Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The democratic legitimacy of 'self-appointed' representatives Montanaro, Laura 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_fall_montanaro_laura.pdf [ 955.42kB ]
Metadata
JSON: 24-1.0071339.json
JSON-LD: 24-1.0071339-ld.json
RDF/XML (Pretty): 24-1.0071339-rdf.xml
RDF/JSON: 24-1.0071339-rdf.json
Turtle: 24-1.0071339-turtle.txt
N-Triples: 24-1.0071339-rdf-ntriples.txt
Original Record: 24-1.0071339-source.json
Full Text
24-1.0071339-fulltext.txt
Citation
24-1.0071339.ris

Full Text

THE DEMOCRATIC LEGITIMACY OF ‘SELF-APPOINTED’ REPRESENTATIVES  by  Laura Montanaro  B.A. (Specialized Honours), York University, 2000 M.A., York University, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate Studies  (Political Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2010  © Laura Montanaro, 2010  Abstract Standard accounts of democratic representation involve both the authorization of a representative by election, and the accountability of elected officials to their constituents for their performance in office. Yet actors such as Oxfam, Amnesty International, and the musician Bono, who make representative claims outside of formal representative institutions – who “selfappoint” – are an increasingly important part of today’s political landscape. On most standard accounts of democratic representation, the absence of formal authorization and accountability renders such activities non-democratic, regardless of any good achieved. Yet the case for their credentials is rooted in a norm that is at the heart of most contemporary democratic theories: those potentially affected by a collective decision should have some say in making that decision. From an empirical perspective, there is a need for a theory of representation that will identify the types of self-appointed representatives that, although unelected, comprise growing and important parts of our political landscape. From an analytic perspective, there is a pressing need for criteria that will allow us to distinguish legitimate from illegitimate claims of self-appointed representatives. To develop the theory required, I develop a general account of representation that identifies representative relationships apart from electoral representation. Because this framework interrupts the close association of representation with elections, we are able to conceptualize actors who claim to represent by self-appointment as representatives, and perhaps even as democratic ones. Furthermore, viewing representation separately from electoral institutions expands our understanding of constituency to include peoples who do not neatly fall within the boundaries of electoral districts but who are affected by their law and policy. In fact, this is where the potentially democratic credentials of self-appointed representation are to be found: in its ability to identify and mobilize affected constituencies around claims of ii  representation. I also conceptualize non-electoral mechanisms of authorization and accountability that may be used to guide, inform, and sanction the self-appointed representative. Understanding the concepts of representation and constituency in this fluid way is a necessary step in developing a democratic theory that is appropriate to the complex, globalizing, pluralistic, and highly differentiated societies within which we now live.  iii  Table of Contents Abstract ...................................................................................................................................................ii	
   Table of Contents .................................................................................................................................iv	
   List of Tables .........................................................................................................................................vi	
   Acknowledgements..............................................................................................................................vii	
   Chapter One: Introduction ................................................................................................................. 1	
   Democracy and its Norms: Political Equality and Self-Determination ............................................ 12	
   Democracy Transformed: Representative Democracy, Self-Determination, and Affected Interests........................................................................................................................................................... 14	
   The Standard Contemporary Account of Democratic Representation ............................................ 16	
   What the Standard Account Does Not Cover ......................................................................................... 19	
   Self-Appointed Representation and Self-Determination...................................................................... 22	
   Chapter Outline............................................................................................................................................. 28	
    Chapter Two: Democracy and Representation ........................................................................... 33	
   Hobbes: Representation as Authorization............................................................................................... 34	
   Burke and Virtual Representation............................................................................................................ 39	
   Pitkin’s Innovation: Representation as “Acting For”........................................................................... 44	
   Representation versus Strong Democracy............................................................................................... 47	
   The Question of Constituency .................................................................................................................... 50	
   Representation, Responsiveness and Empowered Inclusion ............................................................... 58	
   Representation and Civil Society............................................................................................................... 62	
   Conclusion ...................................................................................................................................................... 64	
   Chapter Three: The Concept of Self-Appointed Representation ............................................ 66	
   The Representative Claim........................................................................................................................... 67	
   The Representative Claim in Formal Representation .......................................................................... 69	
   The Representative Claim in Informal Representation ....................................................................... 73	
   The Representative Claim in Self-Appointed Representation ............................................................ 76	
   The Democratic Legitimacy of Representative Claims ........................................................................ 81	
   How Self-Appointed Representatives Might Advance Democratic Norms...................................... 84	
   Conclusion ...................................................................................................................................................... 90	
   Chapter Four: The Concept of Constituency............................................................................... 92	
   The Standard Model of the Demos and its Formation ......................................................................... 96	
   The Affected Interests Principle and Constituency.............................................................................102	
   The Self-Appointed Representative and Democratic Constituency Formation............................112	
   Non-Democratic Constituency Formation.............................................................................................116	
   Conclusion ....................................................................................................................................................120	
   Chapter Five: ‘Who Elected Oxfam?’ Non-Electoral Authorization and Accountability ..............................................................................................................................................................121	
   Electoral Authorization and Accountability .........................................................................................124	
   Beyond Electoral Authorization and Accountability ..........................................................................125	
   Non-Electoral Sources of Authorization ................................................................................................132	
   Sources of Accountability..........................................................................................................................139	
   Conclusion ....................................................................................................................................................149	
   Chapter Six: Self-Appointed Representation in Practice........................................................151	
   Democratic Self-Appointed Representatives ........................................................................................152	
   iv  Surrogate Self-Appointed Representatives ...........................................................................................157	
   Skewed Self-Appointed Representatives................................................................................................165	
   Failed Self-Appointed Representatives ..................................................................................................169	
   Conclusion ....................................................................................................................................................173	
    Chapter Seven: Conclusion............................................................................................................175	
   The Future of Democratic Theories of Representation ......................................................................182	
   Formal Political Roles for Civil Society Organizations ............................................................................... 182	
   Citizen Representatives ............................................................................................................................................ 185	
   Tacit Consent............................................................................................................................................................... 186	
   Representing Future Generations ........................................................................................................................ 187	
   Representative Systems ............................................................................................................................................ 189	
   Concluding Remarks..................................................................................................................................190	
   References ..........................................................................................................................................192	
    v  List of Tables Table 3.1 Dimensions of Self-Appointed Representation………………………………………89 Table 5.1 Electoral and Self-Appointed Authorization and Accountability…………………...148 Table 6.1 Dimensions of Self-Appointed Representation……………………………………..151  vi  Acknowledgements I am grateful to my committee, Mark Warren, Laura Janara, and Bruce Baum, who have been unfailingly supportive and encouraging. I am lucky to have Mark Warren as my supervisor and mentor. Mark provides a model for quality, nuance, and detail that will forever be the standard against which I measure my own work. He has seen this work in all its iterations and somehow remained interested throughout. Mark, to watch your mind at work has been a privilege and a pleasure. Laura Janara sets an example as a theorist, teacher, and human being to which I will forever aspire. Her dedication to her students’ work and wellbeing, her unwavering faith in my abilities, her careful attention – I need a better phrase than “thank you.” From very early on, Bruce took the time to recommend books and articles to read, always encouraging and gently critical. His first concern was always that I enjoyed what I was doing. Quite simply, the best committee I could have hoped for. There are many faculty members at UBC who, although I did not take a class with or, in some cases, even share subfields with, still expressed interest in the work and the student. Barbara Arneil, Ken Carty, Alan Jacobs, Richard Johnston, Chris Kam, Sam LaSelva, Ben Nyblade, Angel O’Mahony, Richard Price, and Lisa Sundstrom all took time to discuss the work and/or the career. They may have forgotten, but I have not. I am grateful also for the financial support of the Department of Political Science, and the Social Sciences and Humanities Research Council of Canada. To my friends and colleagues, a tremendous thanks for countless dinners, movies, and conversations. This experience is made rich not only by people who teach you, but the people you learn with. Finally, to my family, who somehow knew not to ask too many questions – an essential quality in the family of a Ph.D. student – and who never failed to let me know how proud they are. Even though it took six years.  vii  Chapter One: Introduction Democratic representation is usually taken to involve both the authorization of a representative by election and the accountability of elected officials to their constituents for their performance in office. Yet there are an increasing number of individual and collective actors claiming to represent others despite being neither formally authorized nor formally held to account by those they claim to represent. In this sense, these actors are “self-appointed.” Martin Luther King, Jr., for example, claimed to represent African Americans suffering from racial injustice. We might say that King initially “self-appointed”: although he was selected by his congregation and activists to lead, he asserted his broader representative claims outside of formal representative institutions; an electoral constituency did not authorize him; and he was disconnected from formal political authority. To take another well-known example, the musician Bono claims to represent the interests of Africans on the issues of AIDS, debt and trade. As Bono himself pointed out, he has not been elected by any Africans, and is he not formally accountable to them. Likewise, the National Organization for Women takes action to bring about equality for all women – and so arguably claims a representative status beyond its own members. But on what grounds do they represent all women? And the non-governmental organization, Oxfam, presses decision-makers and governments for fair trade rules on behalf of the world’s poor. What entitles them to speak and act for the poor? In this dissertation, I am concerned primarily with the question asked by The Economist: “Who Elected Oxfam?” (2000, 86). The Economist asked this question rhetorically, perhaps facetiously, assuming that “no one elected Oxfam” is both the answer to their question and also the last word about Oxfam’s representative credentials. In this dissertation, I take this question seriously and treat it as the introduction to a family of problems in democratic theory and practice having to do with representatives who might credibly claim democratic credentials,  1  though not as a consequence of formal elections. On most standard accounts of representation, the absence of formal authorization and accountability renders the claims and activities of actors such as King, Bono, NOW, and Oxfam as non-democratic, regardless of any good achieved. Yet, as I shall argue, the case for an actor’s credentials as a democratic representative can and should be developed out of a basic normative intuition that is at the heart of most contemporary democratic theories: those potentially affected by a collective decision should have capacities to influence that decision. From this perspective, actors who claim to represent by selfappointment seem, prima facie, of central importance to democracy – particularly in a complex, globalizing world. Bono, for example, meets with leaders of the G8 countries, who in turn promise to cancel debt for some of the world’s poorest countries. The World Bank has included non-governmental organizations such as Oxfam in their multilateral debt relief discussions. And the National Organization for Women is a “significant and visible presence in Washington politics,” providing “an institutionalized voice to and compensatory representation for the concerns of formerly excluded groups that still have insufficient formal representation in national politics” (Strolovitch 2007, 17). Such self-appointed representatives serve to make these kinds of constituencies politically present, providing a prima facie case for their own importance to the depth and breadth of democracy and representation. If the alternative to selfappointed representation is the political exclusion of historically marginalized groups from these decisions, then there is a need to rethink our received understandings of democratic representation. If self-appointed representatives function to make a constituency politically present, and if I can find mechanisms through which constituencies both authorize and hold accountable those who self-appoint, then I will have made a case for including this important phenomenon within democratic theory. That said, the above examples demonstrate that although we occupy a political landscape in which decision-makers receive actors who self-appoint as representatives of such groups, for 2  the most part we do not understand what it means for groups and individuals to function as democratic, or otherwise legitimate, representatives outside of the formal procedures of electoral authorization and accountability. Actors who represent by self-appointment, such as civil society organizations, non-governmental organizations, foundations, and individual actors, operate as representatives in a complex world, one that is filled with political exclusions either where there is an absence of democracy or where formal democratic institutions do not represent those who have legitimate claims for inclusion. But we lack adequate theoretical frameworks for identifying what these actors are doing when they make representative claims, and we lack the criteria for assessing their legitimacy. As such, there is a pressing need for theory – empirical as well as normative – that will allow us to expand our understanding of the phenomenon of actors who claim to represent by self-appointment, and to distinguish legitimate from illegitimate claims. Generally speaking, democratic representation is rendered clearly recognizable when it is based on the formal relationship of authorization and accountability, and we have articulated criteria that enable us to judge its legitimacy. The fact of fair election is conventionally taken as establishing these relationships, and doing so in a way that is clearly and easily recognizable. As Joseph Schumpeter argues, democracy is a system “for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote” (1947, 269). However, once the representative claims of self-appointed, putative, representatives – especially those claiming to represent democratically – are removed from the context of electoral institutions, legitimacy becomes far more difficult to determine. The question I shall hope to address in this dissertation is this: In the absence of formal authorization of a representative by election, and in the absence of accountability normally established by the reward of re-election or the sanction of removal from office, how do we identify as “representative” and assess as “legitimate” the claims made by actors who appoint themselves as representatives through their claims? How might the concept and criteria of 3  democratic representation be brought to bear on the arena of self-appointed representation? If the appraisals entailed in the notion of democratic representation can speak to the possibility of self-appointed representation, we must theoretically identify the phenomenon – its nature and potentials – and then develop criteria of judgment. The strategy I adopt is relatively straightforward. With respect to representation that will count as democratic, I employ a version of the affected interests principle: representation is “democratic” to the extent that those whose interests are affected by decisions have some capacity to influence those decisions (proportional to the extent to which they are affected) (Goodin 2007). To conceptualize the relevant domain of “affected interests” – what it means to be affected in a way that is democratically relevant – I shall use a definition of “interests” borrowed primarily from the work of Iris Young (2000). As Young develops the concept, interests are defined as “what affects or is important to the life prospects of individuals” (2000, 134), and are different from ideas, principles, and values, which help define the ends a person sets for herself; interests define the means for achieving those ends (Young 2000, 134). I suggest that interests can be identified with the help of the criteria of self-development and selfdetermination. Self-development pertains to both the distribution of resources and also “the institutional organization of power, status, and communication” (Young 2000, 32; see also Young 1990; Sen 1990, 1992). Self-determination involves “participation in making the collective regulations designed to prevent domination” (Young 2000, 33; see also Pettit 1997). Young suggests that the opposites of self-development and self-determination are oppression and domination, respectively. Simply put, we worry about those effects that result in, or are a consequence of, institutionalized oppression and/or domination of interests. Thus, decisions count as relevant to democracy just insofar as they affect the interests of individuals, as defined by conditions of self-development and self-determination. I view representation as a set of relationships through which democracy in this sense can 4  be advanced. My strategy is to borrow the basic concepts employed in electoral theories of representative democracy – concepts such as authorization, accountability, responsiveness, and constituency – and then to ask what these kinds of criteria might mean in the context of those who claim to represent by self-appointment. I will use these familiar criteria for democratic representation to help determine whether, and if so how, such features can be cast off from their electoral institutional forms to operate otherwise in structurally different modes of political representation. As part of this project, I suggest that there are often non-electoral mechanisms of accountability, authorization, responsiveness, and constituency formation at work which, though not based on formal electoral institutions, may contribute to the democratic, or otherwise legitimate, representative credentials of these self-appointed representatives. When these mechanisms contribute to enhancing the conditions of self-development and self-determination of groups and/or individuals, they enhance “democracy.” Using these basic concepts, I theorize an alternative, non-electoral, view of democratic representation at the level of the nation-state as well as globally. I argue that at the level of the nation-state, self-appointed representation may be an important complement to democratic electoral representation because of its potential to be both responsive to constituencies whose self-development and self-determination are affected by decisions and actions, and to also promote their empowered inclusion – constituencies to which electoral representation cannot or does not provide such responsiveness and empowerments. For example, a self-appointed representative such as the National Gay and Lesbian Task Force is responsive to a (nonelectoral) constituency of people (lesbian, gay, bisexual, and transgendered peoples, or LGBT) by working to attain equal rights legislation so that they are full participants in America’s democracy (Task Force 2010). The Task Force is working to improve conditions of selfdevelopment of LGBT peoples. Self-development, as defined by Iris Young, “entails meeting people’s needs for food, shelter, health care, and so on,” and also “using satisfying skills and 5  having one’s particular cultural modes of expression and ways of life recognized depend on the organization of the division of labor and the structures of communication and cooperation” (2000, 32). As Young develops the concept, the opposite of self-development is oppression. By working to end job discrimination, one of its stated goals, the Task Force is responsive to conditions required for the self-development of LGBT peoples whose interests are affected by institutionalized oppression. The Task Force is also working to improve conditions of selfdetermination of LGBT peoples. Self-determination “consists in being able to participate in determining one’s action and the condition of one’s action; its contrary is domination” (Young 2000, 32). By working to ensure civil rights legislation, the Task Force is responsive to the selfdetermination of LGBT peoples (in a way that many electoral representatives currently are not) whose interests are affected by institutionalized domination. I mean this formulation to suggest that effects of institutionalized oppression and/or domination of basic interests produce “affected” peoples as potential “constituencies,” which may then be identifiable and represented by those, like the Task Force, who make claims on behalf of such constituents. Responsiveness to affected interests entails an effort to improve conditions of self-development and selfdetermination. Furthermore, the Task Force promotes the empowered inclusion of the affected constituency by encouraging membership in their organization, as well as participation in rallies, protests, and petitions, so that LGBT peoples may guide, inform, and sanction, that is, authorize and hold accountable, the behaviour of the Task Force. Following this conceptual strategy, I will argue that these two features of representation – responsiveness to, and empowered inclusion of, an affected constituency (a constituency whose conditions of self-development and self-determination are at issue) – if present, enable us to identify an actor who self-appoints as a democratic representative. Beyond serving as a complement to democratic electoral representation, these criteria may be extended to (a) contexts where there is an absence of electoral representation, within states without electoral 6  democracy, within those domains – or for those constituencies – in electoral democracies that are missed by electoral representation, and (b) at the regional and global levels, where the conditions for state-based forms of electoral democracy do not exist. In all of these contexts, self-appointed representatives may function to improve conditions of self-development and selfdetermination of those whose interests are affected by collective decisions – although, of course, they do not necessarily do so. If successful, the theory I develop in this dissertation should allow us to identify and assess these self-appointed representatives as more or less democratically legitimate. If a self-appointed representative is responsive to constituencies whose interests are affected with respect to their self-development and/or self-determination (affected constituencies), then we can begin to talk about a legitimate representative relationship, whether or not elections are involved. And if that actor, in addition to being responsive to affected constituencies, promotes their empowered inclusion, then we may consider that representative relationship as contributing to democracy. What I am arguing, then, is that democratic theorists should not automatically dismiss the democratic potentials of self-appointed representatives, just because they are self-appointed rather than formally elected – nor, indeed, should The Economist or anyone else. We should, at least in the first instance, follow our intuitions that actors like Martin Luther King, Jr. and Bono likely make the world a better place for constituencies of people who would otherwise be excluded from any influence over decisions that affect their interests. And then we can theorize how and why. Martin Luther King, Jr., for example, was responsive to the interests of African Americans, seeking to improve conditions of self-development and self-determination by working on their behalf to end racial injustice. He also promoted their empowered inclusion by providing them with opportunities to authorize his activities (e.g. participating in boycotts, demonstrations, protests), and hold him accountable (e.g. exit the Southern Christian Leadership Conference). Dr. King was both responsive to and also promoted the empowered inclusion of a 7  constituency