SciELO - Scientific Electronic Library Online

vol.4 edição especial índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  


versão impressa ISSN 0011-5258

Dados v.4 Rio de Janeiro  2008


Civil society, participatory institutions and representation: From authorization to the legitimacy of action


Sociedade civil, instituições participativas e representação: Da autorização à legitimidade da ação


Société civile, institutions participatives et représentation. De l'autorisation à la légitimité de l'action



Leonardo Avritzer

Translated by Ana Carolina Ogando
Translation from Dados – Revista de Ciências Sociais, vol.50, n.3, p. 443-464, 2007




In democratic Brazil, political participation has been characterized by the enlargement of civil society's presence in public policies. Such presence generated a series of institutions with the participation of civil society and state actors in the areas of social welfare, health and urban policies. Within these institutions, it began to happen what we can call an institutionalized representation of civil society. This representation is characterized by the inexistence of an explicit requirement of authorization, and the absence of a structure of territorial monopoly in the representation carried out by the actors of civil society. Representation by affinity is proposed as a manner of associating parliamentary representation and representation of civil society.

Key words: civil society; representation; authorization


Dans le Brésil démocratique, la participation politique est marquée par une plus grande présence de la société civile dans le domaine des politiques publiques. Cette présence a engendré une série d'institutions où participent la société civile et des acteurs des États agissant sur le terrain de l'assistance sociale, de la santé et des politiques urbaines. Ces institutions ont fait apparaître ce qu'on peut appeler une représentation institutionnalisée de la société civile, dont les caractères sont tout d'abord l'absence explicite d'autorisation ainsi que l'absence de structure de monopole territorial dans la représentation effectuée par des acteurs de la société civile. Dans cet article, on cherche à discuter une façon de penser la légitimité de cette représentation en proposant la représentation par affinités comme susceptible d'associer représentation parlementaire et représentation de la société civile.

Mots-clé: société civile; représentation; autorisation




Political participation in democratic Brazil has been marked by two important phenomena: the growth of both civil society's presence in public policies and of the so-called participatory institutions. From the standpoint of civil society, diverse actors, belonging to this political field, sought greater presence in institutions known for deliberation on public policies in the areas of health, social work and urban policies since the end of the authoritarian period (Coelho, 2004; Cunha, 2004; Avritzer, 2008; at press). These claims generated a series of hybrid formats that can be characterized by the presence of institutions along with the participation of civil society and state actors in the areas of social work, health, the environment, and urban policies (Coelho et alii, 2006; Abers e Keck, 2006). This presence has grown stronger in the governments that legalized diverse forms of insertion of civil society associations in public policies. These institutions-have, up until now, been analyzed, through the perspective of an increase in participation. In fact, there are more councilors in Brazil than town councilors, and in some cases, such as in participatory budget, the participation in some years reached almost 180 thousand people1. Nonetheless, as the involvement of civil society in social policies grew, a problem became evident: the emergence of new forms of representation.

The participatory institutions that emerged in democratic Brazil gave rise to an increase in representation (Gurza Lavalle, Houtzager and  Castello, 2006), be it through the fact that social actors began to nominate themselves as representatives of civil society or because the State began to institutionally deal with an official representation of civil society. By an increase in representation, I mean the increase in the ways social actors play a role in presenting certain themes in these institutions, such as health or urban interests and the fact that in institutions such as policy councils, some actors are elected with the intention of carrying out a role as representatives of civil society. Therefore, it is not difficult to realize that the representation carried out by the actors of civil society is different from that which is exercised in the representative institutions par excellence, that is in Parliament. There are two aspects which distinguish representation in participatory institutions from representation in parliament: in the first place, there is no explicit authorization requirement, as elaborated by Hobbes, and later developed by Hanna Pitkin. Secondly, there is no structure of territorial monopoly in the representation of actors of civil  society, and likewise there is no supposition of a mathematical equality among all the individuals who compose the representative body2. On the contrary, the representation carried out by civil society is pluralist. Even when it coincides with a given territory within the structure of a council, it also follows other principles which in general make binding decisions in relation to the same theme within the same territory. In this sense, the representation occurring in civil society most resembles the medieval structure of simultaneous overlapping of diverse types of representation (Gierke, 1987)3, rather than a structure of monopoly characteristic of modernity (Pitkin, 1967; Mansbridge, 2003). Thus, in the great majority of times, the representation of civil society is a process of overlapping of representations lacking either authorization and/or monopoly for the practice of deliberation.

The following question arises due to these news forms brought forth by civil society's action: is this proliferation of forms of social representation a distortion of the very functioning of representation or is this simply a case among many others that has served to re-elaborate the very notion of representation, the others being the forms of the overlapping of representation in the European Community (Held, 1995; Cohen and Sabel, 2005), as well as the international action of non-governmental organizations – NGOs such as Amnesty International or Greenpeace? Judging from the recent proliferation of important literature that reexamines the question from this perspective (Abers and Keck, 2006; Mansbridge, 2003; Urbinati, 2006a; Warren and Castiglioni, 2006; Dryzek and Niemeyer, 2006), the response I offer to the question is that it is worth reexamining the principles of the discussion in light of these new practices. This article will be divided into three parts: in the first section, I will reexamine the basis of the discussions regarding representation as presented by Hanna Pitkin and how it became consolidated in contemporary democratic theory. Furthermore, I will address the main elements of this discussion: the presupposition of authorization, the connection between representation and elections, the idea of monopoly and the argument of territoriality. In the second section, I will critically address three recent analyses that attempt to shed new light on this question: Gurza Lavalle, Houtzager and Castello's attempt to defend a concept of virtual representation based on Burke, Nadia Urbinati's recent attempt to propose a non-electoral form of representation based on Condorcet's idea of temporal extension, and lastly, John Dryzek's attempt to defend an idea of discursive representation. In the last section of this article, I will propose a relational concept of representation, in which I will simultaneously attempt to disassociate representation from authorization and associate it to a shared link among social actors, themes and forums capable of integrating them.



