They don't represent us? Synecdochal representation and the politics of occupy movements

The aim of this article is to better understand the function and place of representation in the politics of occupy movements such as Occupy Wall Street and the Spanish 15-M movement. It is often assumed that these movements categorically dismissed any form of political representation, as they criticised representative democracy in its current state and instead experimented with alternative forms of organisation and decision-making. My claim, however, is that these movements employed various forms of representation. In order to establish this, I distinguish between various notions of ‘representation,’ which are then applied to the practices and rhetoric of these movements. Engaging with the work of Ernesto Laclau, I then construct a particular – distinctively synecdochal – concept of representation that I claim is key to an adequate understanding of these movements and their radical-democratic repertoire.

did they propose an alternative view of democracy, but they also put it in practice: a significant aspect of their shared repertoire was that these movements sought to prefigure a radical form of democracy (Bray, 2013, p. 190;Douzinas, 2013, pp. 158-160;Hardt & Negri, 2017, pp. 274-275;Maeckelbergh, 2016;Pickerill & Krinsky, 2012;Smucker, 2014;Van de Sande, 2013;Yates, 2015). By organizing in general assemblies and spokes-councils, for example, they tried to realize, in the here and now, the kind of radical social change that they aspired to bring about on the long term (Graeber, 2013).
These movements' refusal to engage with existing parties and institutions, their horizontalist or direct democratic aspirations, and their attempt to prefigure these ideals in their general assemblies and consensual decision-making procedures, have all contributed to the widespread idea that they were opposed to any form of political representation. Indeed, many prominent political philosophers and social movement theorists have promoted this view of recent occupy movements.
My claim, however, is that this reading is too quick. Not only is it far from certain that all these movements dismissed political representation tout court, but upon closer inspection it becomes clear that both in their rhetoric and practical organization they employed various forms of representation and of making representative claims. In fact, these movements also played a significant representative on their own part, which remains undertheorized in the current debate.
What concepts of representation are implied here? And what lessons for democracy, strategy, and social movement politics may we derive from them?
To answer these questions and to shift the academic debate on these occupy movements, I reconstruct two concepts of representation that can be encountered in the practices and rhetoric of occupy movements such as OWS and the 15-M movement. I focus on one concept of representation in particular that I think is the key to an adequate understanding of contemporary social movements, but that is not sufficiently theorized in the existing literature on these movements: namely, the idea of synecdochal representation, where one of the constituent parts stands for a larger whole.
To address these questions I make the following argumentative steps. In Section 2, I briefly review the contemporary academic debate on occupy movements and their relation to political representation. Should the statements and practices of movements such as OWS indeed be regarded as a rejection of representation, as both their critics and advocates claim, or is their repertoire compatible with at least some forms of representation? In Section 3 I turn to the academic literature on political representation and demonstrate why it is far from obvious what representation means.
Starting from Hannah Pitkin's famous work on representation, I flesh out two different categories of representation, which are both significantly political: representation as "acting for" someone else, on the one hand, and representation as "standing for" something or someone larger than itself, on the other. In Section 4 I apply these notions to the politics of recent occupy movements, and argue why both are present (implicitly as well as explicitly) in their practices and rhetoric. In Section 5 I further develop the concept of symbolic representation as "standing for," focusing on a particular form of representation that I think is the key to our understanding of recent occupy movements and their representative role, but that nevertheless remains undertheorized in the current debate: namely, synecdochal representation. I turn to the work of Ernesto Laclau, who employed this concept at various places in his oeuvre, but never developed a comprehensive theory of it. After reconstructing his concept of synecdochal representation and its political function, I finally apply it in the context of recent occupy movements in Section 6.
Since the emergence of the occupy movements in 2011 and the following years, their significance and success-or the lack thereof-has been hotly debated among academics and activists. It is neither my goal to defend these movements or their strategic and organizational choices, or to downplay their critique of representative democracy in its current state. My aim, rather, is analytical: if we are to understand and appreciate this critique and its political relevance, we first need to acquire a more nuanced and accurate conception of representation and the political functions that it may serve-in the context of radical-democratic protest movements as well.

THE OCCUPY MOVEMENTS
Consider this fragment from an FAQ that was issued by the General Assembly of OWS: Occupy Wall Street is an exercise in "direct democracy." We feel we can no longer make our voices heard as we watch our votes for change usher in the same old power structure time and again. Since we can no longer trust our elected representatives to represent us rather than their large donors, we are creating a microcosm of what democracy really looks like. We do this to inspire one another to speak up. It is a reminder to our representatives and the moneyed interests that direct them: we the people still know our power (New York City General Assembly, 2011, p. 1).
This fragment, like many of the statements issued and slogans used by these occupy movements, is open to various interpretations. According to many political philosophers and social movement scholars, it suggests that OWS defies representative democracy. The statement appeals to the image of "what democracy really looks like": a more direct or participatory form of democracy, which is contrasted with representative democracy in its current flawed state.
One may add to this that the demand for a more radical or participatory form of democracy was expressed even more straightforwardly in their attempts to prefigure such alternatives in practice (Lorey, 2013, p. 79;2014, p. 44).