whose interests were affected by racial discrimination and so can be recognized as a democratic representative. Bono’s contributions differ, or so I shall argue based on the theory I develop in this dissertation. Bono works to hold world leaders to account for the commitments they make to fight extreme poverty, but he does so without promoting the empowered inclusion of the affected constituency in the manner of Dr. King. Bono is responsive to, but does not promote the empowered inclusion of, an affected constituency. I shall suggest that Bono is, indeed, a self-appointed representative, and that in speaking for the poor of Africa, he is responsive to conditions of self-development and/or self-determination. However, the affected constituency lacks the empowerments that would enable them to authorize him and hold him accountable for his claims. Bono is responsive to the interests of an affected constituency but because the affected constituency is not empowered to authorize him and hold him accountable, I shall refer to him as a surrogate representative rather than a democratic one. Likewise, with the case of Oxfam: the organization presses decision-makers to change policies and practices that reinforce poverty and injustice, but for the most part they do so without the empowered inclusion of affected constituencies (though they do work with local partner organizations). The threshold for “democracy” is high, but if the theory I develop here does its job, then we should be able to talk about legitimate representation short of democracy, that is, surrogate representation, based on responsiveness to affected constituencies. But there are self-appointed representatives whose claims are, normatively speaking, not legitimate from the standpoint of affected constituencies, or so I shall argue. Consider the case of Jerry Falwell who self-appointed as the spokesperson for, and helped to create a constituency called, the Moral Majority. The group, founded in 1979, was comprised of mostly fundamentalist, evangelical Christians who, arguably, were not well represented by either of the two major political parties at that time. In creating and speaking for the Moral Majority, Falwell 8  helped to enable an organization that, in turn, authorized his representation, though outside of formal representative institutions. Because the Moral Majority was a membership-based organization, there were also procedures of non-electoral accountability, the most important of which was the potential for exit. However, as I shall argue, Falwell is neither a democratic nor a surrogate representative because he is not responsive to the interests of an affected constituency. The agenda of the Moral Majority included opposition to civil rights for gays and lesbians. With respect to this representative claim on this particular issue (and there are other claims and issues upon which he may be judged differently), Falwell represents one group (evangelical Christians) with the effect of undermining the self-determination and self-development of an affected constituency (gays and lesbians). According to the approach that I develop here, Falwell does represent the Moral Majority on this issue, but to the effect of amplifying the voices of those who would oppress and/or dominate others. Self-appointed representation in this kind of case produces what I call skewed representation. This category follows from relating the representative claim to constituents whose interests are affected with respect to their selfdevelopment and self-determination. That is, we identify affected constituencies as those whose interests (would) suffer from institutionalized oppression and/or domination. On this issue, it is LGBT peoples who are the affected constituency because it is their interests, defined as what affects or is important to the life prospects of individuals, which would be undermined by opposition to civil rights legislation. Civil rights legislation for LGBT peoples would not interfere with the self-development and self-determination of the Moral Majority. The Moral Majority would still be able to express their feelings and perspectives on social life in contexts whether others can listen (self-development), and they would still be able to pursue their lives in their own way (self-determination). According to this formulation, the Moral Majority is not an affected constituency – or, at least, their claims to be an affected constituency are relatively weak when compared to the impact of civil rights (or their lack) on the self-development and 9  self-determination of LGBT peoples. Falwell, then, with respect to this claim – a claim that, in part, constitutes LGBT peoples as a constituency – is a skewed, not democratic representative, because he is neither responsive to an affected constituency on this issue nor do his representative claims and functions promote their empowered inclusion. On this same model, we can see that Falwell promotes the empowered inclusion of the Moral Majority – though at the cost of creating an affected constituency that is, as a consequence of this representation, disempowered. Following this same logic, we will have to say that judging the extent to which a self-appointed representative is democratically legitimate will require assessments of consequences for self-development and self-determination of the constituencies’ interests that are brought into existence, as it were, as a consequence of the representative claim. In other cases, claims of representation simply fail. Many organizations that claim to represent “African Americans” and “women,” as examples, often do not well represent the diverse interests within their organizations. To take one well-known case, from the 1980s on, the National Organization for Women (NOW) was challenged by subsequent feminists – lesbians and women of colour in particular – as being “unrepresentative” of the women for whom it claimed to speak. NOW was, in fact, a largely white, middle class organization that did not well represent the diverse interests among women. Though feminist organizations like NOW claimed to represent all women, they worked mostly on issues that benefitted what Strolovitch calls “advantaged subgroups” – educated, middle class members, for example – rather than “disadvantaged subgroups” – say, low income women and low income women of colour (2007, 3). Such organizations then, are neither responsive to the interests of their disadvantaged constituents nor do they, as largely white, middle class organizations, promote their empowered inclusion, and so fail in their claimed representation of these groups. I shall refer to such cases as, simply, failed representation. This is, in broad outline, the analytic framework of the argument I make in this 10  dissertation. While I develop this argument in subsequent chapters, understanding its importance depends, in large part, on understanding the place and function of representation within democratic theory specifically, and democratic practice more generally. Why has the topic of representation reemerged in democratic theory? Why do I understand self-appointed representatives as assuming an important – perhaps increasingly important – place in the development of contemporary polities? In the remaining sections of this introduction I hope to provide some of this context, first, by outlining democracy as a mode of politics rooted in the norm of self-determination that is concerned that citizens (narrowly understood) exercise self-rule. Second, I discuss how representation grows out of problems of expanding democracy across larger populations and larger political units. Representation is necessary to making democracy work, though it transforms our understanding of democracy as self-determination from an ideal of self-rule to one of a self-legislating demos. That is, as selfdetermination is expanded to include larger numbers of people, the direct democracy of the ancient Greeks, which exemplifies democracy as self-rule, becomes impracticable, and so in the context of representation, democracy is rethought as self-legislation (Bohman 2007). Third, to inform my delineation of the importance of a non-electoral account of democratic representation, I provide a brief overview of Hanna Pitkin’s landmark account of representation – an account that provides important conceptual resources, but also defines a now standard account of political representation that collapses electoral representation and democratic representation. Fourth, I outline the functional gaps left by the standard account of representation, including its inadequacy for the purposes of conceptualizing and assessing nonelectoral representation, which limits its critical purchase as a theory appropriate for assessing current practices of representation. Fifth, I explore what self-determination can mean in the context of self-appointed representation, which operates across borders and across electoral constituencies, and is not bound to self-legislating demoi. I conclude with an overview of the 11  dissertation as a whole.  Democracy and its Norms: Political Equality and Self-Determination Democracy means “a form of government in which, in contradistinction to monarchies and aristocracies, the people rule,” and so importantly, “entails a political community in which there is some form of political equality among the people” (Held 1996, 1). Although we often assume that political equality is, primarily, about the normative status of people within a political community, perhaps the more fundamental moment occurs in the constitution of the political community itself, since it establishes the lines that demarcate those who are included from those who are excluded. This question – who is “the demos” – is fundamental to determining the scope of democracy, and hence the reach of norms of political equality. “Strange as it may seem,” Robert Dahl points out, “how to decide who legitimately make up ‘the people’ – or rather a people – and hence are entitled to govern themselves in their own association is a problem almost totally neglected by all the great political philosophers who write about democracy” (1970, 46). “The people” may not refer to everyone in a given society, but only to a subset of that society known as citizens. For example, the Athenians, at the zenith of their democracy, defined the demos as native-born male adults – a small minority that excluded large categories of people, including slaves, free females, and metics (foreigners living in the city). The lawgiver Solon created a framework in which native, adult males from all walks of life would be counted as citizens. This group was a minority, but one in which political power was widely distributed; that is, citizenship was not for the wealthy alone. Democracy meant rule by the people (adult, male Athenians), including the poor (who are not slaves). In a series of reforms he called isonomia (equality vis-à-vis the law), the Athenian statesman Cleisthenes changed the political organization of the four traditional tribes of Athens,  12  based on family relations, by dividing the citizens into ten tribes according to their area of residence and abolishing surnames, thereby dissolving the remnants of the tribal divisions that existed in Athens. Citizens came to be associated by the deme (regional district) from which they came, rather than by the traditional association to their family or clan. Citizens were chosen by lot (rotation) rather than by kinship or election to serve in political institutions, and the Athenian assembly was open to all citizens who would meet in the agora to participate in debates and decision-making. Cleisthenes also reorganized the boule, a council of 400 citizens whose role was to propose laws to the assembly of voters. Under Solon, the boule was aristocratic in character, excluding the poorest class, and with posts reserved for men from the top income groups. Cleisthenes reorganized the boule so that it had 500 members, 50 from each tribe. This council proposed laws to the assembly of voters, which could be rejected, passed, or returned for amendments by the assembly. Political equality in the polis was intended to operate only among an exclusive group of men, but a group that was not determined by their membership in a clan. This group of citizens made the laws under which they lived: an expression of self-determination as self-rule among equals in a bounded political community. As democratic forms of government expanded to include larger numbers of people, the direct democracy of the ancient Greeks became impracticable, and so required rethinking in the context of incorporating larger numbers of people across larger political units. As democracy underwent a significant transformation, the ideal of democracy as self-rule was no longer feasible. Given that direct democracy, as we have known it in practice, has been exclusionary and suitable to small units, and given that the theory of democracy as self-rule is suitable to small units, how can larger units be democratic? The answer was that they must be representative, which in turn required that the democratic ideal of self-determination be rethought within the context of larger political units.  13  Democracy Transformed: Representative Democracy, Self-Determination, and Affected Interests In the Western history of the practice and theory of democracy, representative democracy is sometimes described as its “first major transformation,” and as “a second form of democracy that enabled it to spread to larger political units and incorporate large numbers of people” (Warren 2003, 223). Robert Dahl writes that as the focus of democratic government shifted to large-scale units like nations or countries, the question arose: How can citizens participate effectively when the number of citizens becomes too numerous or too widely dispersed geographically (or both, as in the case of a country) for them to participate conveniently in making laws by assembling in one place? And how can they make sure that matters with which they are most concerned are adequately considered by officials – that is, how can citizens control the agenda of government decisions? (2000, 93) Dahl answers that the only feasible solution is for citizens to elect representatives and to hold them accountable through subsequent elections (2000, 93). Dahl’s comments repeat John Stuart Mill’s classic statement of the dilemma of democracy in modern mass society. In On Representative Government (1991 [1861]), Mill argued that since not everyone can personally participate, representative government is the ideal form of government. The rise of the nation-state meant that men could no longer meet in local assemblies to enact law. Representative democracy and its institutions enabled the electoral form of democracy to encompass nation-states with large territories and populations, as well as relatively complex governments. The new institutions that facilitated representation “included national parliamentary government, elected representatives, and popularly elected local governments operating within the encompassing elected national governments” (Warren 2003,  14  223). These new representative forms retained the notion of democracy as self-determination but now enacted through electoral institutions. “As a result, the new ideal equated democracy with self-legislation, as the act of the people giving itself laws through the elected legislature that expresses the popular will” (Bohman 2007, 21). The Athenian ideal of self-determination as self-rule thus gave way to an idea of self-determination as a self-legislating demos in which one is simultaneously the author and the subject of the laws, mediated by elected representatives and institutions. Universal suffrage expands the right to vote – interpreted by the model of representative democracy as self-determination – to ensure that (most) citizens literally live under the laws of their own choosing (Dahl 2000, 54). In this way, universal suffrage has come to be the defining feature of democracy, as well as the primary political mechanism of empowered inclusion (Dahl 2000, 3). As Dahl points out, “overwhelming historical evidence” suggests that, “the fundamental interests of adults who are denied opportunities to participate in governing will not be adequately protected and advanced by those who govern” (2000, 77). The right to vote must be extended to all the people who are subject to the law (full inclusion, with few exceptions around competence and transience) in order to ensure that all those who are subject to the law are also its authors. Although “many citizens and philosophers argued that this new institutional form violated the basic democratic principle of self-rule” (Bohman 2007, 20-21), we should note that even the electoral form of representation retains and builds upon a key principle of democracy: the underlying rationale of suffrage was that “[e]veryone who is affected by the decisions of a government should have the right to participate in that government” (Dahl 1970, 49). This principle, to which I refer as the affected interests principle, has a long pedigree. In medieval times, the Latin phrase was, “quod omnes tangit, ab omnibus approbetur,” or, “whatever touches all, must be approved by all.” Pitkin notes that in the thirteenth century, this doctrine “apparently fostered the requirement, in both church and secular government, that 15  extraordinary taxes must have the consent of those taxed” (1967, 85). Combined with an expanding electoral franchise, “representation came to mean popular representation, and to be linked with the idea of self-government, of every man’s right to have a say in what happens to him. And that is how it became embodied in our institutions” (Pitkin 1967, 3). The result is a standard account of political representation, in which elections are seen as the defining feature of democratic representation, justified by the principle of self-determination with respect to affected interests, although narrowly defined, based on residence. While the key point of this dissertation is to expand democratic representation beyond its electoral forms, it is important to note that, restricted though it is, the electoral form still retains these essential normative assumptions.  The Standard Contemporary Account of Democratic Representation In 1967, Hanna Pitkin published her now classic work in which she asks, when should people or individuals feel that they are represented? Her answer is: when there is substantive acting for them. Political representation means “acting in the interests of the represented, in a manner responsive to them” (Pitkin 1967, 209). Democratic representation, however, requires the institutionalization of such responsiveness so that there is not “merely occasional response when [the representative] pleases, but regular, systematic responsiveness” (Pitkin 1967, 234). The definition of representation as a substantive acting for others does not necessarily require elections, though Pitkin does not pursue this logic (Saward 2008, 4). Instead, she suggests that elections, and the authorization and accountability derived from them, ensure responsiveness. According to Pitkin, in contemporary representative democracies and in democratic theory, our preoccupation with elections and electoral machinery is the result of our belief that such systematic responsiveness is difficult, if not impossible, without elections (1967, 234).  16  Without electoral institutions, “the ideal of representation would remain an empty dream, or at most would occasionally recur as a fitful, inexplicable blessing, which we have no power to produce or to prolong” (Pitkin 1967, 239). The election of a representative amounts to an authorization of that representative by the constituency for the proposed policy platform. Accountability is ensured by the possibility of re-election or alternatively, the threat of removal from office; if a representative wishes to be re-elected he or she will presumably be responsive to the interests of the constituents (enact preferred policy) and will be required to give an account of his/her performance. This is a reasonable but somewhat surprising conclusion given Pitkin’s critique of theories of formal representation. These theories are, according to Pitkin, incapable of providing standards for assessing the performance of a representative. Formal theories of representation, in which the vote authorizes the beginning of a term and then provides the possibility of sanctioning or rewarding one’s representative at the end of that term, are an expression of what Nadia Urbinati calls a “yes-no politics.” These theories largely ignore what occurs during representation, and cannot therefore tell us “how a representative ought to act or what he is expected to do, how to tell whether he has represented well or badly” (Pitkin 1967, 58). And, as Urbinati explains it, “In a representative democracy the continuity through the electoral term is the norm we expect representatives to comply with so that we can recognize them, so to speak, and judge them always, not only at the end of their electoral mandate” (2006b, 19). Ultimately, Pitkin conflates what she sees as democratic representation – acting in the interests of the represented, in a manner responsive to them – with electoral democracy – the institutionalization of such responsiveness through elections. The result is what has been termed the “standard account of political representation” (Rehfeld 2006; Urbinati and Warren 2008), in which elections are seen as the defining institutional feature: “elections, open, free, and fair, are the essence of democracy, the inescapable sine qua non” (Huntington 1991, 9). This account is 17  reflected in the work of theorists and political scientists who focus on the formal procedures of authorization and accountability and the proper design of electoral institutions (Amy 1996; Barber 2001; Christiano 1996; Guinier 1994). This is important work: the design of electoral systems may include or exclude, to a greater or lesser degree, those affected by collective decisions (Dovi 2002; Lijphart 1990; Mansbridge 1999; Phillips 1995, 1998; Williams 1998; Young 2000). That is, the design of electoral institutions can serve the norm of selfdetermination by widening the scope of political inclusion. However, this literature, and the standard account of representative democracy on which it turns, narrows the question of inclusion in the demos almost exclusively to a question of the relationship between citizens and their state, almost always interpreted as a relationship between citizens and their elected representatives. Archon Fung explains, Theorists have sought to provide an account of the political rights of individuals living under the legitimate authority of a democratic state. These accounts seek to ground the judgment that adults living under such authority should be treated as equal citizens who have, among other political liberties, rights to participate in the decisions of that state. (2009, 1) The authority of a democratic state is rendered legitimate by including in the demos those who reside within the boundaries of that state, and by providing those individuals with equal political status, expressed in the equal right to vote. It is a key feature of this version of the democratic state that inclusion is generally based upon a principle of territoriality, or residence (Rehfeld 2006). As Sarah Song argues, The pervasiveness of the state-centered view of democracy is a testament to the power of the idea of Westphalian sovereignty in modern political and legal thought. It is based on the principle of territoriality and the exclusion of external actors from domestic authority structures. The territorial boundaries of nation-states have legal and political 18  significance: they demarcate a political community of citizens whose will and interests are taken to form the standards of political legitimacy, and they also define which people and institutions are subject to the laws of the state. (2010, 1) Put another way, the standard contemporary account of political representation restricts representation (and self-determination) to individuals within a given nation-state and to the interests that these individuals have on the basis of their residence.  What the Standard Account Does Not Cover The key limitation of the standard account is that it ties the democratic legitimacy of representation to territorially based electoral democracy rather than to the basic normative question as to who is affected by collective decisions. In part, the problems have to do with complexity. The ideal of self-determination as a self-legislating demos is complicated by the fact that people are increasingly affected by laws to which they are subject but without being their authors. Because of the complexity of political issues that affect peoples across borders and citizenships, such as global warming, disease, and trade, whatever self-determination could entail in a globalizing context, it must reflect more than an idea of people giving itself laws through an elected legislature. In addition to the complexity of issues, the problems are those of borders and boundaries. Even with the universal franchise, elected representation is unresponsive to large classes of the potentially affected in two ways. First, the affected are not always included in the electorate. Because the affected are not necessarily a (geographicallybound) electoral constituency, or a people bound by shared citizenship, the assumption that electoral membership ensures inclusion of the relevant constituency may result in the exclusion of those who are affected – sometimes the most affected – by collective decisions. Global warming and the spread of AIDS, as examples, are issues of global concern that require  19  solutions beyond a particular electoral constituency. Second, even when the affected are included in the electorate, they may still go underrepresented because of the aggregating character of electoral representation. When elected representatives combine many different issues to present a platform, that process of aggregation sometimes “includes” while failing to represent many interests with legitimate claims to representation, that is, interests that are affected by relevant issues. In the United States, for example, the Democratic Party has long been accused of “taking Blacks for granted.” Black votes have often failed to produce representation of relevant issues because they are aggregated with so many others in ways that enable electoral majorities. As Charles Lindblom points out, One must cast a vote for a candidate or party that takes a position on each of many issues, only some of which the voter tends to endorse. The winning party does not know whether it won a voter’s ballot because the voter wants a tax program offered by the party or candidate or because the voter wants to endorse the candidate or party on some other issue. (2002, 72) The effect of such aggregation is that a citizen only has a voice if she or he happens to share the interests of the winning platform (Warren 2008, 58).1 From a normative perspective, representation should reflect more than the representation of interests as defined by residence in a given territory. The mere mention of issues such as global warming, disease, and trade indicates that not all of our interests are easily captured by our residence within the borders of electoral districts within nation-states, or even by the borders of a particular nation-state. Regional and transnational institutions are developed to grapple with these issues and actors who self-appoint step in to fill the gaps in representation that occur when those affected by laws have not given themselves those laws. However, because it collapses 1  Certainly, where they exist, PR and preferential systems mitigate this aggregating effect.  