The modern theory of representation is based on three elements: authorization, monopoly, and territoriality. With the intention of reviewing these three elements, I will discuss the idea of authorization in the theory of representation in this section. In her classic book on the subject, Hanna Pitkin adopted a double strategy for reconstructing the concept of representation: on the one hand, she examined the meanings of the term in modernity, dealing with theatrical and legal representation to political representation; on the other hand, she created an institutional and historic account of the manner in which political representation institutionalized itself in modernity. The concept defended by the author will be the result of the intersection of the two strategies of conceptual construction (Warren and Castiglioni, 2006). The first part of Pitkin's work, in which she reconstructs the origin of the term representation in modernity, is strongly based on her reading of Thomas Hobbes. In Leviathan, Hobbes sought to establish the basis of a non-religious concept capable of breaking free from Christian doctrine. The author examined two secular principles for the notion of representation. The first notion comes from Greece, with the idea of prosopon which means the substitution of one person in the theater by another. The second notion comes from Rome with the idea of the procurator in Cicero. In this case, the procurator represents a client while carrying out three distinct roles: "my own, my adversary's, and the judge's" (Cicero, 1942, chapter III: 104-105). The idea of representation in Cicero involves two elements: that of identification and that of authorization. The procurator identifies him/herself with the condition of the represented before representing the latter consequently creating a relationship of affinity between them. Nevertheless, only the element of authorization gained relevance in the manner Hobbes dealt with representation.

In chapter XVI of Leviathan, Hobbes makes the following affirmation: "Of persons artificial, some have their words and actions owned by whom they represent. And then the person is the actor; and he that owneth his words and actions, is the AUTHOR: in which case the actor acteth by authority" (Hobbes, 1997:125). Here we have both the main elements for a theory of representation and, even if less observed by commentaries on Hobbes' work, important elements for a theory of participation. Hobbes introduces the term action to designate all the acts that authors are responsible for, which can be either a direct responsibility or one transferred by an explicit act of authorization. In the case of representation, the central problem is how to obtain possession of the actions of another actor, a debate, which, as Hanna Pitkin points out, generated an important aspect of the discussion regarding legitimacy of power in the XIX century. In this case, Hobbes is only interested in one trace of this aspect, that which gives legitimacy to the act of authorization: "For that which in speaking of goods and possessions is called an owner […] in speaking of actions, is called author. And as the right of possession, is called dominion; so the right of doing any action is called AUTHORITY" (ibidem). In other words, Hobbes reduces the problem of representation to the problem of authorization and creates a perspective within the democratic theory which will concern itself only with one question: does the actor or political agent have the authorization to act in the name of the represented? Without entering into the merit of this question, which has been widely discussed in democratic theory (Manin, 1997), my objective here is to call attention to the fact that this is only one of the questions stemming forth from Cicero's affirmation. Another question posed is: under what conditions can individuals represent other individuals with legitimacy?

There is still a third and fundamental element in Hobbes' theory of representation: it deals with the differentiation between the limited author and the free author. Hanna Pitkin observes passages outside of the Leviathan, in which Hobbes discusses this point and makes the following affirmation: "[…] we use the word [person] vulgarly, calling him that acteth by his own authority his own person, and him that acteth by the authority of another, the person of that other" (Pitkin, 1993, cap. III: 455). In this case, we should return our focus to two different questions: what is the meaning of assuming or renouncing the authorship of certain actions; and how and when should individuals renounce the authorship of some of their actions and when they should not do so. In addition, what are the  types of actions which are more susceptible of provoking the renouncement of authorship and in which of these actions do individuals tend to maintain their stance of authorship. Evidently, this was not a problem that concerned Thomas Hobbes, given that he was only interested in establishing the fact that an act of transference of authorship is a legitimate act, and as such, capable of establishing legitimate sovereign power.

When the actor maketh a covenant by authority, he bindeth thereby the author, no less than if he had made it himself; and no less subjecteth him to all the consequences of the same. And therefore all that hath been said previously of the nature of covenants between man and man in their natural capacity, is true also when they are made by their actors, representers, or procurators […](Hobbes, 1997:126).

Certainly, in what pertains to representation, the Hobbesian problem limits itself to the act of providing the legitimacy of pacts and agreements signed by the representatives of the actors. However, we the authors of late modernity, do not need to stop where Hobbes stopped. In this case, it is possible to see how the author of Leviathan gives us clues as to how to think about two central questions for a theory of civil society participation: the first is that if we introduce democracy as a variable, politics needs both the actor, who acts in a limited manner by a received authority and which we commonly designate as the representative, and the free actor, who instead of delegating the representation of one's acts, decides to become responsible for them. If the actor who acts on his/her own account is acting on behalf of other actors, this does not mean there is no representation, even if in this case it is presented through identification. In this article, I will call this type of relationship as representation by affinity. Nonetheless, before presenting the main elements of this conception of  representation, I will discuss the manner in which the ideas of election, monopoly and territoriality were aggregated to the idea of authorization throughout the debate on representation.