They experimented with consensual decision-making procedures and forms of public deliberation (Bray, 2013;Cornell, 2012;Graeber, 2013) and formed alternative organizational structures (Castells, 2015;Swann, 2018). They also gave rise to their own infrastructures, to facilitate daily life in the occupy tent camp, providing themselves with means of communication (Gould-Wartofsky, 2015, p. 70), food, shelter, and education (Khalil, 2011, pp. 245-250;Howard & Pratt-Broyden, 2013). In short, these movements tried to embody their ideal of a radically different, democratic society in the here and now (Van de Sande, 2013, 230;Singer, 2016, p. 85). "This is what democracy looks like," another prominent slogan used by these occupy movements, thus can be taken literally. It referred to the concrete organizational forms and practices that these movements implemented within the confined spaces-the "microcosm," as OWS calls it in their above quoted statement-of their occupied squares.
On this basis, some have argued that the politics of OWS and similar occupy movements must indeed be understood to imply a principled rejection of political representation tout court. Simon Tormey claims that the "discourse and repertoire of devices, manoeuvres and gestures" that were developed by OWS and similar occupy movements suggest that they sought to "distance themselves from 'representative politics'" (Tormey, 2015, p. 2). Michael Hardt and Antonio Negri give a similar, rather sweeping, account of these occupy movements. "One novelty of these protests and these refusals," Hardt and Negri claim, "consists in the fact that they immediately make clear that the crisis is not only economic, social, and political, but also constitutional. Representative structures and liberal governance regimes are all thrown into question" (2012, pp. 43-44). In a similar vein, Marina Sirtin and Dario Azzelini argue that one of these movements' most prominent slogans-"You don't represent us!" (¡Que no nos representan!)-must be interpreted as a rejection of political representation tout court: "They don't represent us!" has emerged as a powerful slogan in mobilizations all over the world. We hear it in the US, Spain, Greece, Brazil, Turkey, Slovenia, and even Russia… . The slogans are not phrased as rejections of specific political representatives, but as expressions of a general rejection of the logic of representation. (Azzellini & Sitrin 2014, p. 41, italics added) The underlying assumption here is that representative democracy is essentially an oxymoron. Representation categorically precludes the people from immediate participation in the political process (Hardt & Negri, 2004, pp. 244-245;Rancière, 2006, p. 53). Real democracy, on the other hand, presupposes a horizontalist conception of organization and a rejection of "representative democracy and the empowerment of leaders that such delegation of authority" entails (Sitrin, 2012, p. 74; see also Mason, 2011, p. 45). The horizontalist or radical-democratic aspiration that the occupy movements expressed in both their practices and rhetoric is thus not merely in tension with political representation: the two are fundamentally incompatible.
Interestingly, this analysis is shared by some of these movements' most prominent critics. Chantal Mouffe, for example, argues that their reluctance to engage with any representative institutions or offices reveals a misconception of the state as a "monolithic entity instead of a complex set of relations, dynamic and traversed by contradictions" (2013, pp. 118-119). On a more fundamental level, moreover, their horizontalist aspirations evince a "flawed understanding of politics" in general (2013, p. 78; see also Mouffe, 2005, pp. 107-115). Mouffe holds that a pluralist democracy simply "cannot exist without representation," because political identities are the products of discursive articulation and representative claim-making (2013, p. 126). What these movements ought to challenge, she claims, is the fact that citizens today are no longer offered alternatives to the hegemonic discourse-not our representative institutions and procedures themselves.
I agree with these critics that it is "both analytically confusing and politically unwise to base democratic struggles on a wholesale rejection of all representative politics" (Teivainen, 2016, p. 33). However, I do not think that recent occupy movements such as OWS and the M-15 categorically rejected any form of political representation-nor am I convinced that their political repertoire was devoid of making all forms of representative claims. How are we going to make sense of these movements' practices and slogans such as "We are the 99%," "This is what democracy looks like" or even "They don't represent us!" without employing at least some concept of political representation (Brito Vieira, 2015, p. 506)? In the remainder of this article, I flesh out a few concepts of representation that I think are applicable to these movements. But before doing so let me first briefly elaborate on the various uses and meaning of this-often misunderstood-term.

REPRESENTATION: THE VARIOUS USES OF A CONCEPT
What exactly is representation? If we look more closely at how the term is employed both in academic theory and political practice it may strike us that it has several meanings and uses. As Hannah Pitkin argues in her seminal work The Concept of Representation (1967), these various uses of the term representation are "often intertwined, but their implications and consequences are very different" (Pitkin, 1967, p. 59). Pitkin divides these various uses into roughly two different categories that can be understood as a conceptual distinction similar to that made in several other European languages (Hofmann, 1974, p. 19;Sintomer, 2013, pp. 19-20;Spivak, 1988, pp. 276-277).
In Dutch, German and Italian, for example, representation may be translated as vertegenwoordiging, Vertretung, or rappresentanza, respectively. What is meant, in this case, is the particular relation in which a particular representative (be it a politician, a delegate, or a lawyer) speaks, negotiates, votes, or even decides in someone else's stead. Pitkin refers to this form of representation as "acting for": "the idea of substitution or acting instead of, the idea of taking care of or acting in the interest of, and the idea of acting as a subordinate, on instructions, in accord with the wishes of another" (1967, p. 139). Usually, representation as acting for requires an explicit mandate, although this is not always the case (Mansbridge, 2003). It does, in any case, require a certain degree of responsiveness. That is to say, when understood as acting for, the legitimacy of political representation largely depends on the ability of a representative adequately and accurately to convey the opinions, preferences, or interests of a particular group or constituency that exists both independently and at a certain distance (Manin, 1997) from the representative. In the case of representative as acting for, a legitimate representative "must be responsive to [the one represented] rather than the other way around" (Pitkin, 1967, p. 140). This concept of representation is implied, for instance, in our common understanding of representative democracy.