20  democratic representation with elections, the standard account leaves us without the ability to conceptualize or judge the democratic potentials of those outside of electoral processes who situate themselves as representatives. For all of these reasons, we require a general theory of democratic representation in which the democratic functions of representation are not only identified with electoral institutions as such. The standard account does little to identify, much less assess, the many new kinds of representative claims made by post-national and transnational actors, including the United Nations, the European Union, and treaty organizations, as well as by self-appointed representatives such as NGOs, foundations, and civil society organizations in both formal and non-formal settings. Not surprisingly, the empirical inadequacies suggest a key conceptual inadequacy in the standard account, one identified quite precisely by Andrew Rehfeld: “By wedding representation with the conditions that render it legitimate, the standard account is doing double duty: not only does it tell us when a representative is legitimate or democratic, it also purportedly tells us when a person is a political representative at all” (2006, 3). The standard account, in other words, conflates the conceptual identification of the elected representative with his/her democratic legitimacy. In what sense can we say that non-elected actors, and specifically actors who selfappoint, count as democratic representatives? I mentioned above that at least some selfappointed representatives seem to entail, prima facie, some democratic credentials. These credentials are rooted in a basic normative intuition that is at the heart of most contemporary democratic theories: the affected interests principle. This principle suggests that those potentially affected by a collective decision should have some say in influencing that decision (Cohen 1971; Dahl 1970; Goodin 2007; Habermas 1996; Held 1996; Whelan 1983; Young 2001). Democratic theorists have often suggested that the basic concept of democracy is that 21  those affected by collective decisions should have some influence in making those decisions. The question as to how influence is actualized – what rights are necessary, for example – is highly contested. Robert Goodin (2007), among others, has argued that the affected interests principle is wildly expansionary. Thus, for example, The presidential elections in the United States will heavily affect Iran and Afghanistan – perhaps even more than it will affect a citizen living in Mississippi. So the affectedinterests principle would require that an Afghani have a vote in that election that is perhaps more heavily weighted than that of the American citizen in Mississippi. When the affected-interests principle is invoked within the context of nation-states, the problem becomes one of extending the vote across borders. The norm suggests that there is something intuitively correct about that implication, but we certainly have not figured out the mechanics of how to manage practically such situations. (Mansbridge 2008, personal correspondence) The expansionary character of the principle suggests to Goodin and others that the principle serves best as a norm that underpins judgments about inclusion, rather than a mechanism of inclusion itself.  Self-Appointed Representation and Self-Determination One of the key challenges I take up in this dissertation is to understand representation in a way that helps to actualize the affected interests principle. One of the important features of self-appointed representatives, I shall argue, is that we do not have to figure out the mechanics of distributing votes amongst those affected because the context of the self-appointed representative is primarily discursive and non-formal, and so not bound up with institutions and mechanisms such as electoral districts. This means that these kinds of representatives can, in  22  principle, be much more responsive to effects of collective decisions and actions that do not correspond to electoral constituencies. In contrast, electoral institutions narrow our understanding of representation in ways that hinder our potential to conceptualize representation separate from elections and on behalf of non-electoral constituencies. As a key step in conceptualizing representation apart from electoral institutions and in ways that can include constituencies across boundaries, I rely on Michael Saward’s (2006, 2008) focus on representation as a process of claim making. Representation, Saward argues, should be viewed “in terms of claims to be representative by a variety of political actors, rather than (as is normally the case) seeing it as an achieved, or potentially achievable, state of affairs as a result of election” (2006, 5). In conceptualizing representation as claim making, we move from considering representation as “a factual product of elections” to considering representation as “a precarious and curious sort of claim about a dynamic relationship” (Saward 2006, 5). Rather than focus our attention on forms of representation (delegate and trustee, as examples), as the literature on representation has done, the concept of the representative claim shifts our attention to representation’s dynamics, that is, to what exactly is going on in a relationship of representation (Saward 2006, 5). I shall argue that the dynamics of representative claim making – indeed, one of its most important elements – involve the constitution of peoples as constituencies. It is, in part, this constitutive element that makes it so important to conceptualize the democratic possibilities of self-appointed representatives. In principle, self-appointed representation is not restricted to the boundaries of electoral institutions and so may provide representation of peoples whose interests are not territorially bound, and/or who are not situated within electoral constituencies. Not only does self-appointed representation supplement democratic theory and practice by providing representation for non-electoral constituencies within and across borders, but it also recasts some core premises of democratic theory and practice precisely because it functions as a 23  mechanism of constituency formation, helping to constitute non-electoral constituencies and sometimes, democratic ones – a problem often considered irresolvable within democratic theory. Many theorists argue that the demos must already exist in order for it to decide matters democratically (Whelan 1983; Goodin 2007). The implication is that the demos itself is not democratically constituted. In Chapter Four I shall suggest that the question of constituency has been undertheorized in recent democratic theory. But it was addressed, if indirectly, by earlier theorists such as Machiavelli (1994 [1532]) and Rousseau (1978 [1762]). Machiavelli and Rousseau prescribe a founder and a lawgiver, respectively, to, in fact, constitute the republic, figures that I suggest have interesting parallels with the self-appointed representative. Machiavelli provides us with a simple formula: good laws equal a good state. However, in order for “there to be a people well-formed enough for good law-making, there must be good law for how else will the people be well-formed? The problem is: where would that good law come from absent an already wellformed, virtuous people?” (Honig 2007, 3). Rousseau’s solution to this paradox of democracy is to prescribe a guide who will, at the founding, set the body politic on the right path. This guide is the legislator, whom Rousseau describes as “the engineer who invents the machine” (1978 [1762], 163): this machine, the lawgiver, makes citizens, moulds them in just such a way that they will constitute a healthy republic. He who dares to undertake the establishment of a people should feel that he is, so to speak, in a position to change human nature, to transform each individual (who by himself is a perfect and solitary whole), into part of a greater whole from which this individual receives, in a sense, his life and his being; to alter a man’s constitution in order to strengthen it; to substitute a partial and moral existence for the physical and independent existence we have all received from nature. In a word, he must deny man 24  his own forces in order to give him forces that are alien to him and that he cannot make use of without the help of others. (Rousseau 1978 [1762], 163). Honig (2007) points out, however, that there are two problems with the lawgiver: one, the lawgiver compromises the people’s autonomy, “which Rousseau also seeks to secure as a condition of their ability freely to will the general will” (Honig 2007, 3). Two, the lawgiver only addresses the problem of the original founding. However, the problem of founding, as Honig argues, occurs daily. Every day, after all, new citizens are born, and still others immigrate into established regimes. Every day, already socialized citizens mistake, depart from, or simply differ about the commitments of democratic citizenship. Every day, democracies re-socialize, recapture, or reinterpellate citizens into their political institutions and culture in ways those citizens do not freely will, nor could they. (Honig 2007, 3) There is a paradox in Rousseau’s thought: to be free, the republic must be guided. To be free, they need the lawgiver’s authority. This is not the problem it seems to be, however. Honig argues that the lawgiver “may offer to found a people, he may even attempt to shape them, but in the end it is up to the people themselves to accept or reject his advances” (2007, 6). The constitutive moment proposes to people that they have an issue to which they should attend; that they should become conscious of the issue, and consider how, and to what extent, they share the issue with others – their (proposed) fellow constituents. It is precisely because the claim is made that individuals may be induced to consider their interests, and to say “yes” or “no” to the formulations offered by the representative. That is, the constitutive element of representative claim making may be essential to producing autonomous peoples. For my purposes, the self-appointed representative offers a claim in an attempt to constitute a constituency in relation to a particular issue. The constituency is only constituted, 25  however, if they agree to, indeed, if they authorize, the claim as when, for example, African Americans agreed to participate in the bus boycott organized by leaders of the civil rights movement, including Dr. King, E.D. Nixon, and Ralph Abernathy. Moreover, in cases of selfappointed representatives, these moments of authorization do not only occur once, at the founding, but are continuous. In the case of the Montgomery bus boycott, we might say that authorization occurred every day for as long as the boycott lasted – over 380 days. Put another way, the authorization of self-appointed representatives is not only originary but, as we shall see in Chapter Five, serial and ongoing. And, partly as a consequence of this reflexive process, autonomous peoples come into existence as new kinds of constituencies. Following this same logic, we might also note that Dr. King helped to constitute a democratic non-electoral constituency on the basis of the affected interests principle. Dr. King offered a claim on behalf of African Americans – a constituency of peoples who did not neatly fit within electoral boundaries – whose interests were being undermined by racial prejudice, identifying them as an affected constituency. That constituency supported his claims, in effect authorizing his status as their representative, as I shall argue in Chapter Five. The support of his claims not only authorizes his status as a representative, it also forms a constituency of people that had not existed in quite the same way prior to his claim. When African Americans authorized Dr. King’s claims, they constituted themselves as a people who demanded civil rights legislation and who would do so through massive non-cooperation. In short, through the authorization of representative claims, the people can democratically constitute themselves as a demos. Put another way, understanding self-appointed representatives is key to understanding the means through which people become self-determining through constituting themselves as a demos in relation to an issue that affects them. In the context of democracy, self-determination requires a reflexive form of self-constitution in which, as Bohman puts it, “the terms and boundaries of democracy are made by citizens themselves and not others” (2007, 2). Non26  electoral constituencies cannot always give themselves laws through an elected legislature; they can, however, define the boundaries of their own demos or demoi. I am arguing that this kind of democratic autonomy is a possible, not necessary, consequence of self-appointed representation. As suggested above, Bono, for example, offers a claim to represent Africans suffering from AIDS, debt, and trade, and so helps to constitute that affected constituency. But it is not they who authorize his claims. Instead, it is those relatively privileged individuals in the developed world who join his organization and authorize his claims. This group of people is not the affected constituency but instead a constituency concerned for the affected and acting on their behalf: a surrogate constituency – though a constituency that was certainly brought into being in part by Bono’s representative claims.2 When the Moral Majority coalesced around Jerry Falwell’s claims, they too constituted themselves as a people by authorizing his claims. They are not an affected constituency, but what I term a skewed constituency, because their interests (the means by which a person or group achieves the ends she sets for herself/themselves, involving conditions of self2  It is for this reason that, in contrast to Lisa Disch (2009, 2010), I retain the usage of  “responsiveness.” Disch argues that responsiveness assumes pre-existing constituencies and also treats the representer (the agent) as the sole bearer of agency. For these reasons, she argues, responsiveness is incompatible with a mobilization theory of representation. Disch argues instead for a norm of “reflexivity” which better expresses the relationship of representation as one in which both representer and represented have agency, and which pertains to the quality of public judgment. In the context of surrogate representation, however, I do not think affected constituencies participate in their own mobilization, as much as they are constituted by it, and they are not always privy to (the quality of) public judgment. Though I largely agree with Disch’s assessments, for now, I find that responsiveness remains useful, as I have defined it. 27  development and self-determination) would not be (as) undermined by the existence of civil rights legislation for LGBT peoples. And failed constituencies can be identified when the constituency fails to recognize shared interests with respect to a given issue (even though their conditions of self-development and self-determination would be undermined), and a chance to have them represented. Our collective failure to take up claims with respect to global warming, for example, identifies us as a failed constituency. In these ways, then, I shall begin to develop new understandings of authorization, accountability, responsiveness, and constituency, separate from their traditional meanings in electoral contexts, the topic of Chapter Three. I have suggested that there is no demos without representation, and that self-determination takes on a new meaning as self-constitution, the subject of Chapter Four. I have argued that responsiveness occurs when a self-appointed representative constitutes a constituency on the basis of the affected interests principle, improving conditions of self-development and self-determination. And I have hinted at the nonelectoral mechanisms of authorization and accountability that will be the subject of Chapter Five – mechanisms such as boycotts and memberships – that are ongoing, rather than solely foundational. As we shall see, these mechanisms will help us to clarify the work of selfappointed representatives, and provide criteria for assessing their legitimacy.  Chapter Outline In the following chapters, I make the case for understanding self-appointed representation as part of democratic systems of representation. Chapter Two examines some of the recent literature on representation. First, I discuss the meaning of representation, particularly as developed in Hanna Pitkin’s work, The Concept of Representation. Second, I touch upon the literature that argues for the importance of descriptive and group representation for the purposes  28  of broadening inclusions within electoral democracy. Third, I discuss the emerging literature that has begun to talk about representation in alternative terms. I suggest both where this recent literature will be useful and also how I will move forward to offer a general theory of democratic representation, which is required to explain the phenomenon of self-appointed representation. In Chapter Three, I situate the concept of self-appointed representation in relation to both formal and informal representation, explaining its role in the broader system of representative institutions. This chapter compares the role of the representative claim, which includes three elements I call existential, constitutive, and performative, in formal, informal, and selfappointed contexts. In the formal context, one can make existential claims – “I am a representative of this group” – without an emphasis on the performative. That is, once an actor is elected, s/he is a democratic representative regardless of performance while in office. In the context of the self-appointed representative, the performative aspect of the claim is more directly linked to authorization and thereby to the question of how a representative’s democratic credentials can be earned outside of formal electoral systems. Furthermore, the claim includes a constitutive aspect – the makeup of “this group” – that is assumed in electoral politics, or that takes an electorate as given. In contrast, the self-appointed representative does not simply bundle together the preformed interests of the people; instead, she plays a role in will- and constituency-formation. In Chapter Four, I discuss the potential of the self-appointed representative to form constituencies on the basis of the affected interests principle. Whereas democratic theory traditionally holds that the formation of the demos cannot be democratic, the general theory of democratic representation that I develop flips that assumption on its head. I suggest that through the process of (de)authorizing self-appointed claims of representation, it is theoretically possible for everyone to decide whether or not they want to be part of this and/or that people. When 29  individuals authorize or reject claims of representation, they choose the boundaries of their own demos. The self-appointed representative thus enables an understanding of democracy as an ideal of self-determination in which citizens choose the boundaries of their own demos or demoi. It is for this reason that the traditional paradox does not apply. The self-appointed representative enables flexible, highly sensitive constituency formation, based on issues of interest. Such issue-based constituency formation produces at least three types of constituencies: skewed, surrogate, and affected. These sources of authorization reflect the difference between authorization and legitimacy. I argue that the authority to act as a representative may come from the relevant audience, and perhaps a (skewed or surrogate) constituency, but that what I term constitutive legitimacy comes from her or his responsiveness to the affected constituency. In order to be considered a democratic self-appointed representative the self-appointed representative must both be responsive to and promote the empowered inclusion (authorization and accountability links) of the affected constituency. In Chapter Five, I answer the Economist’s question, “Who elected Oxfam?” The answer is two-fold. First, while it is true that no one elects Oxfam, I suggest non-electoral mechanisms of authorization and accountability that may connect the self-appointed representative and the affected constituency in a democratic relationship of representation. Furthermore, I shall argue that such mechanisms are appropriate with respect to the powers and functions of self-appointed representatives, as discursive and voice representatives. Second, this chapter explores the ‘who’ in “Who elected Oxfam?” As discovered in chapter four, the authority to act as a representative may come from a relevant audience, as well as a skewed or surrogate constituency, but the constitutive legitimacy of the representation comes from its responsiveness to the affected constituency, and the democratic legitimacy of the representation comes from responsiveness to, and empowered inclusion of, the affected constituency. Determining the relationships established by the claim and the mechanisms of authorization and accountability at work 30  provides an answer to this question. In Chapter Six, I develop a normative typology of kinds of self-appointed representatives to answer the question: How would we judge a self-appointed representative to be democratic? I illustrate the critical capacities of the approach with several examples. The typology is based on the following criteria, as developed in the previous chapters. First, the affected constituency must be constituted around the claim of representation, as developed in Chapters Three and Four. Second, the affected constituency must authorize and hold accountable the self-appointed representative via one or more of the mechanisms listed in Chapter Five. On the basis of these distinctions, I develop and illustrate four kinds of self-appointed representation: democratic, surrogate, skewed, and failed. Democratic representation occurs when the self-appointed representative is both responsive to and promotes the empowered inclusion of the affected constituency through identifiable mechanisms of authorization and accountability. Surrogate representation occurs when the self-appointed representative acts on only the first of these dimensions – that is, the self-appointed representative is responsive to, but does not promote the empowered inclusion of, an affected constituency. Skewed representation occurs when the selfappointed representative is unresponsive to the affected constituency and promotes the inclusion of those who would undermine the self-development and/or self-determination of an affected constituency. Finally, if the actions of a representative fail on both dimensions of responsiveness and empowered inclusion, that is, the self-appointed representative is both unresponsive to and exclusive of the affected constituency, despite claiming to be their representative, the representative claims simply fail. The dissertation concludes in Chapter Seven that self-appointed representatives have the potential to address a condition in which the institutions of representative democracy are underdeveloped in most of the world and increasingly inadequate in the remainder. It may be unfamiliar as it differs from our traditional understanding of democratic representation, which is 31  based on formal authorization and accountability. However, the potential of self-appointed representation to be responsive to and promote the empowered inclusion of affected constituencies serves an important function for the many groups, interests, and discourses that are unrepresented in formal political institutions. Self-appointed representation may be essential to the deepening and development of democracy in a complex, post-national world.  32  Chapter Two: Democracy and Representation As argued in Chapter One, there are an increasing number of real-world cases of actors who claim to represent by self-appointment, and who seem to do important political work. We need a theory of representation that identifies and normatively assesses these cases of selfappointed representatives that seem, prima facie, of central importance to democracy. In this chapter, I work toward such a theory by considering the conceptual resources embedded in standing theories of political representation. I survey key elements of such theories to both identify conceptual resources that might be extended beyond accounts focused on standard electoral representation, and also to identify their limitations that will have to be reconceived. First, I discuss the meanings of political representation in Hobbes and Burke, guided by Hanna Pitkin’s well-known analysis. Second, because the territorial notion of “constituency” is both central to, and also problematic for, the concept of political representation, I survey the small literature that addresses the concept of constituency. Third, I examine the conceptual relationships in the literature between political representation and electoral authorization and accountability, focusing on the ways in which the concept of political representation has been made continuous with – perhaps even collapsed into – democratic electoral representation. Fourth, I discuss the small but growing literature that has started to conceive of representation in alternative terms, focusing particularly on the constitutive and reciprocal qualities of representation. Fifth, I comment on the civil society literature that seeks to frame the problem of the legitimacy of civil society actors. Finally, I discuss the framework of the representative claim, which reconceptualizes representation and democratic representation separate from formal representative institutions such as elections. In subsequent chapters I build upon this nonelectoral theory of (democratic) representation to conceptualize non-electoral actors, particularly those who self-appoint, as “representatives,” which will then enable critical analysis and  33  assessment.  Hobbes: Representation as Authorization Until the appearance of Hanna Pitkin’s now classic work, The Concept of Representation, published in 1967, representation was arguably a neglected concept in democratic theory. Despite the importance of the concept, Pitkin points out that few political theorists had actually explained what representation means. In fact, “Hobbes is the only major political theorist who gives a fully developed, systematic account of its meaning…Even John Stuart Mill, who devotes an entire book to representative government, does not consider it necessary to explain what representation is or means” (Pitkin 1967, 4). Pitkin undertakes the project of understanding what representation is, or is like, by examining the meanings it has assumed (1967, 2). Pitkin asks, when should people or individuals feel that they are represented? What reasons can be given for supposing someone or something is being represented? (Pitkin 1967, 9-10). Her method involves tracing the historical uses of the concept of political representation, particularly those concepts that are precursors to contemporary electoral theories of representation, beginning with the articulations of Thomas Hobbes. Key in Hobbes’s account of representation is the act of authorization. To get men out of the state of nature, “Hobbes uses not only the device of the social contract but also (as is less frequently recognized) his concept of representation” (Pitkin 1967, 29). In the state of nature, individuals are equal – an egalitarianism that produces war because they often desire the same scarce resources (Hobbes 2002 [1651], 94-5). Under such circumstances, the life of man in the state of nature is “solitary, poor, nasty, brutish, and short” (Hobbes 2002 [1651], 96). These circumstances motivate rational individuals to exit the state of nature and the risk of sudden,  34  violent death that it poses. Warring individuals can cease to be warring individuals by entering into promises with one another. Out of fear of each other, or out of fear of the victor in war, rational men covenant with each other, transferring that fear to the artificial person – the Leviathan – whom they institute as sovereign (Hobbes 2002 [1651], 149). Fear thus triggers the beginning of civilization. Fear also triggers political representation: the artificial person produced by the covenant is a representation of those who enter into the covenant. The representative qualities of the Leviathan are evident in its ontology: it is an artificial person because it is not the original author of its actions. When a person’s words or actions are considered as his own, “then he is called a natural person; and when they are considered as representing the words and actions of another, then is he a feigned or artificial person” (Hobbes 2002 [1651], 120). The sovereign, then, is an artificial person, one whose words and actions are given authority by the individuals he represents. Individuals covenant with each other to give up the rights they have in the state of nature (with the exception of the right to preserve their lives), and authorize the sovereign to do what is necessary to ensure their protection. In this way, the activity of the representative is authorized – an authorization gained through the covenant, from each and every subject. The sovereign has authority, what Hobbes calls “the right of doing any action,” only “so far forth as is in their commission, but no further” (Hobbes 2002 [1651], 120-21). That is, the sovereign has authority to do anything necessary to ensure the protection of the commonwealth – but cannot take the life of the individual author. When people covenant each with each, that covenant turns the many into one. Hobbes suggests that “this is more than consent or concord; it is a real unity of them all in one and the same person” (2002 [1651], 128). Hobbes explains, A multitude of men are made one person when they are by one man, or one person, represented, so that it be done with the consent of every one of that multitude in 35  particular. For it is the unity of the representer, not the unity of the represented, that maketh the person one. And it is the representer that beareth the person, and but one person; and unity cannot otherwise be understood in multitude. (2002 [1651], 122-23) From the perspective of representation, the image of the Leviathan reverses the idea that the body politic creates the state. It is true that individuals authorize the Leviathan as sovereign, and that without this act of authorization, there is no state. But this act of authorization is performed by individuals, as individuals: they are not (yet) a body politic. Once individuals authorize the sovereign, a fictional entity is created (the state), in which