The theory of representation can be divided into two great moments. The first moment is that in which representation takes on a logical-hypothetical side and in which there is no political institution capable of instituting the act of  representation. The debate surrounding this topic, therefore, became reduced to the discussion regarding the legitimacy of the social contract in the process of constituting a government. The social contract, in this case, constitutes a merely hypothetical act. The emergence of the centrality of representation was presented, in a second moment, by the theory regarding the change of those in power, a theory which has republicanism as its origin (Manin, 1997:44-45). As Bernard Manin adequately claims, European Republicanism never worked with the concept of election but with the idea of lot drawing as the founding principle behind the change of individuals in power. Manin shows how the concept of elections was progressively brought to the center of the republican theory, creating a very change in the concept: instead of being concerned with the legitimacy of the change of individuals in power, the theory of representation became concerned with  the fact that the individual who held power had, in fact, the authority of all individuals, transforming representation into a form of government (idem:92)4. However, the author does not deal with a question that gained centrality in the second half of the XX century: the manner in which elections, as an instrument of representation, acquired monopolistic status inside a given territory.

The concepts of monopoly and territoriality are not inherent to the idea of representation. They only became associated to it throughout the process of the consolidation of the modern State. Originally, representative institutions, at the end of the medieval period and in the beginning of the modern era, operated by the overlapping of sovereignty. In essence, such institutions deliberated with regard to a certain aspect of political order and consequently, such a decision would be implemented in diverse places, generating an imposition of sovereignties or forms of representation (Held, 1995). The process by which representation acquired the monopoly of the capacity to deliberate within the political system is linked to the emergence, strengthening and development of the modern State (Tilly, 1986; 1993; Weber, Gerth and Mills, 1958). Throughout this process, which initially took-place in the coercive and administrative spheres, the modern State will become the only institution with capacity to act within a territory. In addition, it is worth noting that the construction of the modern State was not simply a construction of a homogenous State order, but was in fact, also a process of the homogenization of political communities (Anderson, 1991). In each territorial unity, wherein the establishment of a single State entity occurred, there was also the unification of a language and political community (idem). In the case of France, for example, the French Revolution abolished the Provençal provinces (départements) and the Parliament of Provence, which operated until 1789. Hence it is important to understand that there is no conceptual or institutional relation between the transformation of representation into the main form of operation of political institutions and its transformation of authorization in the only form of organization of the political system within modern States. The latter is only associated with the manner according to which the European states unified themselves around a single, homogenous political community.

In this sense, I can conclude this brief digression on representation in modern politics pointing out the fact that, in its origin, it involves the idea of representation by affinity, a dimension gradually substituted by the idea of a monopoly of representation inside a territory. As the monopolist concept entered into a crisis, various authors introduced other types of understanding. Among these types of understanding, it is worth emphasizing virtual representation, the one having a  temporal expansion of representation and the discursive  one. In the following section, I will analyze and critique each one of these concepts before explaining how we might re-construct the concept of representation by affinity.



Various authors in the field of political theory recently pointed out the limits of the way in which representation operates in contemporary democracies and, at the same time, tried to conceive representation in a distinct manner. In this section, I will address, in a detailed manner, three attempts at proposing a new concept of representation: the virtual concept proposed by Houtzager, Gurza Lavalle and Castello; the one of representation beyond the electoral dimension proposed by Nadia Urbinati; and the discursive approach proposed by Johh Dryzek.

The first of the attempts tries to approach the crisis of representation through the perspective of its dual statute. In a recent article regarding the matter, Houtzager, Gurza Lavalle and Castello, relate its present problems to a constitutive duality between the formation of will and its institutionalization. For the authors, modern political history has been dominated by this duality between

[...] the autonomy of the representative versus the mandate of the represented, the legal institutional component of representation versus its substantive or formation of will component, the weight of the delegation or the element of confidence versus the weight of authorization or the element of consent (Gurza Lavalle, Houtzager and Castello, 2006:56, emphases of the original).

Thus, there would be nothing new with respect to the crisis of representation, and for this reason, the authors sought a solution to these problems, which they present in a classic author from the anti-revolutionary thought of the XVIII century, Edmund Burke. There are two fundamental components to Burke's work: the first one derives from his condition of representative of Bristol constituents. In a speech given upon being elected as representative of the city in Parliament, Burke states that:

[...] the happiness and glory of a representative to live in the strictest union, the closest correspondence, and the most unreserved communication with his constituents. Their wishes ought to have great weight with him; their opinion, high respect; their business, unremitted attention. It is his duty to sacrifice his repose, his pleasures, his satisfactions, to theirs; and above all, ever, and in all cases, to prefer their interest to his own. But his unbiassed opinion, his mature judgment, his enlightened conscience, he ought not to sacrifice to you, to any man, or to any set of men living [...] (Burke, 1774). Para. 1-24

Burke's speech on the autonomy of the representatives during their mandate, a position which prevails in political modernity, is exemplar (Pitkin, 1967; Manin, 1997). Notwithstanding, it can hardly serve the objective which Gurza Lavalle, Houtzager and Castello seem to want to use it for, that of defending virtual representation understood as a form of representation "[…] not formally recognized or accepted" (Gurza Lavalle, Houtzager and  e Castello, 2006:89-90). In this case, the authors seem to lose from sight the other dimension in Burke's work in which the concept of virtual representation seems to apply more strongly to: the defense of representation without elections of pre-French Revolution monarchies.

