution can manifest itself as a ‘part of those who have no part’. My contention is that ‘having no part’ can then only mean ‘having no part in the act of this partitioning itself’.

Todd May has interpreted the concept of the part of those who have no part in terms of the difference between passive and active equality, which he discusses in the context of the debate about distributive justice between Nozick, Rawls, Sen and Young (May 2008). Even in the most radically egalitarian theories of distributive justice, he argues, there remains a separation between a distributor and a passive receiver. The equality that results from it is passive: everybody receives an equal share, without participating equally in the act or process of sharing itself. This puts us on the right path, although Rancière does not focus on distributive issues. Rancière’s understanding of the part of those who have no part follows from the logic of representation itself. Each representation of the community, or each counting of its parts and parties, entails a gap, for example ‘the gap between the part of work as social function and the having no part of those who carry it out within the definition of the common of the community’ (Rancière 1999: 36). In this sense, the partition [partage] of the sensible has a double meaning: it both installs a community and a separation (Rancière 1999: 26). Through the act of partitioning, a part arises that has no part in this very act. The part of those who have no part can surely be the ones who are not counted, or who are not accounted for (as was the case in the 19th century). But if that would be all, contemporary liberal democracy would have no such part. Therefore, it must also be understood as the part of those who have no part in the counting itself.

Again, the comparison with Lefort is interesting. Lefort shows that each time someone engages in political action or speech, she inevitably separates herself from the people in whose name she speaks or acts. Each representation thus installs a division in society (Geenens 2006). Something similar is at stake in Rancière. From the moment a community is instituted, there is a partition and a count is made of its parts and parties. However, political conflict is not merely a confrontation between these heterogeneous parts of the population. Its object, rather, is the representational act of counting itself. The miscount, therefore, is not the mere consequence of a part being excluded or rendered invisible (although that can also be part of what it means to have no part), but resides already in the very act of counting, namely in the fixation of positions, parts and roles and the resulting obfuscation of the contingency and equality which, as we will see below, form the very (non-)ground of each social order. Politics, then, is about making this contingency and equality visible by constituting a political subject that disrupts the distribution of the sensible.

Disagreement

Politics, according to Rancière, entails a very peculiar kind of conflict. He calls it ‘une mésentente’, translated into English as ‘disagreement’. It is a very particular kind of conflict between those who say there is conflict and those who say there is not: ‘From Athens in the fifth century B.C. up until our own governments’, Rancière states, ‘the party of the rich has only ever said one thing, which is most precisely the negation of politics: there is no part of those who have no part’ (1999: 14). According to what Rancière calls ‘the rich’, everybody has her place and her part, no conflict is needed.

However, disagreement is about more than that: its object is the very question whether ‘the ones who say there is conflict’ constitute speaking beings at all. Disagreement, as Rancière understands it,

is less concerned with arguing than with what can be argued, the presence or absence of a common object between X and Y. It concerns the tangible presentation of this common object, the very capacity of the interlocutors to present it. An extreme form of disagreement is where X cannot see the common object Y is presenting because X cannot comprehend that the sounds uttered by Y form words and chains of words similar to X’s own. This extreme situation – first and foremost – concerns politics. (Rancière 1999: xii)

Rancière gives the example of somebody asking ‘do you understand?’ after giving an order (1999: 44). Of course, this is not a question to know whether the receiver of the order has understood its contents correctly. The question actually suggests that ‘[i]t’s not up to you to understand; all you have to do is obey’ (1999: 45). The question is a performative one, it draws a line between different (and inevitably unequal) categories of speaking beings. This is a very paradoxical situation. When one gives an order, one assumes the receiver understands it and is thus equal to the commander in terms of her linguistic capabilities. But through this question, an inegalitarian partition is established (in this case between a superior and an inferior). A conflict can then arise that has a very complex structure. Although the inferior starts a conflict which the superior will probably deny, the inferior at the same time affirms that there is a common language and a common world, and that, as a consequence, mutual understanding is possible, while the superior’s question cited above paradoxically suggests there is nothing to be understood. This was the structure of the conflict the early 19th-century workers Rancière studied were engaged in. They combine a kind of Habermasian ‘communicational intervention’ with a ‘strategic intervention, shifting the power struggle that determines whether utterances can be received as arguments on a common stage’ (Rancière 1999: 55). While the employers affirmed consensus and denied the existence of conflict, they can only do that by not taking the logos of the workers into account. As the workers contest this and engage in conflict, they at the same time affirm the existence of a common world and a common language.