However, all the aforementioned languages also have a second term commonly translated as representationnamely, representatie, Darstellung, or rappresentazione. This is used in the way in which, for instance, a national flag, logo or symbol, or a work of art can be understood symbolically to represent something other or larger than itself.
Pitkin refers to such symbolic forms of representation as "standing for". Neither English nor Spanish or French (which, interestingly, were the primary languages spoken in many recent occupy movements) allows us to draw this distinction: representation, representación, and représentation roughly refer to both concepts of representation: both as acting for and standing for. Even in the predominantly Anglophone academic debate on political representation the two tend to be mixed up. However, both at a theoretical and a practical level, there are important reasons to maintain a conceptual distinction between these forms of representation.
As seen, the most significant difference between representation as acting for and representation as standing for is that the former often implies an explicit mandate and a certain degree of responsiveness between the representative and the thing represented. In the case of acting for, the additive re thus must be taken rather literally. Here, the re-presented is understood to precede ontologically the re-presentative. Representation as standing for, in the symbolic sense, on the other hand, has a more constructivist connotation. Symbols "are often said to represent something, to make it present by their presence, although it is not really present in fact" (Pitkin, 1967, p. 92). This is how a national flag is understood to represent something. Symbols do not correspond to an absent reality that could be presented more or less accurately.
They bring something into being that did not exist as such, prior to (or independent of) its symbolization. A symbol is not merely a source of information, "is supposed to evoke or express feelings appropriate to what it stands for, and what happens to a symbolic figure must be considered as happening to its referent as well" (Pitkin, 1967, p. 99). The establishment of symbolic representations, in other words, is a gradual process that also involves the coming-to-be of certain affective or irrational associations. Thus, that which is represented in a symbolic relation of representation does not precede this relation as such. There is no unarticulated referent outside or behind the particular symbol or claim that represents it. Whereas representation as acting for implies that there is an objective and autonomous referent whose interests can be represented in a more or less accurate way; representation as a symbolic standing for suggests that the referent does not exist independently, it is brought into being through the articulation of a representative claim or relation (Disch, 2011, pp. 106-9).
Although the political function of representation as acting for is more evident, that symbolic representation as standing for is of no less political relevance. Constructivist theorists of political representation, such as Ernesto Laclau, Chantal Mouffe, and Michael Saward, have shown that representation as standing for plays an important role in the establishment of political discourses and identities (Disch, 2019). Saward (2010, p. 15) argues that "the active making of symbols or images of what is to be represented" is a "central aspect of political representation." By articulating a particular image of whom or what they stand for, claim-makers give rise to this precise group or constituency. This evidently does not mean that this constituency has no material pre-existence whatsoever: "there is always a referent. But the real political work lies in the active constitution of constituencies-the making of representations" (Saward, 2010, p. 51). Thus perceived, representation as standing for is an articulatory process (Laclau, 1990, pp. 38-39;2007, p. 98): any representative claim is constitutive of its own referent in that it must be articulated in a particular way. Only in its repeated use does it acquire content and meaning. And, the other way around, the meaning of our words and identities gradually emerges from such processes of signification. Representation as standing for thus can be understood as a "process of identity construction." (Errejón & Mouffe, 2016, p. 112). This is a distinctively political process, one may add, because there is always a competition between various possible articulations. To establish one discourse or collective identity must be done at the expense of others. Or, to put it in Laclau and Mouffe's terms: any symbolic representation seeks to acquire hegemony. Its aim is to "dominate the field of discursivity, to arrest the flow of differences, to construct a centre" (Laclau & Mouffe, 2001, p. 112).
The question, then, is: did the practices of recent occupy movements such as the M-15 and OWS indeed eschew of any form of political representation-as many of their advocates and critics claim? And if not, how can these various forms of political representation as acting for and standing for be recognized in their rhetoric and practices?

MOVEMENTS
As we have seen above, the emergence of recent occupy movements is often interpreted as indicative of a categorical dismissal of political representation tout court (Azzellini & Sitrin, 2014, p. 41;Hardt & Negri, 2012, pp. 43-44;Mouffe, 2013, pp. 125-127;Tormey, 2015, p. 82). It is clear that these movements were critical of representative democracy in its current state. But now that we have distinguished between different forms of political representation, we may ask whether their politics indeed implies a wholesale rejection of all forms of representation?
One important element of the politics of these occupy movements, to start with, is that both in their rhetoric and practices they sought to counterpoise representative democracy with a radically different-more direct or participatory-form of democracy. Their general assemblies, online forms of organization, and horizontalist decisionmaking procedures are attempts to provide a "glimpse of what real democracy might be like" (Graeber, 2013, p. xvii).
This may suggest that these movements were at least very critical of representation as acting for-the idea that some may be formally entitled to speak, act, and decide in someone else's stead.