In Burke's most famous book, Reflections on the Revolution in France, there is a second element which aims to radicalize even more his argument during his the speech made to the Bristol constituents of Bristol against a correspondence between the exercise of mandate and election. Sharing the true horror with which the French Revolution caused among European conservative élites, Burke argues not only against the régime, in which the people are the origin of representation, which is the régime emerging in  France, but also against the very idea that the only legitimate monarchy would be the British one, because only it had held the authorization of the representatives (Burke, 1982:55). According to Burke,

[…] At some time or other, to be sure, all the beginners of dynasties were chosen by those who called them to govern. There is ground enough for the opinion that all the kingdoms of Europe were, at a remote period, elective, with more or fewer limitations in the objects of choice. [...] [Today, the kings acquire sovereignty] […] by a fixed rule of succession, according to the laws of his country; and whilst the legal conditions of the compact of sovereignty are performed by him […] he holds his crown in contempt of the choice of the Revolution Society (idem: 56).

It is not difficult to see, from an adequate interpretation of Burke's intention, the great mistake it is to bring this type of discussion to the debate regarding the form of representation instituted by civil society. With all the strength of his argument, Burke is trying to establish the legitimacy of non-electoral representation carried out by the European kings5. There is no doubt that in this discussion, he was defeated and that the British model of the legitimacy of the monarch through Parliament spread throughout all of Europe. This model, today, is limited by the emergence of new phenomena such as actors from civil society or discursive forms of legitimacy. The question, nonetheless, is that in the anxiety of claiming as legitimate a form of post-electoral representation, Gurza Lavalle, Houtzager and Castello end up retrieving an argument for the legitimacy of pre-electoral representation. By acting in such a manner, they throw out the baby with the water and are not capable of proposing a concept of representation that goes beyond authorization via election. I will now examine other more successful attempts at justifying non-electoral representation.

Nadia Urbinati (2006a) paved the way for a second important perspective with regard to this reflection in her recent book on representative democracy, along with various other contributions6. Urbinati attempts to advance in relation to the manner in which Hanna Pitkin approached the concept of representation, theorizing it from different linguistic uses of the term. The author brings back the method originally used by Pitkin and adds to it a new set of uses to the referred vocabulary word. For Urbinati, the concentration of the concept of representation around questions such as authorization and accountability

[....]is no longer satisfactory due to the role ", transnational and non-governamental actors play ... in advancing public policies on behalf of democratic citizens – that is acting as representatives for those citizens. Such actors speak for, act for and can even stand for individuals within a nation-state." (Urbinati,2006b:7).

Therefore, by using the same method as Pitkin, Urbinati sheds light on the non-electoral, yet legitimate, forms of political representation. The question which presents itself is: how to justify the legitimacy of these new forms of representation?

Urbinati brings two important contributions to the debate by trying to justify a wider concept. The first one is to demonstrate that elections are just one out of the multiple dimensions of representation and of the relationship between State and civil society.

The extent to which interests groups write public policies or play a central role in implementing and regulating public policies is the extent to which the division between formal and informal representation has been blurred. (Urbinati,2006b:7).

In this sense, differently from Gurza Lavalle, Houtzager and Castello, Urbinati takes a step forward in this debate by showing that the contemporary problem of representation is associated with the evolution of political practices that make its electoral component relevant but incapable of encompassing the totality of the relationship between social actors and the State.

Urbinati's second contribution to the debate is her attempt to disconnect the relation between sovereignty and representation by showing the inadequate form which Rousseau associated one dimension to the other. According to Urbinati:

[...] the incompatibility theory [between democracy and representation] is the foster child of the modern conception of sovereignty. Its conceptual coordinates  lay at the core of constitutionalism and the theory of government outlined by Montesquieu and Rousseau, the first theorists to explicitly argue (for divergent reasons) for an insoluble tension s between democracy, sovereignty, and representation (Urbinati, 2006a:6).

Urbinati highlights the fact the Rousseau's model regarding the loss of sovereignty would in reality be a privatist model. In the well-known formulation of the "Social Contract," Rousseau states that the individual is either free to exercise his/her own sovereignty or to delegate it to another individual and in doing so, would make this individual a slave. A great majority of the theories on political participation are based on the contrast proposed by Rousseau, which in truth, has as its model not public representation but, in fact, a contractual and private form of rights alienation (Urbinati, 2003). This is the very problem with the critique of representation associated with sovereignty in Rousseau: he is not capable of developing a private model for a public and he binds himself to an elementary form of the non-delegation of sovereignty. Nonetheless, all the forms of participation, even the most direct ones possible, involve the delegation of sovereignty. Hence the question is precisely to think about which forms of participation are political forms.

Both of Urbinati's contributions to the question of representation are inspired by Condorcet and the substitution of the concept of sovereignty by the concept of  political judgement7. This involves the construction of a hypothetical scenario of political realization of representation that can or cannot be confirmed. As such, it requires a wider scope of temporality for the relationship between the representative and the represented, in which elections would only be one aspect (Urbinati, 2006a:199). The new element of criticism in Urbinati's work would be her attempt to integrate elections inside a wider concept of political judgment, which would involve other temporalities, other non-electoral forms of representation, and even the possibility of revoking the authorization granted. Despite her brilliant criticism regarding the limits of the concept of electoral representation, Nadia Urbinati's contribution to the discussion falls short due to one problem: she is not capable of pluralizing the sources that generate political judgment in a way that integrates the forms of participation to the concept she is proposing. Based on Condorcet, she will propose two forms of expanding representation: temporal expansion, through the referendum to revoke a mandate and the possibility of revising laws (idem:205-206). Both proposals are important and already constitute part of the institutional framework of the Anglo-Saxon world. Hence what makes the solution proposed by Urbinati vulnerable is the fact that she is not able to incorporate, into her perspective of political representation, a new institutionality capable of paving the way for either advocacy8 or the representation of civil society.