Disidentification and equality

As stated, in some of his earlier work Rancière developed a concept of the unrepresentability of the working class. In his later writings, he retains a moment of unrepresentability, but locates it in the first place in the disagreement itself and relates it to how a particular distribution of the sensible functions. It is not ‘differences’ or ‘singularities’ that are unrepresentable, Balibar states when discussing Rancière's work, but ‘the conflict itself’ (Balibar 2009: 104). And, we should add, the conflict is not intrinsically unrepresentable, but only within the distribution of the sensible which is the object of the conflict. The question is no longer whether the people (or the working class) is or is not representable as such, but about how such a regime of representability functions and how it can be contested.

This last question is not an easy one. If a distribution of the sensible makes certain things non-representable, how can a part that has no part then manifest itself and contest this distribution? This is what Rancière attempts to think with the help of three concepts: the affirmation of equality, disidentification and impossible identification. Together, they form the process of what he calls ‘political subjectification’. This is ‘the production through a series of actions of a body and a capacity for enunciation not previously identifiable within a given field of experience, whose identification is thus part of the reconfiguration of the field of experience’ (Rancière 1999: 35).

Political subjectivity is not an affirmation of an identity that is part of the police order of representations. Rancière describes the difference between both with the help of an exemplary dialogue between the revolutionary Auguste Blanqui and a French magistrate in 1832. When the judge asked Blanqui what was his profession, Blanqui answered ‘proletarian’. This is not a profession, the judge stated, remaining within a police logic. ‘It is the profession of thirty million Frenchmen who live off their labour and who are deprived of political rights’, Blanqui replied (Rancière 1999: 37). What interests Rancière is that Blanqui disidentifies with the social category (of workers) he is supposed to belong to. He thereby opens a space for ‘the class of the uncounted that only exists in the very declaration in which they are counted as those of no account’ (Rancière 1999: 38). In a certain way, the part of those who have no part manifests itself through a disidentification with the social roles and positions attributed by a police order.

Political subjectification does not happen through an identification as a ‘class’. even though Rancière continues to use the names of ‘proletariat’ and ‘the poor’. However, he understands the poor as ‘not really the poor. They are merely the reign of a lack of position’ (Rancière 1999: 13–14). The proletariat are those who do not belong to the order of classes, but announce its virtual dissolution in the name of equality. Political subjectification, therefore, is a movement of declassification or disidentification. It is a ‘removal from the naturalness of a place, the opening up of a subject space where anyone can be counted since it is the space where those of no account are counted, where a connection is made between having a part and having no part’ (Rancière 1999: 36). Through disidentification, the miscount is made visible.

The contestation of representation (in casu, of the count of the parts) thus takes place by withdrawing from the very space of representability, and according to Rancière, this requires the affirmation of an equality of anyone with everyone as speaking and thinking beings. It is because ‘the democratic man’ is ‘a poetic being’ that he is ‘capable of embracing the unreality of representation’, Rancière states (2007a: 51). Crucially, this equality is not a demand or a programme to be realized, but merely a postulate (cf. Badiou 2006: 144). It is simply declared, in order to produce practical effects. A case in point is Rosa Parks, who did not demand equality but acted on the assumption of equality when she refused to give her place in the bus to a white passenger in 1955 in Montgomery. This affirmation of equality not only subverts the representations that are constitutive of a distribution of the sensible, but opens a scene where something new can appear. Importantly, in order to become visible, political subjects have to stage their appearance in a very particular theatrical way that can be said to constitute a democratic counterpart to the police conception of representation.