But if we look more closely at both their rhetoric and their practices, these movements turn out to have a much more nuanced understanding of representation as acting for. Take, for example, the OWS communique quoted at the beginning of Section 1. Although this statement implicitly draws a distinction between real democracy and representative democracy as we know it today, it also charges "our elected representatives" for their failure to represent the people's interests adequately (New York City General Assembly, 2011, p. 1). This leaves open the possibility that there are ways to act for or on behalf of the people in a more legitimate or sufficiently democratic way. Arguably, OWS saw a role for itself in this respect: in his memoirs of the movement, OWS cofounder David Graeber (2013, p. 40) suggests that whereas both major political parties in the USA "represent the 1%, we represent the 99% whose lives are essentially left out of the equation." Moreover, it turns out that these occupy movements often employed various forms of political representation in practice. For example, a few months into the occupation of Zuccotti Park, OWS's general assembly was gradually replaced as the most important organ of decision-making by a spokes-council (Bray, 2013). Its regular meetings would be attended by delegates (or spokespeople) from each working group. This task was rotated among the members of each group, and each delegate had a mandate to voice only the working group's decisions (Graeber, 2013, p. 232).
According to Mark Bray, these delegates were therefore not representatives in the strict sense "because they merely communicate the will of the smaller group" (Bray, 2013, p. 88). But arguably it would be more accurate to say that delegation or an imperative mandate is a form of representation as acting for that requires an explicit mandate and a very high degree of responsiveness.
Such forms of representation are by no means incompatible with decentralized organization or even consensusoriented decision-making. Many anarchist and radical democratic theorists have argued before that radical democracy requires various forms of recallable delegation or imperative mandate (Bookchin, 2015, p. 26;Cohn, 2006, pp. 255-256;Teivainen, 2016, p. 23;Wilson, 2014, pp. 147-148). Moreover, the emergence of new, networked forms of communication has done very little to reduce the role of leaders and representatives (Gerbaudo, 2012, p. 159;Nunes, 2014;Smucker, 2017, pp. 184-186). The real question, arguably, is not how such roles can be limited to an absolute minimum but rather how they can be frequently rotated and equally distributed. These movements' prefigurative experiments with decentralized and participatory forms of organization may have given expression to a radically different idea of democracy, but this neither necessarily implies nor requires a full-fledged dismissal of representation as acting for.
What is more, many of these movements' participants have continued to engage with more conventional forms of electoral politics after their square occupations had come to an end. One obvious example is the Spanish left-populist party Podemos, which emerged in the aftermath of the 15-M movement and has had considerable electoral success at both the national and the European level (Errejón & Mouffe, 2016, pp. 70-71). For Podemos and its party leader Pablo Iglesias, this turn to an electoral strategy was no more than a logical next step (Iglesias, 2015, pp. 11-15). More striking, perhaps, is that even some anarchist activists previously involved with the 15-M movement have continued to follow an electoral strategy-in particular at the local level. In the small Spanish city Castellón, for example, anarchists took part in the formation of a new public platform named Castelló en Moviment, which ran in the local elections of 2014 (Ordóñez, Feenstra, & Franks, 2017). Some "small-a" as well as "capital-A" anarchists (Graeber, 2002; see also Gordon, 2008, pp. 23-24) are now holding representative posts in the city council. Their aim is to "change traditional mechanisms of political intermediation to create channels of citizen participation and thus transform local government into a radical, direct municipalist structure" (Ordóñez et al., 2017, p. 10).
The widespread claim that the politics of these movements defied the very logic of representation thus appears to be unfounded (Azzellini & Sitrin, 2014, p. 41). Upon closer inspection it turns out that many forms of representation as acting for were deliberately employed by these movements. This suggests that their radical democratic practices and slogans could "also be read as expressing a demand for better, and more direct and participatory representation to overcome the current 'misrepresentation' of the people's will" (Gerbaudo, 2017, p. 84).
However, there is also a more fundamental (and arguably more theoretical) point to be added here. For not only did these movements' mobilization and internal organization incorporate certain forms of representation as acting for, but in some respects they also fulfilled an important representative role in their own right. Movements such as OWS and the M-15 did not have, or claim to have, any mandate or legitimacy to act, negotiate, or make decisions on behalf of a larger group of people. But they nevertheless did serve as its political representatives in at least some respects. To appreciate this, however, we need to return to the second, more symbolic concept of representation that I described above; namely, representation as standing for, or Darstellung.
The politics of a social movement without any form of symbolic representation is practically unimaginable. Logos, flags, slogans, and banners evidently fulfill an important symbolic function. They not only communicate a message or demand to the public, but activists also derive a sense of identity and community from them. Even the most devoted anarchist, who may reject any form of institutional representation, would surely not deny that their own black flag or their Guy Fawkes' mask stands for something (Cohn, 2006).
The recent occupy movements were no exception to this. Let me once again return to the OWS statement quoted above. As established, the statement is critical of representative democracy in its current condition. But the text also adamantly states that "we the people still know our power" (New York City General Assembly, 2011, p. 1). And this "we the people"-a loaded phrase in US political history-at least implies that the General Assembly issuing this statement stands for a more encompassing "people," and does not merely act in the name of those physically present in Zuccotti Park (Butler, 2015, p. 177). A similar assessment can be made of other prominent movement slogans, such as OWS's "we are the 99%," and (somewhat ironically, perhaps), "they don't represent us!" Both slogans entail a strong representative claim (Saward, 2010) that refers to a larger group of people ("us," "the 99%"), and not merely to those who are physically present at the moment of utterance (Brito Vieira, 2015, p. 506). The prefigurative practices of these occupy movements implied a similar representative claim: the general assembly, alternative decision-making procedures and organizational structures all stood for something larger.