Among these three authors, John Dryzek was the one who best understood the contradictions of the contemporary form of representation. In his book  Deliberative Democracy and Beyond, the author proposed a relevant differentiation between the representation of people and , interests and that of discourses. By doing so, he sought to differentiate his approach to deliberative democracy from the one proposed by John Rawls, which associates liberal constitutionalism to deliberative democracy. According to Dryzek, the discursive dimension, ignored by liberal constitutionalism, is what needs to be elaborated, though separately (Dryzek, 2000:19). In addition, it is necessary to have a design of the discursive dimension that, beyond elections, would be capable of contemplating new forms of discourse that are not necessarily expressed through electoral mechanisms (idem). In this sense, the initial concern of Dryzek's work is to separate the discursive dimension from the electoral dimension and to think about institutional designs that discursive plurality is capable of generating.

In some later writings, Dryzek deals with the critique of representation as a critique of the exclusivity of the demos, the theme of this article (Dryzek and Niemeyer, 2006). Similar to Urbinati's critique, Dryzek points out that the electoral conception of representation assumes that the demos category, as an aggregation of the totality of individuals, would not be capable of uniting the multiple dimensions of modern politics through suffrage. It is exactly this dimension which is being put in check by the emergence of a plurality of discourses that are not necessarily expressed through electoral mechanisms (idem:6). Dryzek's solution is to think about the possibility of creating a Chamber of Discourses existing alongside the forms of representing individuals. It would be necessary to identify a series of different discourses and make room for  them in a chamber where they would be in opposition against each other. As Dryzek and Niemeyer point out, "Members of the chamber of discourses could not be elected, for then they would be representing constituencies of individuals. Another option would be through random selection of members [of that chamber]" (ibidem). Dryzek's proposal advances in relation to Urbinati's in one important direction, which is that of understanding that new actors and new forms of association put in check the functioning of representation based on a demos as the monopolist form of the aggregation of individuals. Dryzek goes one step further by thinking of a chamber of discourses and, thus, breaking with the Habermasian idea of an informal public sphere of a with non-institutional features, as I have already proposed (Avritzer, 2002). Nonetheless, Dryzek's proposal has three important limitations: in the first place, it separates the representation of individuals from that of ideas, which in my opinion seems rather difficult to achieve. It also ignores the fact that one does not only represent discourses, but also interests, values and ideas. Secondly, just like Urbinati, but erroneously, Dryzek believes that civil society is limited to the advocacy of ideas, when in truth, it has become much more common to see an associativism that is linked to the interests, values, and specific proposal of public policies (Warren and Castiglioni, 2006). The concept of advocacy seems to me insufficient to deal with the vast field of non-electoral representation, since non-governmental actors frequently engage in specific politics, creating new political arenas in which their ideas can be implemented. Thirdly, Dryzek ignores that a great part of the time, that civil society is exercising its role of representation, it is supported by deliberative organisms, with which it shares prerogatives with members of the Executive Power9. Therefore, the creation of a chamber that is solely discursive would not solve the problem of the legitimacy of representation. The question is whether to justify or deny the specific representation that civil society carries out in deliberative arenas. In the following section, I will propose a different way of thinking about the legitimacy of the representation of civil society from the ones discussed up until now.



It seems clear that the starting point to create a wider concept of representation that involves both its electoral and non-electoral dimensions, resides in the discussion of the direct relation between representation and sovereignty. If it is true that both of these concepts are in crises, the two crises are motivated by completely different phenomena. In the case of the sovereignty concentrated in the modern State, everything points to the fact that its crisis is inexorable, being caused by the progressive weakening of the State and the ever increasing role of international institutions in the economy and in international exchanges. In all of these cases, the presence of external actors from beyond the nation-state borders is inevitable (Held, 1995; 2003). Now in the case of representation, the question is how to reconstruct it in a manner that integrates its electoral element into the diverse forms of advocacy and representation that have an extra-electoral origin. It is desirable that an adequate reconstruction of the concept of representation reinforce both its electoral and non-electoral elements.  Given this reason, it is important having as a starting point the fact that the situation wherein to reconstruct representation keep in mind the fact that, from now on, it will operate, henceforth, is one of multiple sovereignties.(Held, 1995). Both Urbinati's contribution to thinking about the political as a continuum, in which elections are a relevant moment, although only a moment, and Dryzek's contribution to thinking about the necessity of institutionalizing new forms of discourse are contributions of interest. However, each one of them has an important deficiency: in Urbinati's case, it is the inability to think about the institutionality of the continuum of representation, and in Dryzek's case, it is the inability to think about non-discursive elements in the new forms of representation. A combination of the contribution of both authors seems to me to be the most adequate.

In order to think of a way of articulating these new dimensions, it is necessary to think about the context in which representation can operate and within which both electoral representation and civil society representation will co-exist. It is also important to understand what the role of authorization is in the creation of legitimacy in this new context. In my opinion, the most important element in this debate is understanding that there are diverse types of authorization related to three different political roles: that of the agent, that of the advocate and that of the participant. In all three cases, there is the element of "acting in place of," which was quite emphasized by Hanna Pitkin. Nevertheless, it is important to understand that "acting in the place of" varies according to the perspective and can be justified in different manners. This element, in the case of the agent chosen by electoral process – the classic case of representation –, does not need further discussion in this article (Pitkin, 1967). However the recent changes in the last two cases are essential. Therefore, it is worth discussing their legitimacy.