individuals are considered as one. It is only through having a sovereign (head) that the body of the people is constituted; otherwise, there is only the mass. Thus, the state is something other or more than the body of the people. In fact, Hobbes’ notion of the state (as more than the name of the body of the people) does have something in common with later ideas of representative government: if the state were the same as the body of the people, the state would be a form of direct democracy. Instead, we run representative democracies in which the state is a kind of fiction, as Hobbes suggests (Skinner 1999). Hobbes’s covenant is a popular one: all of the people in the commonwealth have agreed. Despite the popular roots of the authority, he emphasizes that individuals, in giving the sovereign the authority to make decisions, will not challenge those decisions. As Pitkin states, “we read the Leviathan and feel that somehow we have been tricked” (1967, 34). In making this covenant, all rational individuals confer “all their power and strength upon one man or upon one assembly of men, that may reduce all their wills by plurality of voices unto one will” (Hobbes 2002 [1651], 128). One gives up the right to practice discretion – one’s will, as well as one’s judgment – in relation to the sovereign. Again, the sovereign is the artificial person who governs his subjects. And the sovereign can be a single human being (monarchy), or a group of people (aristocracy), and even an entire citizenry (democracy). Yet, it is an authoritarian state. Hobbes 36  is imagining an authoritarian monarchy, an authoritarian aristocracy, or an authoritarian democracy, in which portions of one’s will, including judgment, are transferred to the sovereign, whose will then represents the will of all. If there is to be peace, some person’s or persons’ will must be the will of everyone. Hobbes justifies the abandonment of individual will by emphasizing the stark choice between subjection and death. “For by this authority, given him by every particular man in the commonwealth, he hath the use of so much power and strength conferred on him that, by terror thereof, he is enabled to conform the wills of them all to peace at home and mutual aid against their enemies abroad” (Hobbes 2002 [1651], 129). Uncontested political authority is ensured by this awesome act of authorization, conducted out of fear of one another and/or of the sovereign. Out of fear, they give the sovereign “all their power and strength,” and thereafter cannot challenge the decisions of the sovereign. They are terrified into conformity. This unquestioning acceptance of the decisions of the sovereign, once he has been authorized, would not seem to fit with Pitkin’s view that, “representation implies standards for, or limits on, the conduct of the representative” (1967, 33). Strictly speaking, however, Hobbes’s treatment of the concept of representation does satisfy this definition: there is a limit on the conduct of the representative, and it is that which maintains or preserves life. But outside of this one exception, there is no conduct of the sovereign that could be considered as not representing the wills of constituents. Their wills, their capacities for judgment, are handed over to the sovereign, carte blanche, and they are thus unable to judge whether they are being represented well or badly. The sovereign is authorized and so represents. It is true that in addition to ensuring civil peace and stability, the Leviathan is meant to ensure what Hobbes refers to as “commodious living,” by which he means those things that make human life worth living: human cooperation that can lead to achievement in the arts, commerce, and sciences (Hobbes 2002 [1651], 95-6). People need stability in their world to achieve these things. If they live in a 37  world of all against all – if they live in a natural state, in the state of nature – commodious living cannot be achieved. In choosing a sovereign, they choose a society that can achieve more than they would otherwise be able. Though Hobbes seems to consider the duties of the representative (the sovereign’s ability to ensure peace and secure commodious living), his definition of representation as authorization, as “the one-sided assumption of obligations by the man who is represented” ensures that subjects cannot make any claims on the sovereign, on the representative (Pitkin 1967, 33). The relationship between constituent and representative thus collapsed – the lack of critical distance – makes us feel that “representation has somehow disappeared while our backs were turned” (Pitkin 1967, 37). According to Pitkin, Hobbes “developed too narrow a perspective on representation by approaching it from only one angle, by taking into account only one kind of representing” (1967, 37): representation as any activity after the act of authorization. Although Hobbes has this notion of commodious living that seems to require certain behaviour on the part of the sovereign, ultimately he emphasizes the act of authorization to the extent that any activity of the sovereign is binding on those he represents except harming their lives. Pitkin criticizes Hobbes’ theory of representation as authorization because it entails no assessment of the quality of representation; if one is authorized, that is a sufficient requirement for representation. Hobbes does not consider authorization as a reciprocal process that is ongoing and so might properly reflect the activity of the representative as that activity occurs. Authorization is originary, but not ongoing in Hobbes. Pitkin insists, quite rightly, that whatever else representation involves, it must involve more than a founding act of authorization. And this “more” must avoid collapsing the wills of representative and represented, such that the represented can judge the representativeness of the representative.  38  Burke and Virtual Representation To consider the activity of representation – representation as something more than original authorization – Pitkin discusses representation as “acting for,” a problematic that includes the theory of representation as trusteeship. The notion of representative government as trusteeship holds a position somewhat opposite to the Hobbesian notion of representation: calling the government a trustee implies “that the government must then act for the benefit of the people” (Pitkin 1967, 130). Hobbes sees representation as any activity that occurs after the act of authorization; Burke sees representation as trusteeship, which, Pitkin suggests, “can underline [the representative’s] obligations to others” (1967, 128). Put another way, Hobbes “focuses on the formal authorization of the representative, limiting the formal accountability and substantive obligations of both representatives and constituents. The Burkean or trustee view of representation, in contrast, emphasizes the substantive virtue and expert knowledge of the representative” (Brown 2006, 206). This relationship of trusteeship suggests that constituents have not transferred their will to the representative, though they have transferred their judgment, which the representative holds “in trust.” Theoretically, representatives are obligated, as trustees, to attend to the interests of their constituents, both as individuals and – importantly for Burke – as members of a political community. Yet Burke, like Hobbes, does not emphasize the responsibility or the obligation of the representative to consult his constituents. Trusteeship implies an obligation to the represented, “but it is equally implied that this does not require consultation or responsiveness to their wishes” (Pitkin 1967, 130). Pitkin explains that Burke’s notion of trusteeship “is related to his antidemocratic, elitist hostility to unnecessary extensions of the franchise” (1967, 189). For Burke, then, the representative has no obligation to consult his constituents, except in the very restricted sense that the legislature needs an accurate reflection of popular  39  “feelings.” Even this is not so much a part as a prerequisite of representation, and can be accomplished by any number of agencies. (Pitkin 1967, 189) Representation as trusteeship, at least in Burke’s formulation, does not leave room, it seems, for democracy or democratization, and so has been largely rejected (Pitkin 1967, Mansbridge 2004, Manin 1997). That is, trusteeship implies no accountability to the people and, as Manin argues, “It is the rendering of accounts that has constituted from the beginning the democratic component of representation” (1997, 233). Burke conceptualizes trusteeship as “an elite caring for others,” and as “an aristocracy of virtue and wisdom governing for the good of the entire nation” (Pitkin 1967, 172), without conceiving of representation as involving any consultation with those on whose behalf the trustee acts. From the perspective of democratic theory, however, Burke’s conception of virtual representation speaks directly to the work of this dissertation: I wish to retain the notion that a representative may provide representation to a constituency that is neither defined nor empowered as an electoral constituency. In Burke’s own words, “Virtual representation is that in which there is a communion of interests and a sympathy in feelings and desires between those who act in the name of any description of people and the people in whose name they act, though the trustees are not actually chosen by them” (1792). At minimum, virtual representation describes a relationship between a constituency of people with shared interests and an unelected representative who serves these interests. “Such a representation,” Burke continues, I think to be in many cases even better than the actual. It possesses most of its advantages, and is free from many of its inconveniences; it corrects the irregularities in the literal representation, when the shifting current of human affairs or the acting of public interests in different ways carry it obliquely from its first line of direction. The people may err in their choice; but common interest and common sentiment are rarely mistaken. (1792) 40  We see here the elitism that Pitkin and others have questioned: the people may err in their choice but the trustee can use judgment and wisdom to pursue the common interest, which is “rarely mistaken.” Yet in pursuing this “common interest,” the virtual representative is defining a constituency that would not, or could not, be formed on the basis of electoral mechanisms alone. While it is true that such representation can be unaccountable, and perhaps dangerously so, Burke has also identified a constitutive element in representation. By forming a representation of “common sentiment,” the virtual representative gives voice to latent publics or constituencies that may be excluded from the organization of electoral representation. Yet it is also true, as Pitkin rightly makes clear, that Burke’s formulation of representation carries dangers similar to those of Hobbes with respect to accountability. He offers a vision of a people with a “common interest and a common sentiment,” a judgment that conjures something that, potentially, exists independently of the people’s express opinions. But it is potentially productive in an important respect. Whereas Hobbes emphasizes that the sovereign represents one, united will (because individuals can agree only on the desire to avoid brutal death), Burke argues that it is interests, not wills, that are to be represented. These interests are to be represented in the form of judgments – “wisdom.” Burke said to his own constituency in Bristol, “If government were a matter of will upon any side, yours without question ought to be superior. But government and legislation are matters of reason and judgment” (Burke 1949 [1774], 115, cited in Pitkin 1967, 170). Because politics is concerned with what is right and wrong, Burke argues, wisdom is not to be found in the will. Representatives should not therefore represent the will, but rather interests expressed as judgments. So the activity of the representation is not about representing a person, which Burke thinks would require representing opinions. Instead, it is to represent interests. But, importantly, Burke also conceives of interests as existing independently of the opinions of individuals and groups. They are, as Pitkin notes, viewed by Burke “as something objective and discoverable” 41  (1967, 187). Of course, if they are objective and discoverable, then there is no need to consult one’s constituency; there is no need to render an account. The judgments of the representatives are not constituted in common with constituents; they are not the result of consultation or deliberation, which is why Burke does not see representatives as accountable to constituents, in the sense that he owes his constituents an account. Rather, the representative’s job is to discover interests, and the representative integrity of the representative’s judgment resides in its representation of these objective entities. Yet even virtual representation is not entirely disconnected from accountability. Burke warns that, “this sort of virtual representation cannot have a long or sure existence, if it has not a substratum in the actual. The member must have some relation to the constituent” (1792). For Burke this basis or substratum in objective interests takes the form of representing accurately “the expression of needs or suffering or symptoms, which he refers to as ‘feelings,’ ‘sentiments,’ or ‘desires’” (Pitkin 1967, 183). Pitkin explains why Burke views these elements of constituency as reliable guidance for representatives, though opinions are not: “Unlike opinions, feelings are reliable; and people are seldom mistaken whey they perceive a pain or symptom, be it physical or political. It is only when they attempt to speculate abstractly on the basis of what they feel that they go astray into opinion” (1967, 183). The representative receives this information and then goes to work, acting for the benefit of the people, to address “symptoms.” The representative does not access this substratum – these objects of representation – through consultation with the people. Any number of agencies, including churches and schools, as examples, can access this substratum (Pitkin 1967, 189). Reflected through these agencies, this substratum of feelings can become the data upon which deliberation takes place in Parliament. The representatives are constrained and oriented by these feelings, which are objective and pre-formed (prior to the relationship of representation). The activity of representation only takes place after the data has been formed and conveyed. It is the 42  responsibility of the representative to accurately express those feelings before Parliament in order to deliberate and arrive at a consensus, at the right answer (Pitkin 1967, 188-189). Burke says, “The virtue, spirit and essence of a House of Commons consists in its being the express image of the feelings of the nation” (Burke 1949 [1774], 28, cited in Pitkin 1967, 183). Trusteeship, working directly upon constituents’ “feelings,” in effect bypasses the agency of the person, at least insofar as agency includes judgment. In fact, the “accurate transmission of popular feelings thus is a prerequisite to representation. It is not, to Burke, in itself representation; it is the material on which representation works” (Pitkin 1967, 188). Indeed, interests exist independently of persons: a person need not express, deliberate, or otherwise help define interests; they are already there. The accountability of the representative is therefore to Parliament, not to the people. There is, then, a form of accountability in Burke’s model of representation, but it is not a form of accountability that is recognizably democratic. Thus, although Burke’s notion of representation as trusteeship initially seems to underscore the activity of the representative – a representation as “acting for” – it ultimately does not provide an understanding of the representative’s accountability to his constituents. Interests exist independently of persons and so constituencies have no role to actively undertake in expressing interests to representatives. Representatives may pursue interests even against the will of the constituency, though Burke expects that over time, the constituency will accept that the representative is right, provided that the representative is honest and competent. Because of the objective nature of interests, “Long-range disagreement between representative and constituents can occur only if the representative is corrupt or incompetent” (Pitkin 1967, 180). Of course, the representative will be subject to a judgment of his performance at the time of election – but it is a judgment made by a constituency capable of feeling pleasure or pain; it is not judgment that relies on capacities of the constituency. Like Hobbes, who treats authorization as a foundational moment and not as a reciprocal process, Burke treats accountability as a 43  terminating moment and not as an ongoing assessment of performance. Hobbesian and Burkean accounts of representation provide no meaningful role for the constituent in assessing representation beyond the following: with Hobbesian authorization, people can say “yes,” and with Burkean trusteeship, people can say “no.”  Pitkin’s Innovation: Representation as “Acting For” Pitkin’s approach to defining representation is to ask about the substance of representation and the relationships representation entails. She does so by conceiving of representation as a “substantive acting for others” (Pitkin 1967, 142) in a manner that is responsive to them, but without collapsing the agency, will, and judgment of either represented into the representative or vice versa. Her understanding thus points to “standards for judging the representative’s action, for deciding whether he has represented well or ill” (Pitkin 1967, 142), and so provides what is missing from the Hobbesian and Burkean understandings of representation. In particular, political representation means, acting in the interests of the represented, in a manner responsive to them. The representative must act independently; his action must involve discretion and judgment; he must be the one who acts. The represented must also be (conceived as) capable of independent action and judgment, not merely being taken care of. And, despite the resulting potential for conflict between representative and represented about what is to be done, that conflict must not normally take place. The representative must act in such a way that there is no conflict, or if it occurs an explanation is called for. He must not be found persistently at odds with the wishes of the represented without good reason in terms of their interest, without a good explanation of why their wishes are not in accord with their interest. (Pitkin 1967, 209-10)  44  Building on this substantive conception of representation, Pitkin conceives of democratic representation as requiring the institutionalization of such responsiveness so that there is not “merely occasional response when [the representative] pleases, but regular, systematic responsiveness” (Pitkin 1967, 234). A dictator might choose to respond to the wishes of his subjects – but he is not a representative unless he institutionalizes that responsiveness. This feature of democratic representation, according to Pitkin, explains our preoccupation with elections and electoral machinery; systematic responsiveness is difficult, if not impossible, without elections (1967, 234). Despite her view of the shortcomings of formal theories of representation, Pitkin’s conclusion (to emphasize the activity of the represented and their ability to determine what the representative should be doing) seems to have reinforced a tendency among political theorists and political scientists to focus on the formal procedures of authorization and accountability and the proper design of electoral institutions (Amy 1996; Barber 2001; Christiano 1996; Guinier 1994). This is important work: the design of electoral systems may influence the broader political culture and may encourage cooperation or deepen animosity among social groups within a society (Rabinder James, 2004). Electoral systems also affect the ways in which interests and voices of groups are represented in legislative bodies. Single member plurality systems, for example, underrepresent women and minorities. Proportional representation systems (PR) have been shown to give more scope for women legislators than plurality-majority formulas (Lijphart 1994). Indeed, PR systems have been increasingly adopted to reflect the importance of “descriptive” representation: representation which “depends on the representative’s characteristics, on what he is or is like, on being something rather than doing something” (Pitkin 1967, 61). Most “proportionalists” today follow this descriptive logic: “there is probably no parliament in the world that has not been criticized for the ‘overrepresentation’ of middle-aged 45  males, university graduates, certain occupational backgrounds such as the legal profession or the public sector, a specific ethnic group, etc.” (Andeweg 2003, 148). In order to address this issue of over-representation, the proportionalists design electoral systems to advance inclusion. As Melissa Williams notes, “when historically marginalized groups are chronically underrepresented in legislative bodies, citizens who are members of those groups are not fairly represented” (1998, 3). In the same vein, Iris Young argues that “designers of democratic processes can promote greater inclusion in decision-making processes as a means of promoting more just outcomes” (2000, 17). Though Pitkin wrote her book long before descriptive representation became a topic in democratic theory, her worry that descriptive representation fails to identify mechanisms of accountability remains relevant. For example, the proportionalists who emphasize the similarity between the composition of the electorate’s party preferences and the composition of the legislative body mistakenly “take it for granted that to represent means to resemble or reflect accurately” (Pitkin 1967, 64). Of course, the resemblance of a representative to his/her constituency does not necessarily ensure a substantive acting for that constituency. In fact, if a representative has only to resemble his/her constituency, then representing is not something that one person can do well or badly; it is only that one representative may be more “typical” than another (Pitkin 1967, 90). On this model, “representing means being like you, not acting for you,” and so “leaves no room for representation as accountability” (Pitkin 1967, 89). Pitkin is well aware of the distinction between what she calls “form” and “substance.” She argues that representative institutions (form) may not always fulfill that substantive acting for others (substance). Thus, her book closes with the encouragement to “construct institutions and train individuals in such a way that they engage in the pursuit of the public interest, the genuine representation of the public; and, at the same time, to remain critical of those institutions and  46  that training, so that they are always open to further interpretation and reform” (Pitkin 1967, 240). Although Pitkin’s points are well taken – particularly with respect to accountability – it is important to note that contemporary theorists of descriptive representation are primarily concerned with the representation of voice and perspective within legislative bodies, and value descriptive representation because it provides a mechanism for reflecting the experiences of women, minorities, or other historically underrepresented groups in legislative bodies. Although contemporary defenses of descriptive representation provide no instructions for responsiveness (acting in the interests of the represented, in a manner responsive to them), they do address at least one of the deficits evident in Burke. Whereas Burke portrayed interests as unattached, that is, as disconnected from peoples’ experiences of them, descriptive similarity provides the means through which “substance” is mirrored into a legislative forum (Dovi 2002, 2007; Mansbridge 1999; Phillips 1995, 1998; Williams 1998; Young 2000).  Representation versus Strong Democracy There is, however, another limitation to Pitkin’s articulation of the theory of representation that most contemporary democratic theorists avoid altogether. The debates articulated by Pitkin were largely content to equate representation with electoral politics. Because they do so, democratic theorists have also focused the debate on electoral representation, attending largely to participatory and, more recently, deliberative democracy. Of course, “Initially, neither the concept [of representation] nor the institutions to which it was applied were linked with elections or democracy, nor was representation considered a matter of right” (Pitkin 1967, 3). Its association with popular representation or self-government evolved through parliaments in England. Knights and burgesses were called to meet with the king’s  47  council – a “chore and a duty, reluctantly performed” (Pitkin 1967, 3). Over time, this chore began “to be used as a device for furthering local interests, as a control over the power of the king” (Pitkin 1967, 3). By the seventeenth century, the principle that members of parliament should be elected by males with stakes in the community – principally, property – was widely accepted enough to become key rally cries for the French and American Revolutions (Pitkin 1967, 3). Representation came to mean popular representation. There was another tradition, originating in political theory with Rousseau, that viewed representation, however necessary, as a lesser form of democracy than participatory or direct democracy. Rousseau is a strong critic of representative government, claiming that if we do not directly govern ourselves, we are not free; “the moment a people gives itself representatives, it is no longer free; it no longer exists” (1978 [1762], 199). Following Rousseau’s critique, participatory democrats have tended to view representation and participation as opposites. To use Benjamin Barber’s (1984, see also Pateman 1974) terms, “thin” liberal democracy makes use of representation, while “strong” democracy makes use of direct, fully engaged participation. The exception to this view of representation as a lesser form of democracy came largely from those theorists interested in the exclusion of women and marginalized groups (Mansbridge 1999; Phillips 1995, 1998; Williams 1998; Young 2000). Anne Phillips (1995), for example, hoped to provide policy recommendations to enable the participation of those excluded from politics through the increased representation of marginalized groups. Melissa Williams argues that “marginalized group identity should play an important role in defining legislative constituencies” (1998, 9). Perhaps these theorists might agree that representation is aristocratic, insofar as elections have historically resulted in parliaments and senates that excluded women and minorities. However, the history of political exclusion led to the conclusion that representation was necessary for participation and particularly inclusion. 48  More recently, progressive democrats have criticized the conceptual opposition of participation and representation and its related judgments as to more and less authentic forms of democracy as obscuring what is at stake in expanding democracy. As Iris Young argued, “the assumption that representative institutions are incompatible with deep democracy…stems partly from wrongly opposing participation and representation” (Young 1999, 8). David Plotke’s formulation of the issues is particularly forceful: “the opposite of representation is not participation. The opposite of representation is exclusion. And the opposite of participation is abstention” (1997, 19). Plotke accuses participatory critics of representation, including Pitkin, for linking representation with absence (1997, 29). As Pitkin suggests, “[R]epresentation, taken generally, means the making present in some sense, of something which is nevertheless not present literally or in fact” (1967, 9). The real absence, Plotke suggests, is that representation has not been considered “a relation, one in which both parties are active” (1997, 29-30). Rather than thinking of representation as absence, we should make a distinction between physical and political presence; one who is represented may not be physically present, but is made present politically (Plotke 1997, 30). Such arguments break down the opposition between participation and representation – an important conceptual advance. But they also remained largely committed to the identification of political representation in democracies with electoral democracy. Political theorists were by no means unique in this respect: they were following the more general identification of democracy with electoral representation that was still little questioned within political science. Indeed, most political scientists today would still agree with Schumpeter (1947): democracy is electoral representation (see also Huntington 1991; Linz 1978; Lipset 1959, 1981; Pennock 1979; Przeworksi 1991; Przeworski, Stokes, and Manin 1999). The return to the political theory of democratic representation, however, has had the effect of opening up the concept of representation to questions that have always been intrinsic to 49  the concept, but pushed out of sight, as it were, by the equation of democratic representation with elections. Pitkin was surely right to focus on substance, even if she shares the blame for singularly emphasizing electoral democracy as the medium of representation. In emphasizing substance, however, we can once again ask some fundamental questions. Perhaps the first question is: if political representation is a “substantive acting for others,” who are the “others”? Who are those being represented?  