My claim, therefore, is that the particular form of political representation, standing for, is presupposed in the practices, slogans, and rhetoric of these occupy movements. Much like representation as acting for, this form of representation is not incompatible with their radical democratic aspirations or potentials, as many social movement theorists and political philosophers would argue. In fact, I argue that it is it is implied in their prefigurative, direct democratic repertoire. In the remainder of this article, I aim to flesh out a particular concept of representation as standing for that is at stake here; namely, a synecdochal concept of representation. With Ernesto Laclau, I will argue why this particular form of representation is key to obtaining an accurate understanding of the recent occupy movements.

SYNECDOCHAL REPRESENTATION
Representation as standing for has a much wider use than representation as acting for. Works of art, symbols, protocols, or means of payment all represent something other or larger than themselves. However, as argued above, this symbolic form of representation also fulfills a distinctively political role. It is of elementary importance in the construction of collective identities. Whereas representation as acting for is a typically responsive process where a representative must seek to act or speak adequately and accurately on behalf of the represented object; representation as standing for is a "two-way process: a movement from represented to representative, and a correlative one from representative to represented" (Laclau, 2005a, p. 158; see also Laclau, 2007, p. 98). This is also a distinctively political process because, in order to establish itself, any identity must seek hegemony and thus go forward at the expense of others (Laclau & Mouffe, 2001, pp. 141-145). When Ernesto Laclau speaks of political representation, he often implicitly refers to this much broader concept of representation as a symbolic standing for (Arato, 2013, p. 160).
In his discussion of political representation, Laclau on several occasions invokes the image of the synecdoche: a form of figurative speech that exists alongside other types such as metaphor and metonymy. Laclau employs the differences between metaphor, metonymy, and synecdoche to understand how hegemonic representations are being established (Laclau & Mouffe, 2001, p. 110). Although this figure of speech fulfills an elementary ontological role in Laclau's theory of representation (Laclau, 2005a, p. 72), he never developed a comprehensive account of the synecdoche: references to this subject are scattered throughout his oeuvre. Before this concept can be applied to the representative claims and practices of recent occupy movements, we thus first need to reconstruct the specific (and often implicit) function of the synecdoche in his theory of representation.
But let me first qualify what exactly the synecdoche is, how it is distinguished from other figures of speech such as the metaphor and the metonymy, and why it is relevant in a discussion of political representation. The synecdoche may be regarded as a subcategory of metonymy (Deseriis, 2015, p. 60), hence we must start by distinguishing between metonymy and metaphor. Whereas the latter is based on analogy or association, the establishment of metonymy requires a contiguous relation. For example, in radical circles it is common to refer to police officers as "pigs." This is a metaphor: after all, the officers in question probably do not bear greater physical resemblance to pigs than anyone else. The relation between "pigs" and "police officers" is merely based on association-the perceived brutishness or thoughtlessness of their behavior reminds us of animals. Whether this association is justified or not, is irrelevant (it is common knowledge that pigs are highly intelligent animals). In metonymy, on the other hand, the relation between the signifier and the signified must be warranted by a particular contiguity. Common examples are Wall Street or the White House as references to the financial industry or to a particular branch of the US government, respectively. In both cases there is a direct and concrete relation between them (the stock market is perceived as the center of financial power; the president lives in the White House), but the term used as a signifier does not exhaust this relation. A synecdoche, finally, implies a metonymic relation in that it is based on a similar contiguous relation. But there is one important, additional criterion: in a synecdochal relation the signifier is itself a constitutive part of the larger whole that it signifies.
It requires a pars pro toto relation. Expressions such as "we have hungry mouths to feed" or the use of "wheels" as a reference to automobiles are famous examples of a synecdoche: "mouths" and "wheels" obviously are merely a part of their referent-albeit an elementary one.
Let me return to the distinction between metonymy and metaphor (Laclau, 2014, p. 87). As stated, metaphors are typically based on associations. But this also means that metaphorical representation can be established only as a more or less exclusive, closed relation (Laclau, 2014, p. 88). Because it is based on mere association, many different metaphors can be constructed to characterize a single referent. To return to the example given above: many (but not all) species from the animal kingdom may be equally fit to depict the behavior of police officers. There is no particular reason why the contingent association between the police and pigs has led to such an established metaphor in activist and youth culture (various species could have fulfilled the same metaphorical role). But the fact is that, in order to succeed as a metaphor, pigs had to exclude these other possible associations, and thus had to present itself as the only or most self-evident signifier for the police. A metaphor needs to be continuously reinforced, precisely because it is contingent-and, thus, one among many options. This is very different for the metonymy-which, as we have seen, is based on a contiguous relation. The part that serves as a signifier in a metonymic relation is more closely related to the whole that it embodies, and the contiguous relation between them thus speaks for itself. Everyone with a rudimentary knowledge of US politics will immediately understand what the White House or Washington stand for-both terms, moreover, can readily be used in reference to the same referent. Thus, the success of a metonymy does not depend on its ability to exclude other possible symbolic relations: it remains more open than a metaphor.