Democratic representation

‘Democracy is not a political regime’, Rancière writes in the fourth of his ‘Ten Theses on Politics’ (1998: 231). It is an act of political subjectification that disrupts the police order and modifies the field of the visible (1999: 99). More in particular, ‘[d]emocracy is the designation of subjects that do not coincide with the parties of the state or of society, floating subjects that deregulate all representations of places and portions’ (1999: 99–100). At first sight, political subjectification seems to be about reconfiguring the distribution of the sensible through a process of self-representation: ‘those who have no right to be counted as speaking beings make themselves of some account’ (Rancière 1999: 27). Rather than letting others (mis)count them, they elaborate their own representations. If that would be Rancière’s position, however, he would remain very close to the classical Marxist perspective, in which a social category (the workers) becomes conscious of itself, creates its own representations and disrupts bourgeois ideology.

The Rancierian account of democratic representation is more sophisticated. What is at stake in political litigation, according to him, is demonstrating the difference of the people from itself, and consequently, manifesting the contingency of every social order. What could be called ‘democratic representation’ (a notion that Rancière does not use himself) introduces a ‘supplement’ to the parts and parties of the community. To the ‘exact names’ of the police that assign people to their place and work it opposes ‘improper names’ that articulate a wrong (Rancière 1998: 121). The resulting difference of the people with itself is not the difference between the ideal people inscribed in foundational texts and the ‘real’ people. It is the difference between a configuration of counted parts and the manifestation of a demos, which

attributes to itself as its proper lot the equality that belongs to all citizens. In so doing, this party that is not one identifies its improper property with the exclusive principle of community and identifies its name – the name of the indistinct mass of men of no position – with the name of the community itself. (Rancière 1999: 8–9)

In order to think the specificity of the representation that is taking place here, Rancière takes recourse to the metaphor of the theatre (cf. Hallward 2006): ‘[i]t is a matter of interpreting, in the theatrical sense of the word, the gap between a place where the demos exists and a place where it does not, where there are only populations, individuals, employers and employees, heads of households and spouses, and so on’ (Rancière 1999: 88).

Democratic representation thereby fully acknowledges the ‘as if’ that is part of its theatrical effect. A political subject pretends to be something in order to become it (Davis 2010: 86). Politics means acting as if one is equal in order to really become it. The logic of the police, in contrast, radically goes against this theatrical staging: it is the denial that there is something to be seen. While politics, for example, transforms the street into a space of the manifestation of a subject (the people, the workers, the citizens), the police restores it into a space of mere circulation (Rancière 1998: 242).

As political subjectification is based on disidentification, what is staged is not simply an alternative identity but an ‘interval’ or a gap between identities. The most important gap is the one between a ‘we’ that speaks and a ‘people’ in whose name this ‘we’ speaks. ‘“We are the people”, that does not mean: “we are the masses”, “we are its representatives”’, Rancière (2004a) states in an interview. It means that a group of individuals constitutes a relation between a ‘we’ and the people, a relation between two subjects, and stages this relation as a gap. For Rancière, ‘politics is never an affair of identity. It always stages a gap. And when one says “we are the people,” I would say exactly that we and the people are not the same thing; politics is constituted by the gap between both.’

Disidentification and the staging of a gap can in particular take place by taking up the cause of an other. One example Rancière gives is his formative experience as a young activist supporting the Algerian war of independence in the early ’60s. The basic political act of the solidarity movement, he states, was one of disidentification with the French state and of staging an interval between identities, in-between being French and Algerian (Rancière 1998: 120). A couple of years later, he had chanted together with many other French students that they were ‘all German Jews’, thus expressing their solidarity with Rudi Dutschke. Each time, an ‘impossible identification’ was at stake, of which the identification with ‘the wretched of the earth’ was possibly the most exemplary manifestation.

For Rancière, taking up

the cause of the other as a political figure is first and foremost this: a disidentification with regard to a certain self. It’s the production of a people that is different from the people that is seen, spoken, counted by the state, a people defined by the manifestation of a wrong done to the constitution of the common which itself constructs another space of community. (Rancière 1998: 212)

The same can happen in the political subjectification of the proletariat: it is by taking up the cause of an other that the proletariat can become a name of the universal. Through such a gesture, a subject becomes ‘a figure of citizenship’ (Rancière 1998: 219–20). Similarly, it is by attributing to itself ‘as its proper lot the equality that belongs to all citizens’ that the demos is constituted (Rancière 1999: 8).