The case of the advocacy of collective causes goes beyond the discussions made by Hobbes and Hanna Pitkin regarding the role of the advocate or activist. Until quite recently, the advocate was chosen by the person or group of people and would act according to the precise instructions of these actors. In the last decades, a new concept of advocacy of public or private causes emerged, which has done away this dimension. Non-governmental organizations, which work on behalf of causes outside of their own nation-state, defend actors that did not indicate them for such a role, as is the case of Amnesty International or Greenpeace. In this respect, the advocacy of themes seems to do away with the choice or with any other type of authorization. There are still more problematic cases for a theory of representation, such as those in which some organizations of women's rights-defend the autonomy of women in countries where they do not have such rights, and if consulted, those women might claim they are not in favor of such rights (Kandiyoti, 1991). In all of these circumstances, it is not authorization but affinity or the identification of a group of individuals with a situation experienced by other individuals that legitimizes the advocacy. So one can say that a North-American or European woman has a relationship of identification with the situation experienced by an Indian or Muslim woman, but certainly does not have the authorization to represent such a woman. At best, one can assume that, under open information exchange conditions, the actors involved would have different positions in relation to their own rights, which, in any event, all cases, is only a supposition. In this case, the central element of advocacy of themes is not authorization but rather, a varying relation in its contents  caused by a changing relation between actors and their representatives. If we return to Cicero and his description of the role of the procurator, we can see that the identification with a cause became more important than the explicit authorization to represent the cause. In this case, what the international NGOs are representing is a discourse on the rights of women in general and not a group of specific people.

The third case is that of the representation of civil society. This representation, which has become very strong in areas of public policies in the developing world, occurs as of given the specialization of NGO's in themes and practical experiences. Organizations created by civil society actors and that deal for a long time with a problem in the area of social policies tend to take on the role as the representatives of civil society in councils or in other organisms responsible for public policies. This situation is different from the other two: on the one hand, there are often elections for these representatives, particularly in Brazil, but the electorate has very specific characteristics10. There is a group, in which one finds the origin for the representation exercised by these representatives, but this group may or not include  all the associations related to the theme. In addition, this group may not even be organized in associations. In one case, we are speaking of an almost collective form of representation, and in the other case, we are speaking of a collective and non-institutionalized form of action that generates representation. This last case does not have the characteristics of the mathematical equality of sovereignty, so important to the idea of electoral representation, and does not have the monopolist territorial element, given that it shares its capacity for decision-making with other institutions present in the territory. What is important in relation to this kind of representation is that it has its origin in a choice among civil society actors, frequently decided upon within civil associations. These associations play the role of creating intermediate affinities. In other words, they aggregate solidarities and partial interests (Warren, 2001). By aggregating these interests, they allow for a form of representation by choice, which is different from the electoral representation of individuals or people. The difference between representation by affinity and electoral representation is that the first legitimizes itself in a partial identity or solidarity that occurred previously and led to a constitution of a specific forum. However, it may also include the electoral appointment of civil society representatives within this specific forum.

What provides the legitimacy for representation by affinity? It is provided by the legitimacy of the representative among other actors who act in the same manner. In this sense, the question of partial identities in politics takes on a new role, which was, to some extent, abolished by modern politics11. The pragmatics of legitimacy is different, to the extent that the legitimacy is given by a relation with a theme that generates a specific political body. It is this that relation that generates legitimacy and not the opposite, as in the case of electoral representation. However, among the actors who constitute these specific fora, elections also provide for inter forum legitimacy. Table 1 intends to summarize the different forms of representation discussed here:

If we think about these three aspects of representation, it is possible to see how one can theorize elections in a different manner. In the first place, electoral representation should mean the access to a frame of relationships among different types of sovereignty (Young, 2000)12. In this sense, elections decide the manner in which representative bodies will relate to advocacy and the representation of civil society. This relation can be more or less complementary, depending on the politically proposal elected, even though in Brazil the relation between electoral and non-electoral representation has been one of the most common elements with the last governments. In the case of Brazil, elections have also determined the manner in which one type of representation is capable of legitimizing another. Thus, in Fernando Henrique Cardoso's government, the presidents of the national councils were indicated by the president. Now in Luiz Inácio Lula da Silva's government, the presidents of the national councils are elected by civil society (Avritzer, 2008). This shows that a form of representation may both lend legitimacy to another, as well as question it.

An important aspect of these new forms is that they do not appear pure in contemporary politics. Elections continue to be the most democratic means of choosing representatives, but, once elected, these representatives encounter the advocacy of themes and the representation of civil society. Those representatives who ignore this representation, be it within the national space, or in the international one, tend to de-legitimize themselves among their own electorate and have been frequently incapable of implementing their own agenda13. Therefore the encounter between elected representatives and the advocacy of international NGOs or between elected representatives and representatives of civil society in hybrid institutions is becoming more and more common (Avritzer and Pereira, 2005) in the field of public policies.

These encounters demonstrate that, differently from what Urbinati assumed, the continuum of politics takes on diverse institutional forms that should be part of the discussion. In addition, these encounters concurrently place elected representatives in diverse processes that could be both thematic or interest based, differently from what Dryzek assumed. In this sense, the question posed by contemporary politics should be one of a reduced concern about the legitimacy of these news forms of representation and one of an increased concern about the way in which they should be apply to in a political system governed by multiple sovereignties. The future of electoral representation seems to be increasingly tied to its overlapping with the forms of representation that have their origin in the participation of civil society.