As far as Laclau is concerned, however, this also means that the metonymy has only a limited potential for establishing hegemonic discourses and representations. In a process of hegemonization, a particular representation of society gradually takes upon itself the role of serving as a dominant representation of society as a whole. It must "overflo[w] its own particularity" and become an "incarnation" (Laclau, 2007, p. 72)-a complete and full embodiment of a social totality that always remains an impossibility (Laclau, 1990, pp. 38-39;Marchart, 2007). The problem with metonymies, however, is that they are loosely based on one of many possible contiguous relations, and that the establishment of one metonymic relation does not necessarily advance at the expense of others (as we say was the case with metaphorical relations). Metonymies thus lack the exclusionary potential that is required to establish a hegemonic political representation (Torfing, 1999, pp. 112-113). To acquire discursive hegemony, a metonymy would first need to lose some of its contiguous openness (Laclau, 2014, p. 62;2005a, p. 109). This is where the figure of the synecdoche becomes relevant. Laclau ascribes a particular function to this rhetorical figure. He approvingly quotes Paul de Man, who describes the synecdoche as "one of the borderline figures that creates an ambivalent zone between metaphor and metonymy" (De Man, 1979, p. 63, note 8;see Laclau, 2014, p. 87). Seen from this perspective, the synecdoche is not merely a subcategory of the metonymy: it occupies a middle ground between these other figures in that it constitutes a relation between signifier and signified that is neither contiguous nor analogous-and, therefore, neither open nor closed. The particular part that "takes upon itself the task" (Miller, 2004, p. 220) to embody or stand for a larger whole, is never the only part that could fulfill this role (Laclau, 2000, p. 57;Laclau, Olson, & Worsham, 1999, p. 135). Yet its chance to succeed as a synecdochal signifier does depend on its ability to be presented as the only (or, at least, the most obvious and self-evident) representation. The synecdochal representation creates the illusion of a closure (Laclau, 2014, pp. 87-88). It presents an image of the whole, for instance, by articulating a concept of the people that will never be exhaustive, complete, or even fully possible (Laclau, 2014, p. 97), but that nevertheless presents itself as the embodiment of this totality, and thus serves to constitute it as such.
As far as Laclau is concerned, this is exactly why the rhetorical figure of the synecdoche forms the ultimate basis for our understanding of political representation: it explains how one representative claim or relation can signify a larger whole, which, at the same time, must always exceed it (Laclau, 2005a, p. 72;Deseriis, 2015, pp. 60-61, 216).
Laclau's employs these rhetorical figures literally and not as mere analogies. Political representations are constructed as synecdoches-not merely like them.
Let me give two different examples of how this works in practice. Imagine a local community whose members have a particular shared need or interest. It can be something rather mundane, and initially rather unpolitical: in Laclau's own example they want a bus route through their neighborhood (Laclau, 2005b, p. 36). However, their request is turned down by the authorities. As long as this remains an isolated case, Laclau argues, it will probably not give rise to a stronger counter-hegemonic mobilization. But now imagine that this is only one among many different unsatisfied claims. It turns out that similar requests of other neighborhood residents or surrounding communities have also been denied. Laclau describes how this may lead to the institution of a social frontier: their shared experience of neglect or their frustration with the authorities in question brings people together and constitutes a sense of collectivity, which is counterpoised by a shared understanding or appreciation of the authorities, which serve as its "constitutive outside" (Laclau, 1990, p. 17). The construction of a social frontier not only affects how these communities identify, but leads to a different articulation of their demands. It leads to the establishment of a larger set of claims, "one in which all the demands, in spite of their differential character, tend to reaggregate themselves, forming what we will call an equivalential claim" (Laclau, 2005b, p. 37). Notwithstanding the initial differences between these people and their particular interests, needs or preferences; their shared experience of neglect and their positioning along the lines of a newly constituted social frontier leads them to see a seemingly logical or natural connection between these various claims and demands. This chain of equivalence, however, needs to represented in a particular way: a mere summing up of these various demands will not be sufficient. The representation of an equivalential chain "is only possible if a particular demand, without entirely abandoning its own particularity, starts also functioning as a signifier representing the chain as a totality" (Laclau, 2005b, p. 39; see also Marchart, 2019, pp. 111-116). By virtue of its public appeal or its broad applicability, for example, one particular demand comes to represent synecdochally the entire chain, thus standing for a more encompassing series of demands or identities.
A second example can be found in the political practices of social movements. Imagine a trade union that seeks to organize in a working-class neighborhood that has strong racial and cultural diversity. Initially, the union's campaigns will focus on socioeconomic issues and working conditions. But it turns out that this community is also suffering from an increase in racially motivated violence, and there are no other grassroots organizations focusing on such issues. It is not unlikely that, by lack of an alternative, the union will gradually start to play a more significant role in this struggle as well. Neighborhood residents may thus start to associate the term "labor union" with "anti-racism" and vice versa.
Although there are obvious connections between them, there is no reason why a labor union should necessarily be engaged in anti-racism-and yet, "the union" comes to stand synecdochally for a larger chain of demands, struggles, and identities, including racial equality and anti-racism: First, despite the differential particularism of the initial two kinds of struggles and demands, a certain equivalential homogeneity between them is being created. Second, the nature of the trade unions changes in this process: they cease to be the pure expression of sectorial interests at a given moment, and become more-if a variety of equivalential articulations develop-a nodal point in the constitution of a "people." … Third, the term "trade union" becomes the name of a singularity …: it no longer designates the name of an abstract universality, whose "essence" would be repeated, beyond accidental variations, in all historical contexts, and becomes the name of a concrete social agent, whose only essence is the specific articulation of heterogeneous elements which, through that name, crystallize in a unified collective will (Laclau, 2007, pp. 109-110).