The division of the people appearing through such staging is called by Rancière a form of ‘democratic supplementation’ (Rancière 2006: 77–8). It adds an element that disrupts the clear distribution of parts and parties of the community. To the existing titles to govern (based on tradition, possession, expertise etc.), democracy adds ‘a supplementary title, one common to those who possess all these titles but also to those who do not possess them’ (2006: 46). The ‘scandal of democracy’ is to show that this title is actually the absence of title, or, in other words, ‘that the government of societies cannot but rest in the last resort on its own contingency’ (2006: 47). Democracy, therefore, is the name of an ‘an-archic supplement’ (2006: 58) that discloses that democracy cannot rely on one single principle and therefore cannot be identified with a juridico-political form (although it is of course not indifferent to juridico-political forms). The logic of the supplement reveals the fundamental contingency and indeterminacy of each social order, its unfinished nature, the absence of arkhê (Rancière 1999: 15). ¹

Democracy is the point where it becomes visible that there is no ultimate foundation or legitimacy, and that inequality relies on nothing else but the contingent equality of anyone with everyone. Even in the most hierarchical forms of command a moment of equality is presumed, as has already been argued above. In order to obey a command, it is required that one understands both the meaning of the command and the fact that one ought to obey (Rancière 1999: 16). A basic equality of intelligences is therefore required as an invisible presupposition in each social order: ‘Inegalitarian society can only function thanks to a multitude of egalitarian relations’ (Rancière 2006: 48). In a certain way, the command, as the central device of the police, contains its own opposite, the core of politics: ‘you need equality to explain inequality’ (Rancière 1998: 160). That is also why a confrontation between politics and the police is possible, and why they can become commensurable, even though the ‘experience of this common measure is an extreme experience’ because equality ‘is an exception’ (Rancière 2007a: 88).

It is by disclosing the contingency of each order that it is possible for a ‘part’ to stand for the whole (Žižek 2000: 188). This is a representative gesture of a very peculiar kind, as it seems to be limited to disrupting existing representations and identities and showing a glimpse of their unacknowledged (non-)ground in contingent equality. At the same time, however, the distribution of the sensible can be transformed as a result of such a democratic event. As Etienne Balibar has stated, ‘representation in this sense [what I have called the ‘democratic’ sense] is not what takes place within a pre-existing representative framework, but which creates it, and therefore deforms or subverts it’ (Balibar 1998: 214). How exactly this transformation occurs, and how democratic struggle results in the establishment of a new social order, is less elaborated by Rancière. Following Joseph Jacotot, Rancière stresses that the equality of anyone with everyone can never become a social order in itself (Rancière 1998: 162, 1999: 79). The affirmation of equality is always a singular act ‘that cannot consist in any form of social bond whatsoever. Equality turns into the opposite the moment it aspires to a place in the social or state organisation’ (Rancière 1999: 34). Each social order requires the particular type of representation that is the count of parts and parties of the community, and, as has been shown above, such a count inevitably entails a miscount and a denial of equality.

Rancière’s strategy is thoroughly post-foundational (Marchart 2007): the affirmation of equality does nothing but disclose a void, the contingency and lack of foundation of each order. At the same time, however, he admits that in a social order there can be no vacuum (Rancière 1999: 35). As a result, the political is about the constant possibility of a clash between police foundations and the political act of disclosing their lack of ground. Rancière concedes that such clashes can make a police order better: preferable is the order ‘that all the breaking and entering perpetrated by egalitarian logic has most often jolted out of its “natural” logic’ (Rancière 1999: 31). But how the representations within a police order can be partly or gradually rendered contingent (and more democratic) remains unclear. What Rancière does provide, however, is a sophisticated account of democratic representation as an event, recasting the terrain of the political in terms of the ineradicable confrontation of democratic and police representations.