1. This data, referring to 2004, was obtained by adding the adhesion of the participatory budgets in the cities of São Paulo (80 thousand people), Porto Alegre (30 thousand), Belo Horizonte (30 thousand) and Recife (40 thousand). The fact that almost 200 thousand people were involved in participatory politics shows that participation is a relevant form of exercising political sovereignty in Brazil today.

2. This is a founding element of the theory of representation dating back to its origin. Representation appears in Locke and even in Rousseau as associated to a principle of a mathematically established equality, on the basis of which the vote of each individual has exactly the same weight. See Rousseau (1997).

3. Otto Gierke (1987) called attention, for the first time, to the fact that the structure of sovereignty at the end of the medieval period was an overlapping structure of sovereign entities. The State, local governments, and institutions such as the Catholic Church would simultaneously decide on distinct issues  within  the same territory without any of the institutions claiming total monopoly of sovereignty in that given territory. The association between territory and monopoly only appeared with the modern State. David Held (1995) recently observed that a return to the medieval concept of superimpositions/overlapping of sovereignties, is due to the creation of the European Community.

4. Discussing the debate regarding representation is not the same as discussing the virtues and the problems of representative government. The difference resides in the question of the monopoly of representation among those who defend representative government as the only form of government. Bernard Manin's study concentrated more on the second dimension, which is understood as the autonomy of those governing in relation to the will of the represented (Manin, 1997:6). Nadia Urbinati criticizes the reduction of the debate on representation to the functioning of representative government, by affirming that it is incorrect to assume that the singularity of representation resides in elections. According to Urbinati, these are parts of the process of establishing representation, and in this sense, representative government cannot be reduced to electoral representation. See Urbinati (2006a:9).

5. It is also worth noting, that during the process of independence in the United States, the British crown used the concept of virtual representation to defend the idea that the interests of the citizens in the thirteen colonies were being represented in the British Parliament (see Wood, 1969:180). The British discussion shows the correction of the reconstruction of the concept of representation by Manin, which assumes the identity among representation and authorization for all representatives.

6. Also see a series of articles published in the journals Political Theory and Constellations. In those articles, Urbinati dealt with questions such as advocacy and representation and she criticizes of the model of representation presented in Rousseau. See Urbinati (2000; 2003; e 2006b).

7. One of the peer reviewers of this article disagreed with the argument presented here, holding that the substitution of the concept of will for the concept of judgment proposed by Urbinati would not imply in the disassociation between sovereignty and representation. This author disagrees with this interpretation not only because Urbinati the author explicitly affirms this (Urbinati, 2006a:6) but also because it is very difficult for the concept of judgment to articulate itself with that of sovereignty, given that the latter demands an explicit authorization to "act in the place of another". The concept of judgment, given its temporal extension, implies that each citizen place him/herself in the place of the sovereign and judge him/her. In this sense, there is indeed a disassociation between sovereignty and representation. See Urbinati (idem:105).

8. One of the peer reviewers of this article correctly suggested that the translation of the term advocacy does not have the same meaning in the Portuguese language. He or she suggests, in its place, the use of the terms militancy or activism. Despite the correction of the linguistic observation, I chose to maintain the term advocacy throughout most of this text, because militancy or activism in the Portuguese language seems to be more closely linked to the intensity of certain forms of political action of the left than the manifestation of ideas or actors. In certain passages in which advocacy seemed to me to be completely inadequate, I added the term activism.

9. This is without a doubt the case of Brazil, but it seems to be the case in the developing world in general. Forms of participation of civil society in Peru, in Argentina, in India, and even in the United States, in the so-called habitat programs, function in the same manner. The exception, which may be the case Dryzek had in mind, are the parallel meetings of the United Nations, in which civil society meets separately from the organisms that exercise the representation of countries. See Panfichi (2003).

10. There are different cases of elections that should be emphasized, such as the case of the elections of the housing council in São Paulo, during Marta Suplicy's government, in which more than 30 thousand people voted. There are also cases in which some councils established into norm what is a representative of civil society, such as the health council of the city. See Avrizter (2004).

11. Until the beginning of the modern era, all forms of representation of interests were particular by definition. The different forms of corporative representation that survived in some countries in Europe, until the beginning of the XIX century, are a good example of forms of particular representation. The modern State dissolved these forms in the representation of individuals, believing that this would decrease particular interests.

12. Iris Young, in her book Inclusion and Democracy, dealt with the idea of representation as a relation, but in a distinct manner from that which we are proposing here. According to Young, the relation of different types involved in representation limits itself to different types of relations between the representative and his/her electorate. See Young (2000: 128).

13. Among the most important examples, we can emphasize both the question of the rights of women in many countries of the Arab world and environmental questions in various countries, among which Brazil. The public audiences, introduced in Lula's government, with regard to questions with a strong environmental impact, such as the construction of the highway BR-163 and the transposition of the waters of the San Francisco river, are an example of the necessity of elected governments to legitimize themselves in order to implement policies in areas such as the environment.

* Revised by Gilda Stuart




ABERS, Rebecca N. and KECK, Margareth E. (2006), "Muddy Waters: Decentralization, Coordination and Power Struggle in the Brazilian Water Management Reform". International Journal for Urban and Regional Research, vol. 30, nº 3, pp. 601-622.