What we see here, again, is how a particular struggle or demand comes to stand for a much broader equivalential chain-without losing its particularity at the same time (the labor union in Laclau's example will continue to advocate good working conditions). What this example stresses, however, is that the construction of a broader identity thereby also transgresses the level of concrete demands or claims. The trade union here synecdochally represents not only the concrete agenda or programme of an organization, but more broadly a collective identity, a particular articulation of "social justice," "equality," or other values and ideals, and a strategic view of how these must be brought about. In short, it comes to stand for a more encompassing political repertoire.
How, then, can Laclau's concept of synecdochal representation help to understand the politics of recent occupation movements? It is far from obvious to apply Laclau's thought in this context: both he (2014, p. 9) and his co-author Chantal Mouffe (2013) have strongly criticized these movements for their perceived refusal to engage with representative politics. My claim, however, is that their work could nevertheless be employed in an analysis of these very movements and their prefigurative practices (Decreus, Lievens, & Braeckman, 2014, p. 138). In the next section, I give a few examples of how these movements established such synecdochal representations.

THE OCCUPY CAMP AS A SYNECDOCHE
Although both Laclau and Mouffe have been critical of occupy movements such as OWS and the Spanish 15-M movement, I hold that their theoretical framework can well be employed to acquire a better understanding of these movements and their political repertoire (Decreus et al., 2014, p. 138). Laclau's concept of the synecdoche, in particular, can help us to recognize the forms of political representation that were implied in these movements' practices and statements. Let me give a few examples, focusing on OWS.
A first striking example, of course, is that of Occupy's famous slogan: "we are the 99%". This slogan allegedly was inspired by Joseph Stiglitz's (2011) article in Vanity Fair, in which he argued that only 1% of US Americans receives nearly a quarter of the nation's yearly income, and that it owns 40% of the nation's total wealth. A blog on Tumblr greatly contributed to the popularization of the idea that society could be divided between a 99% and a 1% (Graeber, 2013, p. 41). 3 On this website, thousands of people, mostly from the US, posted pictures of themselves, accompanied by a brief statement that described their individual situation: giving testimonies of debt, unemployment or systematic underemployment, poverty, chronic illness, and war trauma. Their testimonies were all undersigned by the slogan "I am the 99%." Although these were all individual accounts of many different situations and experiences, the slogan suggested that all these stories had something important in common. It seemed that, in a way, every single personal story could embody or encompass all the others-they all represented the entire 99%. Whereas this idea of the 99% suggests that they together constituted a majority, the I in "I am the 99%" also suggested a synecdochal, representative claim: my story is representative for what we all are going through (in spite of the obvious differences between us).
In this way, the slogan and the idea of a 99% constructed what Laclau and Mouffe call an equivalential chain between various stories and identities on the basis of a relatively vague and undefined experience that they all seemed to have in common. This shared experience, which perhaps could best be described in terms of deprivation or precarity, came to represent synecdochally a wide variety of concrete inequalities, instabilities and injustices-and thus gave rise to a newly articulated collective class identity (Dean, 2014, p. 385).
Another example is the role of occupation in these movements' repertoire. One grievance shared by many of these movements is that over the past decades public space-both in the literal sense of a physical space, and in the more abstract sense of a public sphere-has become increasingly inaccessible. In all parts of the world public spaces and services are being privatized, and this development in many ways corresponds to the establishment of an empty and watered-down concept of citizenship. The occupation of public space was not merely a strategic instrument, but in one way or the other, the occupied space itself was also often at stake (Butler, 2015, p. 126). As Wendy Brown argues, it is no coincidence that the occupation of space played such an important role for various movements around the world, which all "repossessed private space as public space, occupied what is owned, and above all, rejected the figure of citizenship reduced to sacrificial human capital and neoliberal capitalism as a life-sustaining sacred power" (Brown, 2015, p. 220). For instance, Istanbul's Gezi Park was initially occupied in defiance of the city's plan to restructure it-and, more generally, to protest against the regime's attitude towards its citizens' right of assembly and freedom of press. New York's Zuccotti Park, to name another example, is not only conveniently located only a few blocks from Wall Street, but it is also a privately owned public space (Foderaro, 2011) that bears the name of its proprietor. Of course, this is a much more widespread dynamic, and the occupation of a single square will do very little to turn the tides against the privatization of public space. But it does synecdochally represent a critique of the appropriation of public space (and, in extension, of neoliberal privatization in general) and gives expression to a demand for its protection.
One last important, synecdochal function could be ascribed to the concrete forms and techniques of organization that were prefigured in the Occupy camp, such as the iconic general assemblies and its consensus-oriented decisionmaking procedures. These structures, procedures, and techniques not only served an important organizational function, they also fulfilled an, even more important, articulatory or symbolic role. The search for consensus at general assemblies, for example, well exceeded its instrumental or strategic purposes. As social movement theorist and historian Andrew Cornell argues: [C]onsensus functions as a synecdoche-a part rhetorically standing in for a greater whole. In this case, the whole that consensus stands for is a participatory, egalitarian, self-determining movement, on the one hand, and, on the other, a society with the same characteristics. (Cornell, 2012, p. 164, italics added) Graeber makes a similar point when he argues that OWS's direct democratic decision-making procedures and horizontalist structures were understood to "offer a model, or at the very least a glimpse of how free people might organize themselves, and therefore what a free society could be like" (Graeber, 2013, p. 233; see also p. 43, pp. 190-1).