The Question of Constituency By combining Pitkin’s view of representation as “substantive acting for others” with more contemporary concerns about the representation of peoples as group members, we can form a new, more radical question, albeit one that Pitkin herself does not pose: What is the nature of constituencies, and how are they constituted? The standard account of representation in Anglo-American liberal theory tends to assume territorial constituencies: individuals are represented primarily on the basis of their residence. The “substance” that is represented, on the standard view, is the interests that arise on the basis of where individuals live. However, constituencies based on territory can mask non-territorial issues, such as religion, gender, and ethnicity, and extra-territorial issues, such as global trade and the environment, each of which affect groups and individuals in ways relevant for representation. Of course, as Urbinati and Warren argue, “nonterritorial interests are not new to democratic theorists” (2008, 390). “The main object of disagreement of the Weimar Republic, for example, was whether representation should represent individuals or corporate interests” (Urbinati and Warren 2008, 390). That said, it is still true that in “modern constitutional democracies…the older corporatist views of parliaments and representation have given way to the representation of individuals whose only commonality is residence. Thus, legislatures attend to nonresidential constituencies only  50  indirectly” (Urbinati and Warren 2008, 390). Representing individuals and the interests that they have on the basis of their residence only indirectly captures the range of non-territorial interests that also affect groups and that also require representation. The standard assumption within democratic theory is that constituency formation is predemocratic. Robert Dahl, for example, simply states that we cannot solve the problem of the domain and scope of democratic units from within democratic theory…The criteria of the democratic process presuppose the rightfulness of the unit itself. If the unit itself is not proper or rightful—if its scope and domain is not justifiable—then it cannot be made rightful simply by democratic procedures. (1989, 207) Yet, notwithstanding Dahl’s claim, the question of constituency is emerging as a problem for investigation within democratic theory, and about which democratic claims can be made. Andrew Rehfeld (2005), for example, suggests that political representation should no longer be territorially based. The definitions of electoral districts, he argues, comprise one of the most important political decisions a system can make, though one that is not, typically, viewed as part of “democracy.” Electoral districts are defined by the state as a condition for “democracy” rather than being conceived, as they should be, as a key element of democratic decision-making. “When constituencies are defined by ‘where people live,’ representatives have incentives to maximize territorially defined resource distributions and advance territorially specific interests” (Rehfeld 2005, 152). Thus, territorially based districts amount to a decision to skew law and policy in favour of local partial interests and against broad, non-territorially-organized public interests. As an alternative, Rehfeld suggests assigning citizens of large nations into random groups, each of which would have a single representative. In this way, Rehfeld challenges the unquestioned assumption of territorial electoral constituencies as the foundation for political representation. 51  Rehfeld’s (2006, 2009) recent work has moved to account for non-electoral and nondemocratic relationships of representation, by distinguishing the question of representation separate from electoral institutions. Rehfeld observes that standard accounts of political representation “explain why one is or why one fails to be a representative at all by reference to democratic norms: a representative is purportedly someone who looks out for the substantive interests of those who elected them through free and fair elections” (2006, 2). Yet other sorts of actors are commonly received as representatives, despite the absence of electoral authorization and accountability, and despite the possibility that they may or may not actually pursue the interests of those they represent. Rehfeld offers the example of the United Nations receiving actors as political representatives “whether or not they were selected according to democratic procedures” (2006, 1). Less formal examples include non-governmental organizations and environmental groups. Rehfeld’s point expands to many other kinds of post-national and transnational actors as well, such as the European Union and various treaty organizations, which make representative claims and provide representative functions for extra-territorial constituencies. To account for an actor’s status as a representative when that actor has not been elected, Rehfeld outlines a “general theory of political representation” that focuses on the existence and judgment of an audience, and in so doing takes an important step in theorizing representation beyond electoral democracy. “Representation,” Rehfeld argues, “really does happen whenever a particular audience recognizes a case that conforms to whatever rules of recognition it uses, regardless of whether these rules are just or unjust, fair or unfair, legitimate or illegitimate” (2006, 4). According to Rehfeld, The Audience is the relevant group of people who must recognize a claimant as a representative, and the relevance of the group will always depend on the particular Function of a case of representation. Representation is always in service to some purpose 52  or function; it is never “had” just to have it—a representative does not merely “stand for” another, she “stands in for another in order to perform a specific function.” (2006, 5) The audience, in short, is that group of people who must recognize a claimant as a representative, determined by “the function” of representation. In Rehfeld’s example, if a representative’s function is to stand for Libya and express policy preferences before the voting members of the World Trade Organization (WTO), then the WTO is the relevant audience as a result of that function. And in accepting that representative, the WTO does not have to appeal to criteria of legitimacy in order to recognize that this is, in fact, a relationship of representation, but instead can recognize a case on the basis of whatever rules it employs. According to Rehfeld’s formulation of the concept, then, “political representation, per se, is not a particularly democratic phenomenon at all” (Rehfeld 2006, 2). According to Rehfeld’s account, the reason that Bono is, in fact, a representative of certain Africans (what I term the affected constituency) is because he claimed to be, and the relevant audience (leaders of the G8 countries) accepts his claim. Indeed, so long as the relevant audience does accept him as their representative, he is the representative of certain Africans whether or not those Africans accept him as such. By itself, then, considering representation from the perspective of the audience does not provide us with grounds to make judgments about the legitimacy of non-elected relationships of representation – a line of reasoning recognized by Rehfeld. He argues that the classic trustee/delegate debate collapses together three distinct elements: aims (whether the representative-lawmaker aims at the good of all or the good of a part), sources of judgment (whether the representative-lawmaker relies on his or her own judgment or the judgment of a third party to determine the substance of those aims), and responsiveness (the degree to which the representative-lawmaker is responsive to sanctions). This conflation obscures a more complex political landscape (Rehfeld 2009, 215). These conceptual distinctions must be maintained, argues Rehfeld, in order to provide a general theory 53  of representation that can be applied to non-electoral and non-democratic contexts. Rehfeld suggests that these distinctions apply to any decision maker, whether king or citizen – not only representatives. What is particular to representation, however, is that it “may make salient the particulars of the three distinctions in a way that is important for democratic governance. So, for example, being a political representative might make one’s ‘constituency’ the salient (and normatively important) referent” (Rehfeld 2009, 229). I pursue this line of argumentation in Chapter Four. I am concerned that the affected constituency should be the constituency that determines both the authority and the legitimacy of representation, if we are seeking to identify democratic sources of authority and legitimacy – though, of course, there are many other (nondemocratic) kinds of sources, including, as examples, traditional, charismatic, religious, and some kinds of legal sources. I depart from Rehfeld in my effort to determine the democratic legitimacy of non-electoral representatives and, in particular, self-appointed representatives. Rehfeld’s “general theory of political representation” is useful for helping us to recognize examples of non-electoral representation that the standard account, by collapsing political representation with elections, cannot; but it does not delineate criteria to judge the legitimacy of such representation. It is correct to say that the authority to act as a representative is derived from the relevant audience, but there are examples of self-appointed representatives whose authority to act as a representative is derived not only from its audience but also from its constituency. Jerry Falwell, to name one, derives authority both from the audience (state leaders) and from his constituency (the Moral Majority). However, we should think of the legitimacy of the representative as derived from the representative’s responsiveness to those affected. Specifically, I call a self-appointed representative constitutively legitimate if she is responsive to a constituency whose interests are affected along dimensions of self-development and self-determination. And when the representative is both responsive to, and promotes the empowered inclusion of (is authorized by 54  and held accountable to), the affected constituency, that representation is democratically legitimate. It is this relationship between representative claimants and a democratically conceived constituency – as those potentially affected by collective actions – that is missing from Rehfeld’s account, leaving a gap between his (appropriately) more general and analytically precise account and democratic theory. John Dryzek and Simon Niemeyer (2008) offer another key development in the theory of democratic representation: constituencies can form around discourses. They suggest, for example, that Bono’s representation claim makes sense “not in terms of representing African people, nor in terms of representing a place called Africa, but rather in terms of representing a discourse of Africa” (2008, 481). A discourse, they suggest, “can be understood as a set of categories and concepts embodying specific assumptions, judgments, contentions, dispositions, and capabilities” (Dryzek and Niemeyer 2008, 481). Discursive representation, they argue, is especially appropriate when a well-bounded demos is hard to locate. Rehfeld and Dryzek and Niemeyer provide important movement toward making the concept of constituency into a key problematic in the theory of representation. Nevertheless, they do not directly address the question put so clearly by Dahl: the question of how constituencies – demoi – are formed. But we now have the outlines of what to look for: constituencies can be brought into being by the act of representation itself. Rehfeld, Dryzek, and Niemeyer do not take this step that would formulate the problem; they do not consider representation as a constitutive activity that brings constituencies into being. In this sense, their approaches are continuous with received theories of representation. As Laclau argues, “The main difficulty with classical theories of political representation is that most of them conceived the will of the ‘people’ as something that was constituted before representation” (2007, 162-63). Laclau asks, “What happens if we have weakly constituted identities whose constitution requires, precisely, representation in the first place?” (2007, 160). Representation may not only 55  be the act of transmitting a will, or the act of respecting the will once it has been determined through deliberative procedures, but may be, in fact, the activity of constituting “the people.” Laclau’s formulation is exactly right: “[T]he construction of a ‘people’ would be impossible without the operation of mechanisms of representation” (Laclau 2007, 161). It is this point that I intend to formulate. Michael Saward’s (2006, 2008) work on representation is certainly the most advanced with respect to the constitutive qualities of representation as well as the question of the legitimacy of non-electoral representatives. Saward offers an important articulation and analysis of the concept of representation as a process of claim making. The framework of the “representative claim” shifts attention from elections to “a broad range of representative claimants, asking how, why and whom they represent (if anyone), without our very definitions determining whether and to what extent they constitute cases of representation” (2008, 4). Frustrated by the limits of mandate-independent, trustee-delegate conceptualizations, Saward emphasizes that “both of these perspectives assume a fixed, knowable set of interests for the represented: the capacity to be a ‘delegate’ or a ‘trustee’ is built precisely upon the more or less transparent knowability of the interests of the represented” (2006, 301). In fact, representation involves the act of “depicting a constituency as this or that, as requiring this or that, as having this or that set of interests” (Saward 2006, 301). A representative claim has been made when: “A maker of representations (M) puts forward a subject (S) which stands for an object (O) which is related to a referent (R) and is offered to an audience (A)” (Saward 2006, 302). The maker of representative claims may be the same as the subject, but Saward separates them in order to capture the idea that the maker depicts or portrays him/herself in a certain way as well – as a politician who adequately represents his/her constituency, for example. As Saward explains, a maker of representative claims (M) could claim to represent (S) the interests (O) of a group of people (R) to an observer 56  or intended audience (A). Or, a maker of representative claims (M) could claim to stand for (S) the preferences (O) of a country or region (R) to an intended audience (A), and so on. Saward outlines legitimacy criteria for non-elected claims of representation that include “connecting” criteria, “confirming” criteria, and “untaintedness” criteria. “Connecting” criteria refer to the position of the claimant within a broader set of conventionally legitimate institutions. For example, in the UK, the House of Lords is defended on the grounds that it is subordinate to the elected House of Commons. “Confirming” criteria focus on whether the claimant identifies a particular constituency and how constituencies respond to this claim. For example, Saward asks, do the claims of Bono or Martin Luther King evoke a clear and reasonably bounded sense of constituency? What evidence is there that such claims are accepted or rejected? And “untaintedness” criteria focus on the immediacy and authenticity of claims. Claims deliberately located outside of formal institutions, for example, attain an authenticity in part because they bypass those institutions that have historically marginalized or excluded particular interests, and so are “untainted” by them. For example, the representative legitimacy of the “No Sweat” movement – a movement that organized boycotts of college and university licensed apparel in order to establish overseas labour monitoring regimes – stemmed in part from its directness. The movement bypassed governments in order to directly focus activism on sweatshop conditions and remedies (Saward 2008, 15-20). An important advantage of Saward’s approach is that he conceives of the concept of constituency in generic terms rather than as a traditional electorate. He deploys confirming criteria to query whether there is a constituency that could reasonably be expected to respond to a claim of representation. However, the problem of democratic legitimacy can be more precisely and directly framed than does Saward by linking it to a normative concept of a democratic constituency, that is, a constituency defined by the affected interests principle. Such a concept of constituency formation would enable an alternative conception of democratic representation, beginning with the problem of who needs 57  representation – namely, those affected. I will argue this case in subsequent chapters.  Representation, Responsiveness and Empowered Inclusion In addition to the framework of the representative claim, other elements in a theory of democratic representation provide resources for an inquiry into self-appointed representatives. As I suggested in the introduction, self-appointed representation may be an important complement to electoral representation, because of its potential to promote both responsiveness to, and empowered inclusion of, those affected by collective decisions and actions, in ways that electoral representation cannot. As I shall argue in subsequent chapters, these two functions, if present, suggest that certain kinds of self-appointed representation are democratic. These criteria may be extended to contexts where there is an absence of electoral representation, both within states without electoral democracy and at the global level, where they translate into democratic representation for voices with legitimate claims to representation but which would otherwise remain excluded. To make this argument, I will further conceptualize representative relationships that, while still democratic, are not grounded on a formal act of authorization nor on a formal mechanism of accountability of the kind achieved by electoral representation. Lisa Disch’s (2009) recent work provides a generic understanding of responsiveness that enables me to identify its workings in non-electoral contexts. In practice, “responsiveness” refers to congruence between the attitudes of constituents and of representatives on policy questions (Eulau and Karps 1977, 233; Miller and Stokes 1964). As Disch explains, The assumption behind this model of measurement is twofold. First, it is that representative government is democratic insofar as it responds to stable and coherent citizen preferences, principally as measured by opinion polls (Manza and Cook 2002, 631). Second, it is that citizen preferences are sufficiently independent of the process of  58  representation to mark the difference between a manipulative popular regime and a democratically responsive one. (2009, 1) These assumptions do not hold when faced with the question of self-appointed representatives: in this instance, no stable and coherent constituency predates the activity of representation. As Disch points out, the two assumptions are too restrictive to capture generic forms of political representation. As such, I instead work from the baseline view that representation is integral to the process of constituency formation. In the context of the self-appointed representative, as one example, there is no constituency until a claim of representation is offered, and the people authorize that claim – or not. Indeed, the process of rejecting the claim may simply be to disavow that they are a constituency; potentially, however, that disavowal may constitute a constituency as one that recognizes shared interests but who disavow the putative representative. The content of the claim must resonate with, or act as a lightning rod for, a people, causing them to rally around, or unify around, the claim. Representation, then, is not a relationship between a principal and agent in which the agent is the sole focus of activity. Pitkin initially focuses on the activity of representatives (they should act “in the interests of the represented, in a manner responsive to them”), but she ultimately emphasizes the activity of the represented and their ability to determine what the representative should be doing. The reason for this shift in focus is likely a result of the fact that citizens will differ in their opinions on what constitutes the proper activity of a representative, and therefore require mechanisms for expressing approval or disapproval of a representative’s actions. This formulation of the principal-agent relation emphasizes the importance of not just responsiveness but also accountability. My understanding of the relationship of representation suggests that the interests of an individual or a group are formed by and through the relationship of representation, and so negates the idea of a canny agent who employs a principal. Put simply, representation, in 59  general, is not a dyadic relationship between a principal and an agent, where the agent is the sole focus of agency. Instead, as Disch explains, “Political representation…is uniquely reciprocal. It is an activity where agency cannot be the property of either the represented and representative” (2009, 5). As we shall see in Chapter Three, even in formal contexts, an actor must offer a claim of representation (agency) and a constituency of people must authorize that claim (agency) in order to produce a relationship of representation. Importantly, it is only a claim until it is authorized. The authorization of the claim is significant because it solves a traditionally assumed limitation of democratic theory – so neatly stated by Dahl – that the question of the constituency itself is not susceptible to democratic interrogation. I shall argue that self-appointed representation has the potential to encourage the formation of the “right unit” (to use Dahl’s formulation) from the perspective of the democratic norm of affected interests. Those whose interests are affected by collective decisions are not always neatly captured by, for example, electoral districts, and so do not have a say in the laws to which they are subject. From the standpoint of democratic theory, the point of democratic representation is to include the voices and perspectives of those potentially affected by collective decisions in making those decisions. Once representation is viewed as a process of claim making, representation is not necessarily tied to formal representative institutions, such as elections and electoral districts. Therefore, “the people” can be expanded to include those who do not neatly fall within such boundaries, and/or can be formed along non-territorial interests. When those affected by the representative claim are empowered to authorize that claim, and if they do so, a democratic constituency is formed. This conception of constituency is implied, though obliquely, in Nadia Urbinati’s discussion of “Representation as Advocacy.” “Too often,” she argues, representation has been given merely an instrumental justification and has been seen as a pragmatic expedient to cope with large territorial states, or as a useful “fiction” by means 60  of which the method of division of labor has been adapted to the function of government. (2000, 758) Instead, Urbinati argues that representative democracy is “not simply prudentially necessary but also valuable in itself” because of the reflexivity it introduces into representative processes (Urbinati 2000, 768). According to Urbinati, because it “lacks simultaneity in political deliberation and decision making” (2000, 765), “representation creates distance between the moments of speech and decision and, in this sense, enables a critical scrutiny” (2000, 768). If we combine Saward’s framework of representation as claim making with Urbinati’s emphasis on the distance between the moment of speech (claim) and decision, the self-appointed representative emerges as a figure that offers a claim of representation that may then be authorized, thus generating a constituency (or not). Moreover, Urbinati helps us to see that representation provides a critical distance between the moment of speech and decision. This critical distance is what provides representative relationships with the space within which the elements of the claim can be judged: the existential, constitutive, and performative aspects that identify the self-appointed actor, the constituency, and the good/purpose/issue of concern, respectively, to be discussed in Chapter Three. Despite the value of her commentary on representation as a process mediated by time and judgment, Urbinati’s understanding of representation as advocacy remains unclear about the constitutive features of representation with response to interests. Democratic theorists have tended to label “advocacy” those activities that do not fit the traditional understanding of representation: an actor disconnected from formal politics, representing an unbounded demos on a particular issue or set of interests. Of course, once we start to separate the understanding of representation from elections, the activity of representation is understood in much broader terms, and includes a broader sphere of activity, which may include advocacy. According to Pitkin, “We need representation precisely where we are not content to leave matters to the 61  expert; we can have substantive representation only where interest is involved, that is, where decisions are not merely arbitrary choices” (1967, 212). In this sense, representatives can be advocates. [A]dvocates plead the causes of others or defend a cause or proposition. Advocacy captures what is unique about these transnational networks: they are organized to promote causes, principled ideas, and norms, and they often involve individuals advocating policy changes that cannot be easily linked to a rationalist understanding of their ‘interests.’ (Keck and Sikkink 1998, 8-9) Like the standard account of representation, the “advocacy” concept of representation implies that interests pre-exist the representative relationship, and that this pre-existing interest is what is represented. In contrast, as I shall argue in Chapter Four, much of the work of representation involves constituency formation.  Representation and Civil Society There is yet another literature of direct importance to my project – the literature concerned with the democratic legitimacy of civil society actors such as non-governmental organizations (NGOs), public figures, and other kinds of representative claim makers. While I will discuss much of this literature in subsequent chapters, here I shall note that the discourse of civil society assumes an account of self-appointed representatives that is demanded, as it were, by the assumed democratic burdens and functions of civil society (Cohen and Arato 1992; Edwards 2009; Habermas 1996; Price 2003; Warren 2001). That is, much of the civil society literature shows that democracies “work” when civil society associations formulate issues, needs, and voices in ways that complement electoral democracy (Warren 2001, chap. 4). One of  62  the more impressive attempts to identify criteria of legitimacy of civil society actors is that of Ruth Grant and Robert Keohane, whose work I shall discuss in Chapters Five and Six. Most other attempts to define the legitimacy of civil society actors as democratic representatives, however, are not as successful. For example, Hugo Slim suggests that “legitimacy might be defined as ‘the particular status with which an organization is imbued and perceived at any given time that enables it to operate with the general consent of peoples, governments, companies and non-state groups around the world’” (2002, 9). Slim continues, “From this working definition, it can be further observed that an NGO or human rights group’s legitimacy is both derived and generated. It is derived from morality and law. It is generated by veracity, tangible support and more intangible goodwill” (Slim 2002, 9). But “the particular status with which an organization is imbued and perceived…” may be disconnected from legitimacy, whether constitutive or democratic. In other words, the status of an organization – its public reputation – may have little to do with their responsiveness to affected constituencies. Legitimacy cannot be derived precisely from an actor’s public reputation; public reputation might serve as a means of accountability, in that an actor may be concerned that his/her reputation may suffer if s/he does not act in the interests of the constituency. Public reputation, if positive, may even provide a reason for an audience and/or constituency to authorize that actor’s claims. Slim’s definition of legitimacy, then, is more properly a definition of authority. But authority provided by an audience and/or constituency is constitutively legitimate only if the actor is responsive to an affected constituency. Authority is democratically legitimate only if it is the affected constituency that offers it. As Rehfeld argues, representation really does happen whenever a relevant audience recognizes it, regardless of whatever rules of recognition that audience uses – whether those rules are legitimate or illegitimate. The authority of a representative is derived from an audience (and sometimes also from a constituency), but as I have suggested, and as I will argue, the 63  constitutive legitimacy of the representative comes from their responsiveness to an affected constituency. If the representative acts on behalf of the affected constituency, there is constitutive legitimacy to that activity that is derived from a consideration of interests affected by conditions of self-development and self-determination. And if the representative is both responsive to the affected constituency and also empowers them to employ mechanisms of authorization and accountability, this is democratic legitimacy. This theory of non-electoral representation, I shall argue in subsequent chapters, provides criteria by which to assess the authority and the legitimacy of self-appointed actors, including those civil society organizations of particular concern in the literature on NGOs.  