ANDERSON, Benedict. (1991), Imagined Communities: Reflections on the Origin and Spread of Nationalism (2ª ed.). London/New York, Verso.

AVRITZER, Leonardo. (2002), Democracy and the Public Space in Latin America. Princeton, Princeton University Press.

___ (ed.). (2004), A Participação em São Paulo. São Paulo, Editora Unesp.

___. (2006), "New Public Spheres in Brazil". International Journal of Urban and Regional Research, vol. 30, nº 3, pp. 623-637.

___. (gone to press), A Participação Social no Nordeste. Belo Horizonte, Editora da UFMG.

___  e PEREIRA, Maria de Lourdes Dolabela. (2005), "Democracia, Participação e Instituições Híbridas". Teoria & Sociedade, special issue, pp. 16-41.

BURKE, Edmund. (1774), Speech to the Electors of Bristol at the Conclusion of the Poll.  Available at:, accessed on 15/7/2007.

___. (1982), Reflexões sobre a Revolução em França. Brasília, Editora da UnB.

CÍCERO. (1942), Oratorium. Cambridge, Loeb Classical Library.

COELHO, Vera Schattan R. P. (2004), "Conselhos de Saúde enquanto Instituições Políticas: O Que Está Faltando?", in V. S. R. P. Coelho e M. Nobre (ed.), Participação e Deliberação: Teoria Democrática e Experiências Institucionais no Brasil Contemporâneo. São Paulo, Editora 34, pp. 255-269.

___ et alii. (2006), Participação e Distribuição de Serviços Públicos de Saúde no Município de São Paulo. Relatório de pesquisa. Cebrap, São Paulo.

COHEN, J. and SABEL, Charles. (2005), "Global Democracy". NYU Journal of International Law and Politics, vol. 37, nº 4, pp. 763-797.

CUNHA, Eleonora S. M. (2004), Aprofundando a Democracia: O Potencial dos Conselhos de Políticas e Orçamentos Participativos. Dissertação de mestrado. Departamento de Ciência Política, UFMG, Belo Horizonte.

DRYZEK, John. (2000), Deliberative Democracy and Beyond. Oxford, Oxford University Press.

___ and NIEMEYER, Simon. (2006), Discursive Representation. Paper presented at  Rethinking Democratic Representation workshop. University of British Columbia,  18th-19th May.

GIERKE, Otto. (1987) [1960], Political Theories of the Middle Age. Boston, Beacon Press.

GURZA LAVALLE, Adrian, HOUTZAGER, Peter and CASTELLO, Graziela. (2006), "Representação, Pluralização da Representação e Sociedade Civil". Lua Nova, nº 67, pp. 49-103.

HELD, David. (1995), Democracy and the Global Order. Cambridge, Polity Press.

___. (2003), Global Covenant: The Social Democratic Alternative to the Washington Consensus. Cambridge, Polity Press.

HOBBES, Thomas. (1991), Leviatã ou Matéria, Forma e Poder de um Estado Eclesiástico e Civil. São Paulo, Abril Cultural.

KANDIYOTI, Deniz. (1991), "Women, Islam and the State". Middle East Report, nº 173, pp. 9-14.

MANIN, Bernard. (1997), The Principles of Representative Government. Cambridge, Cambridge University Press.

MANSBRIDGE, Jane. (2003), "Rethinking Representation". American Political Science Review, vol. 97, nº 4, pp. 515-528.

PANFICHI, Aldo. (2003), Sociedad Civil, Esfera Pública y Democratización en América Latina: Andes y Cono Sur. México, Fondo de Cultura Económica/Universidad Veracruzana.

PITKIN, F. Hanna. (1967), The Concept of Representation. Berkeley, University of California Press.

___. (1993), "Hobbes Concept of Representation", in P. T. King (ed.), Thomas Hobbes: Critical Assessments: Politics and Law. London, Routledge (vol. 3).

ROUSSEAU, Jean-Jacques. (1997) [1762], The Discourses and Other Early Political Writings. Cambridge, Cambridge University Press (Ed.Victor Gourevitch).

TILLY, Charles. (1986), The Contentious French. Cambridge, Harvard University Press.

___. (1993), Coercion, Capital and European States: AD 990-1992. Oxford, Blackwell (Studies in Social Discontinuity Collection).

URBINATI, Nadia. (2000), "Representation as Advocacy: A Study of Democratic Deliberation". Political Theory, vol. 28, nº 6, pp. 758-786.

___. (2003), "Continuity and Rupture: The Power of Judgment in Democratic Representation". Constellations, vol. 12, nº 2, pp. 194-222.

___. (2006a), Representative Democracy. Chicago, The University of Chicago Press.

___. (2006b), Representation. Paper presented at Yale Political Theory workshop. Yale University, 9th November.

WARREN, Mark. (2001), Democracy and Associations. Princeton, Princeton University Press.

___ and CASTIGLIONI, Dario. (2006), Rethinking Representation: Eight Theoretical Issues. Paper presented at Rethinking Democratic Representation workshop. University of British Columbia, 8th-19th May.

WEBER, Max, GERTH, Hans H. and MILLS, C. Wright. (1958), From Max Weber: Essays in Sociology. Oxford, Oxford University Press.

WOOD, Gordon S. (1969), The Creation of the American Republic, 1776-1787. Chapel Hill, The University of North Carolina Press.

YOUNG, Iris M. (2000), Inclusion and Democracy. Oxford, Oxford University Press.