This does not mean that these movements always consistently used, or claimed to use, direct democratic decisionmaking procedures. As we have seen above, their activist repertoire often comprised various forms of representation as acting for. But the point is that, by experimenting with alternative forms of organization and decision-making, these movements did give expression to their demand for a more participatory and legitimate form of democracy than we currently have in our electoral-representative state. The general assembly, "jazz hands," "the people's mic" and other techniques used by these movements, synecdochally represented a much larger set of grievances and demands.
The point is not simply that the synecdochal form of representation serves as an analogy for how these practices symbolically represent something larger than themselves (Franks, 2003, p. 19). Regardless of their concrete appreciation of representative democracy in its current state, the practices of protest movements are always discursively mediated. In other words, the exact way in which their practices and organizational structures stand for a more encompassing social ideal or critique, must find expression through language, memes, images, or symbols. The occupation of public space, the implementation of consensus-oriented decision-making procedures and the prefigurative realization of new organizational structures could acquire a synecdochal meaning because they were accompanied by-and, thus, mediated through-particular representative claims, demands, slogans, and symbols. Statements such as "this is what democracy looks like," "we are the 99%," or "they don't represent us," inscribed these experiments with a particular synecdochal meaning and representativeness. It is through the articulation of these statements that these prefigurative experiments came to stand for larger than themselves: a particular concept of democracy, a notion of the 99% or the people.

CONCLUSION
What is the role and relevance of political representation in the repertoire of occupy movements, such as OWS and the 15-M movement? Since the emergence of these movements in 2011 this has been a subject of debate among political theorists and social movement scholars. Many advocates and critics of these movements have argued that their horizontalist aspirations, their radical democratic rhetoric and their prefigurative approach all imply a principled rejection of political representation per se. I instead argued that, if we want to appreciate these movements' views of political representation we first must be able to disentangle its various uses and meanings: representation as acting for (or Vertretung) on the one hand, and representation as standing for (or Darstellung) on the other. The former often implies an explicit mandate and always requires some degree of responsiveness between the representative and the represented.
Representation as standing for, on the other hand, is a distinctively symbolic form of representation in which the thing represented acquires a particular articulation through the process of representation.
I argued that both forms of representation were in fact deliberately employed by these occupy movements: representation as acting for can be encountered in their use of recallable delegation and imperative mandate. Representation as standing for, on the other hand, can be recognized in the way these movements synecdochally represented a broader idea of who they stood for and what they aspired to realize in the future. Examples of such synecdochal claims can be found in their activist practices on the one hand-the occupation of public space, the general assembly-and in their accompanying rhetoric and slogans-"This is what democracy looks like!" "They don't represent us!" "We are the 99%"on the other. Thus, to read these movements' slogans and practices as principled rejections of political representation is beside the point.
However, these findings do give rise to new questions. Given that representation as standing for does not require any responsiveness, then how must the legitimacy of representative claims be assessed and safeguarded (Brito Vieira, 2015; Saward, 2010)? And does a synecdochal form of representation, in which a part claims to stand for a larger whole, sufficiently cater to its diversity and plurality? Or may it instead facilitate authoritarian or even totalitarian forms of leadership (Arato, 2013)? Finally, if the politics of recent occupy movements indeed consisted of many forms of political representation, then how exactly must their critique of representative democracy in its current state be appreciated?
To what, if not a purely direct or participatory form of democracy, did these movements aspire? If their slogans and practices must indeed "be read as expressing a demand for better, and more direct and participatory representation" (Gerbaudo, 2017, p. 84), then what would this alternative view of democracy look like in practice? Eight years after the emergence of these movements, many of these questions remain to be answered.
One elementary step toward a productive discussion of these questions, however, is that we first acknowledge the role that various forms of representation played in these movements' politics and rhetoric. Only then will we be able to draw meaningful lessons from their experiences on the future of democracy and social movement politics.

NOTES
1 Representation as standing for can be divided into two more subcategories, as it can have both a descriptive and a symbolic function. If, as Pitkin holds, representation literally means that something absent is made present, then a descriptive form of representation allows us to "draw accurate conclusions about the represented, gather information about the represented, because it is in relevant ways like the represented" (Piktin, 1967, p. 81, italics added). One could think, for example, of a passport photo or a map from which such information can be derived. It has been argued that Pitkin turns a blind eye to the political potentials and implications of descriptive representation (Mansbridge, 1999). But for the sake of brevity, I focus only on her concept of symbolic representation as standing for. 2 Pitkin also reconstructs a third, more formalist concept of representation, that has been of significant historical influence, but which no longer seems to apply to our current use of the term. From this formalist point of view, representation is a transaction-some person, institution or state formally receives all authority over others. In this case, the "defining criterion for representation lies outside the activity of representing itself," and although in some cases the representative can be held accountable for her deeds, they cannot be understood to misrepresent the wishes, interests or identities of the represented.
In other words, in this case "the defining criterion for representation lies outside the activity of representing itself" (Pitkin, 1967, pp. 58-59).