Conclusion This, then, is where the literature leaves us, and where I shall go in subsequent chapters: the framework of representation as a process of claim making provides the space for understanding non-electoral actors, including those who self-appoint, as representatives. According to this framework, a representative does not have to be an elected official; a representative can be any actor who claims to represent others. Because claim making can occur outside, or alongside, electoral institutions, it also provides the space for understanding a nonelectoral constituency as “the people” who may require representation. This framework challenges an assumption made by the standard account of representation: a territorial electoral constituency is necessarily the relevant demos. From the perspective of representation as a process of claim making, representation is not necessarily tied to formal representative institutions, such as elections and electoral districts, and so “the people” can be expanded to include those who do not neatly fall within such boundaries, and/or can be formed along nonterritorial interests. That is, representation as a process of claim making provides the space for  64  understanding a non-electoral constituency as “the people” who may require representation. From the standpoint of democratic theory, the point of democratic representation is to include the voices and perspectives of those affected by collective decisions in making those decisions. Those who are affected by collective decisions are not always neatly captured by, for example, electoral districts, and so do not have a say in the laws to which they are subject. If the problem of representation is considered in light of this normative concern about the inclusion of those affected, then the democratic potential of the self-appointed representative has been found: in the formation of non-electoral, democratic constituencies. This is the case I shall make in the following chapters.  65  Chapter Three: The Concept of Self-Appointed Representation In this chapter I develop the concept of self-appointed representation in two ways: I develop a generic conception of representation that enables us to identify the relationships that comprise self-appointed representation, and I distinguish self-appointed representation from other kinds of political representation. I begin this chapter by outlining the existential, constitutive, and performative elements of the representative claim. Second, I compare the representative claim and its three elements in three contexts: the formal, informal, and selfappointed contexts. This comparison elucidates the functions that self-appointed representation serves in relation to those systems of representation with which we are more familiar. Third, I introduce the problem of assessing the legitimacy of each element of the claim, when operative outside of representative institutions. Fourth, I compare the democratic possibilities of selfappointed representation to electoral representation. I suggest that those who defend electoral representation on the grounds of voting equality beg fundamental questions of inclusion and exclusion, which are located at the level of constituency formation. I conclude the chapter by arguing that self-appointed representatives, like formal and informal representatives, offer claims of representation that, although they occur outside of institutionalized contexts, classify them as political representatives, though not necessarily as democratic representatives – a challenge I develop in subsequent chapters. My conceptual strategy is to employ and develop two concepts discussed in the previous chapter: Michael Saward’s (2006) “representative claim” and Andrew Rehfeld’s (2006) “audience.” Building on these two concepts, I point out that in self-appointed contexts, the claim is creative: it constitutes a constituency, bringing into existence a relationship that we can then judge to be more or less democratic. I shall argue in this chapter that we can refine the bases of this kind of judgment by looking more closely at what happens when an actor makes a  66  representative claim. When an actor makes a claim of representation, she performs three acts: she self-identifies as a representative (existential act), constitutes a constituency (constitutive act), and performs/identifies a good/purpose/issue (performative act). Building on Saward, I shall suggest that the constitutive dimension of the claim proposes a constituency, which, of course, may or may not in fact be (the most) affected by the issue as defined by the representative claim-maker. The performative element enacts the representative relationship, such that it can become an object for an audience or constituency to judge. Distinguishing between the audience, the constituency the representative claims to represent, and the affected constituency, provides two benefits. First, it allows us to detach the concept of democratic representation from elections, potentially expanding the concept to those who self-appoint. And two, it provides the means by which to judge the activity of selfappointed representatives as democratic, or not – an argument I introduce in this chapter and develop in subsequent chapters.  The Representative Claim To understand the nature of a self-appointed representative, I begin with an analysis of the representative claim, the first moment in establishing self-appointed representation. When an actor offers a claim of representation, she performs three acts. First, the actor is performing an existential self-identification as a representative; she claims, “I am a representative of this group.” Second, in identifying a group she claims to represent, she is also identifying that group as a potential constituency, such as a group composed of “the poor,” or “Africans suffering from AIDS,” or “victims of human rights abuses,” or “evangelical Christians,” and so on. Third, a representative claim is also a performative identification of a good/purpose/issue, which would be achieved if the claim were to be successful. For example, “the poor need food and water;  67  give it to them.” As part of the performative claim, those affected by the claim are identified/constituted, which is, I shall argue, a key element of the potentially democratic functions of self-appointed representatives. Take, for example, the case of the International Committee of the Red Cross (ICRC), which self-appoints as representatives of, among other groups, prisoners of war. According to their website, the ICRC works “with the authorities to ensure that people deprived of their liberty [prisoners of war] are treated humanely and according to recognized international standards, which forbid torture and other forms of abuse” (ICRC 2010). This mission statement exemplifies all three elements of the claim: the existential self-identification as a representative (the organization known as the Red Cross); the constitutive identification of a constituency (prisoners of war); and the performative identification of a purpose or issue (in order to ensure the humane treatment of prisoners of war). The performative act suggests that the audience is “the authorities” which, according to the ICRC website, “can mean a range of people – from prison guards and military commanders to rebel leaders and armed opposition fighters,” but may also include “heads of state or government” (ICRC 2010). And the performative act also identifies the affected as prisoners of war. In the case of the ICRC, the constitutive and the performative claims identify the same group – prisoners of war. But there are examples in which a representative claims to represent one group while affecting another. To take an example I introduced in Chapter One, one of Jerry Falwell’s claims was that he represents the Moral Majority who is in opposition to state recognition and acceptance of homosexuality. This claim demonstrates three acts: existential self-identification of a representative (Jerry Falwell); constitutive identification of a constituency (the Moral Majority); and performative identification of a purpose or issue (opposition to state recognition and acceptance of homosexuality). The performative element of the claim reveals that the audience is composed of state leaders, and the potentially affected are 68  gay and lesbian peoples. In this case, then, the represented constituency and the affected constituency are distinct: the Moral Majority is represented, and gays and lesbians are those potentially affected. What does it mean to represent one group while aiming to affect another? And what does it mean to represent and affect the same group? I flag the possibility of an actor constituting two constituencies: one that he claims to represent and the other that he affects and/or aims to affect. This is not a phenomenon unique to self-appointed representation, but it is easier to identify in the case of self-appointed representation than in formal, electoral contexts, because electoral representation builds on the idea that represented constituencies are also the affected constituencies. That said, the distinction makes clear that electoral representation carries the same possibilities: an electoral representative can act on behalf of her electoral constituency, but the act can have effects on constituencies beyond its boundaries, producing, as it were, another constituency of those affected. In the following sections, I build on these observations by examining the way in which the representative claim functions in three representative contexts: formal, informal, and selfappointed representation, outlining how each element of the claim operates in these respective contexts. Comparing the representative claim and its three elements in these three contexts will provide insight into the gaps of representation that are left by formal and informal representation, and how self-appointed representatives step in to fill those gaps.  The Representative Claim in Formal Representation By formal representation, I refer to all forms of representation that have the following qualities. One, representation is about the representation of citizens within their governments. Two, representation is organized through elections and around electoral districts. I include in this category those representative officials – cabinet officials, parliamentary committee staff, the  69  General Accounting Office in the U.S., and the like – that are appointed by elected officials, and whose representative legitimacy might be said to derive from elections. Although appointment potentially raises new kinds of issues, for the purposes of this dissertation I shall treat these representatives as a derivative category of electoral representation. Formal representation is often conceptualized as a principal-agent relationship in which the principal (voter) exercises power over the agent (representative) insofar as the authority granted to the agent can be revoked. The existential claim (“I am a representative of this group”) is a promissory statement (the representative announces his/her candidacy), which is then validated or invalidated by the election. The constitutive claim (the makeup of “this group”) is largely assumed in formal politics: it is the electoral constituency comprised by the relevant district. The performative aspect of the claim is the representative’s platform – the list of actions that the representative and/or party support (which they hope will appeal to the electorate for their votes). A fair election is considered to be the defining institutional feature of democratic representation, serving as the criterion that makes representation democratic. Thus, in electoral politics, the validity of the existential claim (“I am a representative of this group”) is determined through election. The usual formulation of democratic representation – the electoral authorization and accountability of a representative to a preconceived constituency – therefore collapses the distinction between the existential and performative aspects of the representative claim. Put another way, no matter what the performance of the representative for the duration of her time in office – even if she is unresponsive to the interests of the electorate – she may still claim to be a democratic representative because she was elected. Of course, an electorate can remove representatives or parties from office as a response to poor performance, and the threat of that removal from office is in itself, at least theoretically, an incentive for the representative to serve constituents. Nevertheless, it remains true that formal representatives can claim to be 70  democratic by virtue of electoral authorization and accountability irrespective of their performance. Even if a representative were removed from office because of his unresponsiveness, that electorate would not claim he was an undemocratic representative; he was elected and removed from office as per the rules of the game. Put simply, once an actor is fairly elected, he is a democratic representative regardless of whether or not he achieves any good while in office. Simply by looking at how and where representation occurs, the democratic limitations of formal representation are revealed. Because formal representation collapses together and conflates the existential and performative aspects of the representative claim, formal representation does not require the establishment of the constitutive “this group” because the electoral constituency is preconceived/assumed in some way. In fact, almost all representative democracies assume constituencies based on territory without considering alternatives (Rehfeld 2005, 9). Urbinati and Warren note that, The central feature of the standard account [of representation] is that constituencies are defined by territory; individuals are represented insofar as they are inhabitants of a place (Rehfeld 2005). Beginning with the formation of the modern state, territorial residence became the fundamental condition for political representation—a condition more inclusive than status- and corporate-based representation. (2008, 389) The result is that electoral representation is often bounded in ways that capture the interests of those who reside in a given territory, though even then, only insofar as their interests are territorially bounded. Decisions and actions by representatives that affect those outside of the electoral district, or which affect issues that are non-territorial in nature, are disconnected from affected constituencies. Moreover, representation by territory or residence provides a simple egalitarianism (by virtue of the equal vote), but it may produce or enable or fail to name many inequalities in other dimensions. This vision of representation is therefore unsuitable for global 71  politics, but may also be unsuitable for the representation of peoples and issues that go unrepresented even within a nation’s boundaries. That is, because formal representation attends primarily to the representation of constituencies based on their residence, it often operates in a context where an electoral constituency and a constituency who is affected are mismatched. For example, law and policy (or the lack thereof) on air pollution emissions made in the United States directly affects Ontario’s smog, which in turn affects people not represented in the decisions and policies. Ground-level ozone (smog), in turn, “has been linked to health deficiencies including reduced lung capacity in healthy adults and children, and increased respiratory infections such as bronchitis and pneumonia, as well as increased hospitalizations for asthma and lung disease” (Government of Ontario 2005, 45). Though affected by these policies and laws in the United States, residents of Ontario are not defined as the affected constituency by the terms of formal representation, even though they are quite directly affected. Likewise, with the permission of the Canadian government, in 1987, a giant nickel smelting operation in Sudbury, Ontario built a giant smoke stack to disperse toxic air pollutants away from the region. The stack had the effect of increasing local environmental quality. It also had the effect of decreasing the air quality in Scandinavia, and increasing their problems with acid rain. The Scandinavians had no say in this decision, though they bore the effects. Normatively speaking, democracy defines “the people” as those whose interests are affected (in relation to good x or y) rather than those who are inhabitants of territory, or who share citizenship (though the affected may share territory or citizenship) – an issue to which I return in Chapter Four. The implication is that electoral constituencies may be undemocratic when they do not capture those interests that are affected by a given law or policy, nor provide that affected constituency with an opportunity to influence those decisions and actions that affect them. Electoral representation, which is meant to enable the self-determination of peoples, undermines this norm of democracy when it excludes affected interests from decision-making. 72  To sum up, the framework of the representative claim and its three elements provides insight into the potential gaps in the practice of democratic representation in formal contexts. The first gap is that the identification of a representative as democratic is disconnected from his/her performance in office. A formal representative is considered democratic because she is fairly elected. In formal contexts, then, “democratic” refers to the form of representation (the electoral process) rather than to its substance (performance), though it hopes the former ensures the latter. The second gap in the democratic practice of representation is that the constituency is largely assumed. Assuming the electoral constituency as the right unit overlooks the possibility of constituencies who may be affected by law and policy of which they are not authors. In other words, territorially based electoral constituencies do not always capture affected interests. This is not to suggest that electoral constituencies are necessarily the wrong unit with respect to certain laws and policies. It is to suggest that, from the perspective of democratic theory, the unit needs to be justified – though, once the question is put, it is clear that in very many important instances, the democratic justifications fail.  The Representative Claim in Informal Representation Informal representation, as I shall use the term, occurs when a formally authorized representative extends the use of that authority to act as a representative for those with whom he or she has no electoral relationship. An example is Barney Frank, a Democratic Member of Congress from Massachusetts who, in addition to his formal activities as a Member of the House of Representatives, acts “as a surrogate representative for gay and lesbian citizens throughout the nation” (Mansbridge 2003, 523). Informal representation occupies a position somewhat inbetween that of formal representation and self-appointed representation. Unlike the former, and like the latter, informal authorization is not granted on the basis of promises made through  73  formal channels, such as electoral mandates. Instead, and in common with self-appointment, informal representation has an entrepreneurial element: the informal representative independently chooses to supplement his/her formally authorized position on an issue of interest. In this case, the formally elected representative self-identifies (existential claim) as an informal representative in relation to a particular non-electoral constituency (constitutive claim, where the constituency here is composed of gays and lesbians), for the purpose of advocating their interests in the U.S. House of Representatives (performative claim, which identifies gays and lesbians as those affected). Strictly speaking, Frank’s activities on behalf of lesbian and gay interests occur beyond the formal legislator-constituent relationship for which he was formally authorized; however, his electoral constituents are sympathetic to his position and understand that “issues concerning gay and lesbian discrimination are important to [him]” (Mansbridge 2003, 523). Unlike self-appointment, informal representation retains an electoral foundation: voters, though not necessarily the affected constituency, can choose whether or not to support the representative’s supplementary activities through elections. Frank can legitimately infer from his re-election that his electoral constituents approve of his work (performance) in the area of lesbian and gay rights – if they did not approve, they could choose to hold him accountable by organizing to vote him out of office.3 Frank has also received informal authorization from those he affects (lesbian and gay interests) as well as from his formal constituency, through discursive 3  My thanks to Jane Mansbridge for emphasizing the point that non-elected representation may  depend quite heavily on tacit consent – that is, on the represented not actively repudiating these informal acts of representation. This is beyond the scope of this dissertation to discuss, but what seems of primary importance is not that the constituents have demonstrated their approval of Frank’s work; rather, it is that they have not voted him out of office. 74  means such as letter writing and public opinion, and they hold him accountable through similar means. Ultimately, however, he remains accountable to his formal electoral constituency, which can remove him from office, thereby stripping him of his ability to act as a representative in the U.S. House of Representatives for lesbian and gay interests. Like self-appointed representation, the authorization of informal representatives occurs outside, or alongside, electoral institutions, and beyond fixed electoral constituencies. The primary authorization and accountability mechanisms of both formal and informal representation, however, are based upon elections and fixed constituencies, even though in this case, the fixed constituency serves as a sympathetic proxy for the informal one. By extending the use of his authority to act as a representative for those with whom he has no electoral relationship, the informal representative represents those whose interests are not captured by territorial electoral constituencies. This type of ad hoc activity responds to the fact that electoral constituencies do not always capture those who are (most) affected by a given law or policy, or those who are without representation because they face inequalities in dimensions that are not captured by territory or residence – such as issues of identity, religion, gender, and so on. Actors like Barney Frank provide non-electoral constituencies with an opportunity to influence those decisions and actions that affect them, and provides a voice for their interests in order to shape those decisions and actions that may affect them. Of course, this type of informal representation does not capture affected interests outside of a given electorate. That is, actors like Barney Frank extend the norm of affected interests to non-electoral constituencies (lesbian and gay interests), but still within the broader boundaries of American citizenship. Informal representation helps to fill some of the gaps in the practice of democratic representation found in formal contexts in at least two ways. First, whereas a formal representative is considered to be democratic solely because she is fairly elected, an informal representative may be judged as democratic or not on the basis of his or her performance. 75  Barney Frank may be considered a democratic representative not only because he was elected as a result of a fair election (the form of representation), but also because he is responsive to and promotes the empowered inclusion of affected constituents (the substance of representation). Of course, because informal representatives are ultimately accountable to their electoral constituencies, Frank’s electoral constituency could interfere with or prevent Frank’s responsiveness to his informal constituency, should they disapprove. Second, whereas formal representation assumes a territorially bound constituency as the affected constituency, informal representation questions that assumption and enables representation to expand to affected constituencies, across electoral districts, though still within the limits of American citizenship.  The Representative Claim in Self-Appointed Representation I argue that self-appointed representation is a subset of non-electoral representation that is connected primarily to civil society and the public sphere, and is disconnected from formal political authority, whether or not it is organized through electoral democracy.4 Democratic theorists might be unsurprised to hear that grassroots actors may fulfil the functions of a selfappointed representative: a constituent who initiates a movement on behalf of others, speaks with individuals on the street, gathers signatures for petitions, and so on, is self-appointed. Indeed, such constituents reflect the Tocquevillian preoccupation with participation in local associations as the key mechanism for a healthy democracy. But self-appointed representation may also be a top-down activity in which an elite actor claims to represent others for political purposes. Here, I am interested primarily in representative claim makers who may not have, at  4  Ambassadors represent non-democratic countries in the UN, but because their authority relies  on formal authority (even if non-democratic), they are not self-appointed. 76  least initially, an organized constituency at all, but who seek entrepreneurially – through their claims – to bring such a basis into existence. Be it an individual, non-governmental organization, or foundation, the self-appointed representative offers a claim of representation that is explicit or implicit. In some cases the representative claim is a clear statement. To return to the example of Bono introduced in Chapter One, in an interview with the London Evening Standard, it was suggested to Bono that he represents the 18-30 year olds who listen to his music. Bono responded: Outside of that I represent a lot of people who have no voice at all. In the world’s order of things, they are the people that count least. They are the 6,500 people who are dying of Aids in Africa every day for no good reason. They haven’t asked me to represent them. It’s cheeky but I hope they’re glad I do. (Iley 2005) With this now-famous claim, Bono fulfils the three acts of the representative claim. First, he self-identifies (indeed, self-appoints) as a representative (“I represent a lot of people who have no voice at all”). Second, he identifies a constituency (those people “who have no voice at all”). And third, he identifies a purpose/issue (representing “people who are dying of Aids in Africa every day”), which locates the affected constituency (those dying of AIDS) as well as the audience (though not clear from his statement above, Bono represents this constituency of people who are dying of AIDS to leaders of developed nations, including former President Bush who established the President’s Emergency Plan for AIDS Relief). However, the representative claim may not be as explicit as Bono’s but still exist and function as a representative claim. For example, Human Rights Watch does not specifically claim a representative relationship. That is, the organization does not make an explicit existential claim, such as “We represent.” Yet it is not particularly controversial to imagine that an actor who claims, as Human Rights Watch does, to “challenge governments and those who hold power to end abusive practices and respect international human rights law,” and actually 77  functions in that capacity, represents those who are victims of human rights abuses. The mission statement of Human Rights Watch serves to embed it in a representative relationship with a specific group by fulfilling the conditions of the existential identification (though implicit), the constitutive identification of a constituency (victims of human rights abuses), and the performative identification of a good/purpose/issue (challenging governments to end human rights abuses). The self-appointed representative self-identifies as formal and informal representatives do, though without the formal authorization that follows from the votes of an electoral constituency. Herein lies the quality of self-appointment. As I discuss further in Chapter Five, some methods of authorization, though not derived from a formal electoral constituency, may provide authorization (and perhaps even democratic authorization). However, because the political authority of a self-appointed representative is not derived from the electorate, from the standpoint of democratic theory, it is unclear whom the self-appointed representative is representing and on what authority. For this reason, some of the literature on representation has referred to the activity of self-appointed representatives as “discursive representation” (Dryzek and Niemeyer 2008; Keck 2003). A discursive representative is one who “[makes] audible the multiplicity of voices whose presence at the table is recognized as appropriate or necessary for deliberation to take place” (Keck 2003, 51). Keck applies this term to an institutional process in which there is no constituency or population represented but rather, positions and ideas (2003, 45). Similarly, Dryzek and Niemeyer (2008) suggest that Bono’s representation claim makes sense, not in terms of representing African peoples, but only in terms of representing a discourse of Africa (2008, 481). However, Bono’s claim does identify a constituency of people: those who are suffering from AIDS in Africa, as well as a group of people concerned about them (the 1830 year olds). The constitutive act of the claim may do more work in the self-appointed context than in formal and even informal contexts because those who self-appoint are not necessarily 78  assuming a given territorial constituency or a group of constituents within a broader electorate. The constitutive aspect of the claim – the makeup of the constituency – is not necessarily assumed in self-appointed representation as it is in electoral politics. The self-appointed representative is not institutionalized to represent a specific people as are our formal representatives; instead the self-appointed representative constitutes the people she represents through the claim of representation. She identifies – offers a definition of – the people and issues to be represented, and they and/or a relevant audience accepts or denies the claim. Amnesty International, for example, claims on their website to have more than 2.8 million members and supporters in more than 150 countries and regions who are concerned about internationally recognized human rights for all (Amnesty 2010). These 2.8 million people are united around Amnesty International’s claim to stop the abuse of human rights, and they authorize that claim by becoming members. The representative claim is offered, and then authorized, and so creates a constituency of people concerned about the abuse of human rights. That is, representation constitutes a constituency (the 2.8 million people concerned to stop human rights abuses), and also constitutes the affected constituency (victims of human rights abuses). The self-appointed representative does not simply bundle together the preformed interests of “the people”; instead, she plays a role in constituency-formation – a role that will be discussed further in Chapter Four. For now, we begin to see that the self-appointed representative provides not only discursive representation of positions and ideas, as when Greenpeace claims that, “this fragile Earth deserves a voice” (there is no human constituency or population represented in the usual sense), but also provides voice representation, as in the case of Amnesty International. Most self-appointed representatives make claims for which they seek authorization: even Greenpeace asks people to donate, sign up, and/or get involved in their campaigns. And at the moment that a representative claim is authorized by a constituency, voice representation is added to discursive 79  representation. In other words, the self-appointed representative potentially retains the connection to a constituency by suggesting who is to be represented (voice). But this constituency of people, as we have seen in the examples of Amnesty and Bono, is not necessarily a formal electorate and so is not a traditional source of political authority and legitimacy. Whereas formal representatives retain a traditional political authority where they are understood to be an authorized representative of a state, self-appointed representatives are actors who make public claims outside of organized state sovereignty. Their authority is derived from their audience – that group of people who must recognize a claimant as a representative according to whatever rules of recognition they use to generate authorization and, hence, authority (Rehfeld 2006) – and sometimes from the constituency they claim to represent. In selfappointed contexts, the constituency is usually not institutionalized, so that the content of a claim has much to do with defining its existence, substance, and boundaries. For example, Jesse Jackson’s RainbowPUSH Coalition claims on its website to, among other things, advocate on behalf of the poor and hungry, a claim that is “addressed to government, business, education and religious leaders throughout the world” (RainbowPUSH 2010). The audience that must recognize Jesse Jackson and the RainbowPUSH Coalition as a representative is composed of these government, business, education and religious leaders to which they appeal. The constituency that recognizes Jesse Jackson and the RainbowPUSH Coalition as a representative is composed of those concerned on behalf of the poor and hungry. The affected constituency is the poor and hungry themselves. It is because the audience (and the constituency) receives Jesse Jackson as a representative of the poor and hungry that he has the authority to act as the representative of the poor and hungry. He is a constitutively legitimate representative because he is responsive to the poor and hungry and he is a democratically legitimate representative because the affected constituency recognizes him as their representative when they participate in his calls for direct action. 80  It should be noted that formal and informal representatives, too, rely on an audience for their authority, though in these contexts, the audience is institutionalized and generally predictable – usually, the majority of the electorate (or, the majority of the votes received from the electorate), though the audience may also be Parliament or Congress, as examples. And though the rules of recognition in formal contexts are thought to be democratic (elections), this is not always the case: the 2001 election of George W. Bush demonstrates that “even if [American citizens] should not have recognized Bush as President, as long as they do, he is their representative” (Rehfeld 2006, 14). So the political authority of all political representatives comes from the audience. As Rehfeld suggests, “The standard, democratic account thus turns out to be merely a special case of the more general phenomenon: political representation arises simply by reference to a relevant audience accepting a person as such” (2006, 2).  The Democratic Legitimacy of Representative Claims The standard account of representation fails to indicate which kinds of the more general phenomenon – the kinds I am classifying as self-appointed representatives – benefit from democratic authority. According to Rehfeld’s account, the reason that Bono is, in fact, a representative of certain Africans is because the leaders of the G8 countries (the audience) accept him as such. Indeed, so long as the audience does accept him as their representative, he is their representative whether or not those Africans accept him as such. Rehfeld clarifies that political authority is derived from a relevant audience. But what kinds of claim ought to be acceptable to different audiences? Nothing that I have described thus far necessarily makes selfappointed representation “democratic,” or recognizes in it democratic legitimacy. An (unelected) actor who offers a claim of representation may gain a following as a result of that claim, and makes demands on behalf of another person or group of people. This describes an  81  instance of (non-electoral) representation, but does not clarify whether it is constitutively legitimate (responsive to an affected constituency) or democratically legitimate (both responsive to and empowering of the affected constituency). How, then, might such non-electoral representation be rendered democratically legitimate? I move beyond Rehfeld to suggest that democratic political authority is derived from the affected constituency. It may be that how an affected constituency feels about Bono standing in for them ought to matter, and that is how the norms of democracy or justice get introduced into the question of the democratic legitimacy of representation. More generically, from the perspective of democratic theory and practice, we want to understand the demos as a democratic source of political authority just insofar as the demos is the affected constituency. My approach is to suggest that democratic legitimacy is reciprocal. That is, the democratic legitimacy of selfappointed representation can be determined by its relationship of responsiveness to and empowered inclusion of affected constituencies. More specifically, self-appointed representation is constitutively legitimate when it is responsive to (promotes the interests of) the affected constituency. Further, self-appointed representation is democratically legitimate when it is both responsive to and also promotes the empowered inclusion of the affected constituency. We can analyze this formulation of democratic legitimacy in terms of its demands on the representative claim. Each of the elements of the representative claim (existential, constitutive, and performative) has conditions of legitimacy. Let me begin with the existential claim, “I am a representative of this group.” A basic conditions of its democratic legitimacy is publicity: the claim must be made public in order to be authorized (or not) and for accountability to be established. If an actor simply acts on behalf of others without enabling the affected constituency to exercise capacities of authorization and accountability, then the claim fails this condition of validity. Publicity is, in part, a matter of representative claims becoming known to the people who are constituted by them. Publicity is also part of the process through which 82  affected people are constituted as a demos by the claim. There is a reflexive element to publicity: unlike the standard account of representation in which (private) preferences are registered through elections, the concept of publicity points to people making collective decisions in a public space (Habermas 1996). With respect to the constitutive dimension, the representative claim is legitimate if the claim constitutes the group on the basis of the affected interests principle. As already suggested, this basic normative intuition that is at the heart of most contemporary democratic theories suggests that those potentially affected by a collective decision should have capacities to influence that decision (Cohen 1971; Dahl 1970; Goodin 2007; Habermas 1996; Held 1996; Whelan 1983; Young 2001). By this standard, the self-appointed representative is a constitutively legitimate representative if she is responsive to those whose self-development and self-determination are or would be affected by collective decisions. Furthermore, in order to be considered democratically legitimate, the affected constituency must be empowered with mechanisms of authorization and accountability to guide/sanction/inform the claims of the selfappointed representative. If the constitutive “this group” is not composed of those affected but rather by those concerned for the affected – Bono’s claims, for example – the claim may still be constitutively legitimate but it is not democratic. This constitutive form of legitimacy is based on a consideration for the interests of the constituency affected by conditions of selfdevelopment and self-determination. The claim has still formed an affected constituency but the constituents are not those empowered with mechanisms of authorization and accountability. Instead, a concerned constituency acts as a proxy on their behalf. In this way, we can speak of a constitutive legitimacy based on a representative’s responsiveness to the affected constituency, while democratic legitimacy is determined by both responsiveness to, and also empowered inclusion of, the affected constituency. While I shall return to these points in Chapter Six, the key point here is that the constitutive aspects of the representative claim identify the 83  constituency as well as the representative relationships which can then be subject to normative assessment. In the context of the self-appointed representative, it is the performative aspect of the claim that is linked more directly to the normative assessment of representation than it is in electoral politics and thereby to the question of how a representative’s democratic credentials can be earned outside of formal electoral systems. I argued above that an elected representative is considered democratic regardless of his or her performance in office: the fact of a fair election is equated with democratic legitimacy. But for a self-appointed representative to be deemed (constitutively) legitimate, he or she must be responsive to those who are affected. Further, to fulfill the requirements demanded by democratic legitimacy, the self-appointed representative must be authorized by and held accountable to those who are affected. In this way, the definition of democratic representation is decoupled from formal representative institutions and is instead linked to standards of constitutiveness and performance, that is, to responsiveness to the right unit, and to the empowered inclusion of that constituency. This is to say that democratic constituency formation begins with those who need representation – namely, those affected. These standards provide the criteria by which judgments can be made about representation that operates beyond formal representative institutions.  How Self-Appointed Representatives Might Advance Democratic Norms While I shall develop these points in more detail in subsequent chapters, here I want to introduce the ways in which self-appointed representatives play a key role in advancing the basic norm of democracy, the inclusion of affected interests, or the affected interests principle. Self-appointed representatives have the potential to contribute to democracy by acting as representatives for those affected interests that may otherwise be excluded from relevant  84  decisions. If self-appointed representatives include affected constituencies in contexts where they would otherwise have no voice in the decisions that affect them – either because a group is not included in a given electorate or the boundaries of a nation-state, or because a group is (historically) marginalized even within their own electorate or nation-state boundaries – then we have good normative reasons to conceptualize and consider such activity within democratic theory. The affected interests principle as a criterion of legitimacy emphasizes the extent to which self-appointed representatives are responsive to the self-development and selfdetermination of affected interests. This criterion differs from that used to assess the legitimacy of elected representatives, particularly, winning a fair election. However, there are democratic norms that are embedded in the idea of winning a fair election that confer democratic legitimacy, which are not exhausted by the electoral form. To rethink representation so that it may conceptualize and assess self-appointed representatives, I consider the criteria that make representation democratic in the context of elections, and determine whether, and if so how, such features can be decoupled from elections. First, electoral representation is considered to be egalitarian and inclusive owing to the universal franchise. Where there is universal suffrage, each citizen has an equal opportunity to vote and each vote is considered to be equal. The concerns about self-appointed representation are that: Although there are ways to subvert its thrust toward equality, an equally weighted vote, especially in terms of mandated or compulsory voting, is about the best instrument we have for equalizing individual power; within the informal realm, there is no formal egalitarian pressure  85  such as that embodied in the vote; and in the informal realm, inequality is likely to be even worse than in the formal one.5 While these objections are, strictly speaking, correct, they beg the question as to the adequacy of voting-based equality to democracy conceived as advancing the affected interests principle. As suggested above, we must always ask: “Equality with respect to what goods?” Voting based on residential constituency formation produces equality only with respect to those interests that are captured by residential location, and is often compatible with exclusion and inequality with respect to interests that fall outside of this kind of constituency formation. Building on this serious democratic deficit, the argument for self-appointed representation is as follows. First, self-appointed representatives do not replace systems based on the equal distribution of votes. Rather, they step in to fill gaps left by formal representation – gaps that represent the difference between formal voting equality and what is normatively required by the affected interests principle. For example, the National Gay and Lesbian Task Force supplements 5  One response might be to take a pluralist perspective, and suggest that the more groups that are  participating in politics, the better. This perspective would suggest that self-appointed representatives have the potential to include more groups in politics – whatever the effects of self-appointed representation on egalitarianism. This is one of the potential benefits of what I term surrogate and democratic self-appointed representatives: these kinds of representatives potentially include groups that have historically been marginalized. I am concerned, however, that self-appointed representatives do not only include more historically excluded groups in politics; rather, they may provide yet another venue of representation for skewed constituencies: groups that are already well served by existing structures of power, and/or would disproportionately affect the self-development and self-determination of an affected constituency. 86  the representative government of the United States, providing representation for LGBT peoples who are not always well served by their formal representatives. Second, in an imperfect, multilayered, complex world, there may be no option to replace these mechanisms with more egalitarian votes. That is, self-appointed representatives function to fill in gaps where representation is non-existent, at the global level and also in states without representation. Third, imagining a world without self-appointed representatives, and only with egalitarian votes, may leave issues unaddressed, leading to more, not less, inegalitarianism, depending on how those votes are counted. That is, self-appointed representatives do not include everyone in a manner that promotes egalitarianism. Likely, they often provide another venue of representation for the privileged. However, they do have the potential to represent those who need representation: those whose interests are not well served by existing structures of power. In other words, selfappointed representation contributes to egalitarianism and inclusiveness when it provides representation to those whose interests are excluded by the design of electoral constituency formation or are subject to oppressive conditions of power. In addition to egalitarianism and inclusiveness, another feature that makes elected representation democratic is that elections are a mechanism by which a constituency can authorize its representatives. In fact, according to the standard account, election is the authorization mechanism that provides representation with its democratic element. There is no electoral constituency that can formally authorize, prospectively or retrospectively, the claims of the self-appointed representative. It is true of course, that the self-appointed representative does not receive electoral authorization. But there are other mechanisms that may provide democratic legitimacy. Indeed, a self-appointed representative does make claims for which s/he seeks to gain authorization – through followings and memberships, as examples, which will be discussed in further detail in Chapter Five. While it is true that an electorate does not authorize a self-appointed representative, 87  those who are affected and who do not otherwise have access may provide authorization. Again, electoral representation masks this potential difference between an electoral constituency and an affected constituency because it assumes the electorate is the affected constituency, which is not always the case. The lack of electoral authorization, then, is both the defining feature of a selfappointed representative and also necessary to its potential democratic function. It is precisely because they are not restricted to the limits of electoral constituencies that self-appointed representatives are able to represent affected constituencies – the topic of Chapter Four. Elections also function to hold representatives accountable to their constituents for their performance in office. Many theorists and political scientists argue that it is accountability that makes representation democratic: “It is because those of us in modern democratic societies can easily deprive [representatives] of power – depose them, if you will – at certain intervals that they have (at least theoretically) the incentive to rule in a way responsive to our interests” (Barber 1984, 145; Grant and Keohane 2005). Some scholars suggest, however, that the effectiveness of electoral accountability is often exaggerated (Spiro 2002): voter turnout is often low, and there is a large literature that suggests the average voter is often uninformed and/or apathetic (Bartels 1996; Campbell, et al. 1960; Delli Carpini and Keeter 1997; Lupia and McCubbins 1998; Luskin 1990; Sniderman, Brody, and Tetlock 1991). Given this reality, some non-election based forms of accountability may do as well as, and perhaps better than, electoral accountability on many measures, including legitimacy. And, as with authorization, it is not that there is necessarily an absence of accountability with respect to actors who self-appoint. Rather, it is that accountability (and authorization) does not come from the traditional, institutional source: the vote of the electorate. Instead, self-appointed representatives may receive authorization from, and be held accountable to, not only an audience, but also non-electoral constituencies and potentially from constituencies based upon the democratic norm of affected interests. 88  All of this is to suggest that self-appointed representatives advance democratic norms when they are responsive to affected constituencies, and when they empower affected constituencies with mechanisms to authorize their claims and hold them accountable for their actions. These criteria – responsiveness to an affected constituency and empowered inclusion – produce four ideal-types, as shown in table 3.1. Table 3.1 – Dimensions of Self-Appointed Representation Empowered Inclusion Responsiveness to Affected Interests +  -  +  Failed Representation  Skewed Representation  Surrogate Representation  Democratic Representation  To summarize, in order to be considered a democratic representative, the self-appointed representative must be both responsive to and promote the empowered inclusion of the affected constituency – that is, a constituency whose interests are affected with respect to their selfdevelopment and self-determination. Put another way, the relationship between the selfappointed representative and the affected constituency is “democratic,” not only if the former is responsive to the latter, but also if the self-appointed representative promotes the empowered inclusion of the affected constituency in order that they may guide, inform, and sanction the representative. If the self-appointed representative acts on only the first of these dimensions – that is, the self-appointed representative is responsive to, but does not promote the empowered inclusion of, the affected constituency – I call this surrogate representation. If the selfappointed representative is unresponsive to the interests of the affected, and promotes the inclusion of those who would undermine the self-development and/or self-determination of the affected, I term this skewed representation. Finally, if the actions of a self-appointed  89  representative fail on both dimensions of responsiveness and empowerment, that is, the selfappointed representative is both unresponsive and exclusive of the affected constituency, despite claiming to be so, representation simply fails. To be clear, I am in no way suggesting that the self-appointed representative is necessarily better than, and/or should replace, elected representatives. Rather, my aim here is to show how the self-appointed representative may step in to fill the gaps left by formal and informal representation: where it is underdeveloped, inadequate, entirely absent, or undemocratic, both at the level of the nation state as well as globally. My aim is to also provide the criteria by which we may judge the activity of the representative. Providing these distinctions – democratic, surrogate, skewed, and failed – helps to clarify where self-appointed representatives work to democratic effects and in democratic ways, and where they do not. As I shall argue in Chapter Four, however, to understand self-appointed representatives as potential supplements to formally democratic institutions in ways that deepen democracy will also require that we rethink the fundamental premise of electoral democracy: the concept of constituency itself. Only then can we begin to fill in the currently large gap between democratic inclusion based on the affected interests principle and the actual performance of democratic institutions.  Conclusion In a case of self-appointed representation, the self-appointed actor self-identifies – indeed self-appoints – as a representative of a particular constituency in relation to a good/purpose/issue. This is not a democratic phenomenon of necessity. Self-appointed representatives may or may not provide representation of affected constituencies. And the selfappointed representative may or may not choose to imbue the affected constituency with the normative status of a citizen, which would allow the affected constituency to create (authorize)  90  and regulate (hold accountable) obligations between them. It is when that actor self-appoints as a representative of an affected constituency in relation to a good/purpose/issue that he may be considered a constitutively legitimate representative, as I shall argue in more detail in the next chapter. And that actor is democratically legitimate only when the affected constituency is empowered to utilize mechanisms of authorization and accountability in order to guide their self-appointed representative (the subject of Chapter Five).  91  Chapter Four: The Concept of Constituency In this chapter, I respond to an assumption prevalent in democratic theory: that “the people” cannot be democratically constituted. Against this claim, I argue that through the process of authorizing and rejecting the claims of those who represent by self-appointment, people can choose the boundaries of their own demos/demoi. The traditional model of the demos, based on the borders of nation-states or other kinds of existing collectivities such as communities, is not necessarily wrong in its definition. But from the perspective of democratic theory, these borders are arbitrary and so need to be justified. As James Bohman suggests, “requiring a democratic justification of borders breaks the constitutive connection between sovereignty and self-determination on the one hand and bounded political communities on the other” (2007, 178). I shall argue in this chapter that the definition of the demos – that is, who is included or excluded from influence in collective decision-making for any particular issue – should itself be subject to democratic justification, in particular, the justification following from the affected interests principle. In traditional democratic theory, this moment of constitution is, typically, viewed as pre-democratic, in large part because of the logical problem that a demos must exist for it to then decide matters democratically. Here, I shall theorize an alternative conception of demos formation, revealed by the phenomenon of self-appointed representation: every time a representative makes a claim, they propose a demos that is potentially affected by the issue identified by the claim. If this demos is able to authorize their representative and hold them accountable, we have an understanding of self-determination separate from formal inclusion as related to government. Under ideal circumstances, self-appointed representatives would engage in a highly responsive process of constituency-formation, constituting alternative, issue-defined, serial, overlapping peoples – in this way, working around the (apparent) paradox that a demos cannot itself be constituted democratically.  92  My point of departure in this chapter is this paradox, which, I shall argue, is a “paradox” primarily because standard theories of democracy overlook the role of representation in constituting the demos. Indeed, as often noted, standard theories simply assume that “the people” somehow already exists. As Margaret Canovan argues, “‘the people’ as an entity or group capable of exercising power is/are not readily available. Far from being a given, it/they has/have to be in some way constructed, mobilized or represented to be in a position either to wield power or to be checked in doing so” (2005, 88-89). Representation, I will argue, is essential to democracy because “the people” does not exist without representation. Current theories of non-electoral forms of representation are more attuned to the constitutive aspects of representation than the standard account, but they either avoid the question of how to form democratically legitimate constituencies outside electoral processes (Rehfeld 2006, 2009; Saward 2006, 2008), or beg the question by focusing on the representation of discourses (positions and ideas) rather than peoples (Dryzek and Niemeyer 2008). In contrast, by combining the affected interests principle with an account of the constitutive elements of representation, this chapter shows that democratically legitimate constituencies may be formed outside of the electoral process, and without begging the question of inclusion and exclusion in the formation of demoi. I shall argue, simply, that self-appointed representatives – those actors who make claims of representation outside of electoral processes – may link non-electoral constituencies to the basic democratic principle that those affected by collective decisions should have some capacity to influence those decisions. Properly conceived and theorized, self-appointed representation neither restricts itself to existing geographical constituencies nor forfeits the question of how to understand non-electoral constituencies in democratic terms. Rather, the phenomenon of self-appointed representation encourages an alternative conception of democratic representation, one that begins with the problem of who needs representation from a democratic perspective, namely, those (potentially) affected by 93  collective decisions. I am not arguing that the mere fact that an actor offers a representative claim is sufficient to this normatively desirable function. I shall argue that a self-appointed representative constitutes a democratically legitimate constituency when it is composed of those affected by collective decisions and actions and who, in turn, authorize the claim. In Chapter Five, I shall provide an account of the ways in which we might understand the relationships of authorization and accountability, which might potentially provide linkages between a self-appointed representative and a constituency. In Chapter Six, I show how the relative successes and failures of these relationships result in more and less successful forms of representation. In this Chapter, however, I am concerned with the constitutive features of representation in general, and of selfappointed representation in particular. I suggested in the previous chapter that, logically, from the standpoint of the representative claim, all authorization is retrospective – the claim is offered and then authorized. The temporal distance between the claim and its authorization implies a process of reflection on the claim. A self-appointed representative offers a claim of representation that needs to resonate with a constituency (along its existential, constitutive, and performative dimensions) in order to then receive authorization. The claim is offered, scrutinized, and then authorized or not, which suggests a reflective self-assertion on the part of those who authorize the claim. The self-appointed representative provides a particular understanding of a group (constitutive) with respect to a good/issue (performative), and that group signs on to (authorizes) that understanding of themselves, or not. In this chapter, I emphasize the possibility that, through these processes of authorizing and rejecting self-appointed claims of representation, individuals can, in principle, decide whether or not they want to be part of this and/or that people. For example, when Martin Luther King, Jr. called on all Americans to protest racial injustice, Americans each had the opportunity to choose whether or not to authorize that claim and whether or not to continue authorizing that 94  claim, thus constituting themselves as a people concerned with civil rights, or not. In other words, when individuals authorize or reject claims of representation, they constitute themselves as a people in relation to the representative claim, thus choosing the boundaries of their own demos. Self-appointed representation enables an understanding of democracy in which citizens themselves define the boundaries of the demos (Bohman 2007, 2). Put another way, representation claims do not merely reflect the will of constituencies that somehow already exist. Rather, political wills are formed through representation – in particular, through the relationship proposed by the representative claim, and then accepted (or rejected) by the constituency or constituencies that the claim brings into existence. From the standpoint of democratic legitimacy, this kind of constituency formation, potentially, reflects inclusion based on the democratic norm of affected interests because it forms a constituency of people by virtue of their domination and oppression by racial injustice. Of course, self-appointed representation does not necessarily reflect this norm: the claims of Dr. King were rejected by many and so formed, in addition to the affected constituency who had been oppressed and dominated, a reactive constituency of people who did not support civil rights. This chapter argues that, contrary to standard theories of democracy, constituencies can be democratically constituted on the basis of the affected interests principle, and that selfappointed representation is an integral part of constituency formation. I first address models of constituency formation that are alternative to the affected interests principle, focusing on the traditional model of the demos as the population of a nation-state, and I outline their limits with respect to democratic theory. Second, I defend the affected interest principle against the literature that suggests it is infeasible as a principle of demos formation. Third, I analyze how self-appointed representation can entail the democratic formation of constituencies. Fourth, I discuss the non-democratic constituencies that are formed when the represented constituency is not composed of those affected. Finally, the chapter concludes that when the affected can 95  choose the boundaries of their own demos/demoi through authorizing (or rejecting) the claims of self-appointed representatives, peoples can be constituted democratically.  The Standard Model of the Demos and its Formation The question of how constituencies are formed is one that is often overlooked by democratic theorists. Robert Dahl observes, I think this is because they take for granted that a people has already constituted itself. How a people accomplishes this mysterious transformation is therefore treated as a purely hypothetical event that has already occurred in prehistory or in a state of nature. The polis is what it is; the nation-state is what history has made it. Athenians are Athenians. Corinthians are Corinthians, and Greeks are Greeks. (1970, 46) Of course, to the extent that “the people” is produced by the circumstances of geography and history, peoples are often arbitrary from the perspective of democracy (Canovan 2005, 110; Goodin 2007, 48). As Canovan argues, “Although the borders of polities have been delivered by historical contingencies, politicians and theorists usually find it convenient to take them for granted and assume that they contain a political community – a people. But this is not necessarily so, and in some cases it is very obviously not so” (2005, 110). Yet fundamental to democracy is that the boundaries of a polity are also the boundaries of inclusion that determine who will, and will not, have a say in its decisions. As Philipe Schmitter points out, the question of boundary “has long been a conundrum of democratic theory since it provides no convincing ex ante mechanism or ex post justification for determining the boundaries of that unit” (2009, 487). Coincident with the model of the demos as one produced by circumstances of geography and history, political representation is traditionally conceived as based on residence and seen to  96  reflect the will of constituencies that already exist by virtue of their location (Rehfeld 2005). We assume that representation makes sense in relation to our residence because we think our interests are defined by our common location. We assume that territory matters (more than other features of our social relationships) because we live in this place with these people, and so we think we will most likely be affected by the actions of, and share interests with, those who live in close proximity to us. “Even as late as the seventeenth century, communities sent delegates to English parliament because it was by local communities that individuals primarily organized their lives” (Rehfeld 2005, 69). Even then, however, territorial representation was adopted simply because it was efficient to do so; not because there were clearly bounded communities of interest. As Goodin notes with understatement, the “correlation between territoriality or nationality or history and shared interests is far from perfect” (2007, 48). It is not necessarily the case – indeed, it is increasingly not the case – that I will be affected by the decisions and actions of those who live nearby, or who share a history, or who share nationality, as much as I will be affected by the decisions and actions of those with whom I share none of these features. Territorial representation was adopted at the American founding without justification and, indeed, is used in virtually every democratic nation without much thought to its alternatives (Rehfeld 2005). Of course, the fact that justifications have not been offered does not mean that the territorial bases of constituency have no justification at all. Andrew Rehfeld (2005) considers four potential justifications for adopting territorial representation, in addition to assumed communities of interest. First, territory as a means of constituency definition can be justified on the grounds that it might encourage local attachment: “when a nation becomes very large it is difficult for citizens to feel any attachment to the larger polity…whose seat of government is often hundreds if not thousands of miles away” (Rehfeld 2005, 59). A second justification might be the protection of property as the first and most fundamental interest of citizens: “constituency lines might be drawn around real estate” (Rehfeld 2005, 60), preventing 97  government encroachment on their property. Third, territorial constituencies might serve the purpose of enabling citizens to choose their own electoral constituencies by changing their residence. Fourth, and perhaps most compelling, territorial constituencies enable democratic practices within the constituencies: “organizing citizens by where they lived enabled the efficient transmission of local information to the national legislature, citizen deliberation, and the tradition of elections as being tangible moments of consent” (Rehfeld 2005, 62). In fact, Rehfeld argues it is only the last of these justifications that received support at the American founding: “the founders gave constituency division by territory per se no normative anchor. It simply functioned as an important means to these other democratic ends” (Rehfeld 2005, 62). Just as in England in the seventeenth century, territorial organization was just how representation happened to be executed, given technologies, not because the territory was a site of clearly bounded, shared interests. Rehfeld usefully demonstrates that an individual or group is always represented, not as an individual or group, but as an individual or group within a pre-established constituency (see also Schmitter 2009, 487). That pre-established constituency was not (and is not) always defined by territory – historically, it has taken the form of guild or status, as examples – but over time, territorial representation persisted without justification, “as a habit of mind, a historical remnant no longer serving its original purpose of representing communities of interests, but so ingrained that it was never seriously challenged” (Rehfeld 2005, 9). Despite these fundamental observations, Rehfeld does not challenge the larger pre-established constituency, that is, he does not ask “why the citizenry as a whole deserve their inclusion within a particular state” (Schmitter 2009, 487). As noted above, in addition to assuming the population of a given state as a demos, most democratic theorists argue that the formation of the demos itself cannot be democratic. Democracy is a method of group decision-making that logically cannot be brought to bear on 98  the initial membership of the group (Whelan 1983; Goodin 2007). It is only once the demos has been determined by undemocratic means that the business of democracy can begin. That is, democracy, in its first instance, is commonly seen as necessarily undemocratic. Thus, democratic theory has an inherent limitation: “Boundaries comprise a problem…that is insoluble within the framework of democratic theory” (Whelan 1983, 16). In his article on “Enfranchising All Affected Interests, and Its Alternatives,” Robert Goodin (2007) looks outside democratic theory, to the literatures on contractarianism and political obligation, seeking a solution to the problem of constituting the demos. Contractarianism suggests that the demos “ought consist of all and only those persons each of whom is not rejected by any of the others as a member” (Goodin 2007, 42). This formula does not provide an analysis of what are (and are not) good reasons for constituting the demos, which involves the (legitimate) exclusion of others, and so would arbitrarily exclude large numbers of people (Goodin 2007, 41-42). A somewhat opposite problem occurs with the literature on political obligation, which suggests that “all and only those persons who are (legally) obliged or (morally) obligated to obey a body of laws ought be entitled to membership in the demos making those laws” (Goodin 2007, 42). This formula does not exclude people from membership in a demos who ought rightly be excluded. There are all sorts of people who are legally obliged and morally obligated to obey laws even though they are rightly not entitled to membership in making those laws. Visitors to a foreign country, for example, are people who are rightly bound by laws but are not, and perhaps ought not, be entitled to a vote in making those laws (Goodin 2007, 42). More generally, the problems with assuming a demos based on the population of a bounded territory are normative, empirical, and conceptual. The standard account of the demos and its representation is normatively inadequate because the affected constituency is not always and/or entirely territorially bound. In fact, however a constituency is defined, whether it be 99  along territorial, racial, ethnic, gender, or religious lines, the affected constituency is not always or entirely captured, depending, of course, on the issue of representation at hand. It is for this reason that Goodin argues that constituting “a demos on the basis of shared territory or history or nationality is thus only an approximation to constituting it on the basis of what really matters, which is interlinked interests” (2007, 49). As Goodin explains, the affected interests principle is the standard by which inclusion in the demos is measured. As such, democratic constituency formation should begin with who needs representation, namely, those whose interests are affected by a collective decision, and, from the standpoint of democratic representation, they should have some say in influencing that decision. As long as we continue to view inclusion in relation to government and in relation to the boundaries of a nation-state, exclusion will occur that is unfair from the perspective of democratic theory. As I shall argue below, self-appointed representation may provide a partial solution to this normative inadequacy by founding constituencies on the basis of the affected interests principle. With Laclau, I argue that that the “task of the representative is, however, democratic, because without his intervention there would be no incorporation of those marginal sectors into the public sphere” (Laclau 2007, 158). In particular, if self-appointed representatives function to enact the affected interests principle, they augment democracy. Empirically, the standard account does little to identify, much less assess, the many new kinds of constituencies represented by post-national and transnational actors, including the United Nations, the World Bank, the European Union, and treaty organizations, as well as by civil society organizations in both formal and non-formal settings. For example, Greenpeace represents more than 2.9 million members from all over the world who want solutions to global environmental problems. This is a non-electoral, non-territorial constituency of people who has authorized Greenpeace through membership, donations, and volunteering to “ensure the ability of the earth to nurture life in all its diversity” (Greenpeace, 2010). Furthermore, there are non100  territorial and extra-territorial issues such as global warming and global trade that do not fall neatly within the traditional models of the demos and its territorial representation, but that affect people who need representation. Dahl refers to this problem when he asks whether there is not after all some wisdom in the half-serious comment of a friend in Latin America who said that his people should be allowed to participate in our elections, for what happens in the politics of the United States is bound to have profound consequences for his country. Do not dismiss his jest as an absurdity. In a world where we all have a joint interest in survival, the real absurdity is the absence of any system of government where that joint interest is effectively represented. (1970, 51) If the empirical problem is the inability, from the perspective of the traditional understanding of the demos, to conceptualize (and judge) non-electoral and non-territorial constituencies, then – as I shall argue – we need to develop a theory of representation that can respond to the fact that representative organizations are now little bound to existing geographical constituencies. The conceptual problem of the traditional model of constituency is the mystery of the origins of “a people” that somehow already exists as a politically canny subject who comes to the ballot box with interests predetermined. Self-appointed representation can provide the missing link between the formation of a people (whether affected or not) and their selfdetermination. As part of the representative claim, the representative offers a definition of a people that must resonate in order to receive authorization from them. The claim of the representative acts as a lightning rod for the people, or mobilizes a people (perhaps peoples) around the claim. As Laclau precisely states, “the construction of a ‘people’ would be impossible without the operation of mechanisms of representation” (2007, 161).6 Following 6  See also Ankersmit 2002; Disch 2010; Laclau 1996; Laclau and Mouffe 1985; McClure 1992;  Saward 2008. 101  Laclau, it is to this constitutive feature of representation that we must attend if we are to address the problems of inclusion and exclusion that come with any defined demos.  The Affected Interests Principle and Constituency From the perspective of democratic justification, inclusion in (and exclusion from) a “people” or a constituency should be justified not by residence but by the affected interests principle: those whose interests are affected by decisions should have capacities to influence those decisions (proportional to the extent to which they are affected). Goodin argues, correctly in my view, that the affected interests principle is the norm by which we should understand inclusion in the demos. Goodin’s formulation, with which I agree, enables a fair amount of proportionality in determining boundaries. While those potentially affected by collective decisions should be included, it is possible to develop cases in which a demos is over- or underinclusive (or both) with respect to interests. Under-inclusion occurs when a person is excluded from the demos whose interests are affected by the decision of the demos (Goodin 2007, 49). Notice, however, that is just to say that the ‘all affected interests’ principle is the principle that is being implicitly employed in judging all those other methods of constituting the demos, on the basis of territoriality, nationality, history, or whatever. The “all affected interests” principle is the standard by which the adequacy of those other approximations is invariably assessed. (Goodin 2007, 49) Where I depart from Goodin is that he considers the affected interests principle only in the context of casting votes for political representatives. In other words, Goodin, like Rehfeld, is still considering inclusion in relation to voting-based representative government, whereas I will use the affected interests principle to conceptualize and judge the many new kinds of nonelectoral constituencies represented in contemporary politics.  102  As mentioned in Chapter One, the principle of affected interests has a long pedigree. The early medieval phrase is “quod omnes tangit, ab omnibus approbetur,” or, “whatever touches all, must be approved by all,” from the Code of the Emperor Justinian, around 534 A.D. Democratic theorists since have often suggested that those affected have rights regarding the decisions that affect them. But exactly what those rights are is highly contested. In this section, I engage with some of the common criticisms leveled at the affected interests principle to achieve three related aims: first, to develop the principle; second, to defend the principle against its detractors; and third, to provide an understanding of the principle in the context of the selfappointed representative. In achieving these aims, I begin to make the case for understanding self-appointed representatives as (potential) founders of democratic constituencies on the basis of the affected interests principle. One objection to the affected interests principle is that who is affected would change with each decision (Whelan 1983, 19; Dahl 1970, 49). Further, and related, Dahl asks, “how is the citizen who is affected by so many different units of government, each with its own procedures and officials, to devote much time and energy to any of them?” (1970, 50). Notably, whereas formal representation involves the drawing of boundaries around peoples, selfappointed representation involves the drawing of boundaries on the basis of issues. In principle, boundaries would change with each issue, following the effects issues have, or potentially have, on individuals. As Warren explains, It has become clear that boundaries may be based on issues, as they increasingly are under doctrines of subsidiarity (the notion that political units should match the scale of problems with which they deal), and in emerging global institutions and forums. In such cases, “the people” is constituted and reconstituted as a self-governing collectivity in a different way for each kind of problem and its effects – say, for purposes of occupation, defense, control of pollution, schooling children, or regulating public health. Implied in 103  this kind of boundary is a complex form of citizenship, in which individuals have multiple memberships, depending upon the nature and domain of collective decisions. (2006, 385) This understanding of “the people” or “peoples” – as one formed not by boundaries of nationstates but by affected interests (which may cross boundaries or be extra-territorial), brings into question the traditional focus of political representation on individuals in a specific location. On this view, who “the people” is varies, depending on how an issue affects individuals. A people with a democratically legitimate claim to representation is comprised of those whose selfdevelopment and self-determination would be potentially affected by an issue. It is true that those entitled to have influence in collective decisions would change with each issue. The concerns about this fluidity are, however, not (typically) normative, but rather functional: shifting “peoples” or constituencies would be infeasible. Certainly this is mostly correct in the context of fixed, territorially based constituencies. But the concern is mitigated in the context of the self-appointed representative because this type of constitution and reconstitution of the people already occurs in response to representative claims. For example, individuals may choose to donate time and/or money to the National Organization for Women (NOW), constituting a people who are concerned about equality for women. Further, some may allow their membership fees to lapse, effectively reconstituting this constituency. Another set of individuals, and perhaps an overlapping set, may also sign petitions for Amnesty International, supporting the work of those who campaign against human rights abuses. The work of constituting the decision-making unit with respect to particular issues is accomplished by the representative claim and its authorization, and with no more effort than is required of an individual to choose which concern(s) is a priority and requires representation in addition to the representation s/he receives as a member of an electorate (where available). Nancy Fraser has noted this advantage in the case of participation in publics: “the unbounded character and 104  publicist orientation of publics allows for the fact that people participate in more than one public, and that the memberships of different publics may partially overlap” (1990, 70). The same holds true, potentially, in the case of informal representation, as well as self-appointed representation. A second objection to the affected interests principle is that determining who is affected is both a logical and procedural impossibility: [B]efore a democratic decision could be made on a particular issue (by those affected), a prior decision would have to be made, in each case, as to who is affected and therefore entitled to vote on the substantive issue…And how is this decision, which will be determinative of the ensuing substantive decision, to be made? It too should presumably be made democratically – that is, by those affected – but now we encounter a regression from which no procedural escape is possible…Thus to say that those who will be affected by a given decision are the ones who should participate in making it is…to propose what is a logical as well as a procedural impossibility. (Whelan 1983, 19) I suggest that representation and the representation claim interject in this process of identification of the affected, halting the logical and procedural regress that concerns Whelan. The self-appointed representative identifies a community of the affected in relation to an issue and offers a claim, which may then be authorized by the affected. For example, when Susan B. Anthony founded the American Equal Rights Association and later, the National Woman Suffrage Association, with the aim of securing women’s enfranchisement through a federal constitutional amendment, women were able to authorize that claim through their membership, effectively constituting themselves as an affected people in relation to the claim. It is in this way, through the authorization of claims, that the constituency whose conditions of selfdevelopment and self-determination are potentially at stake constitutes itself as a demos. The American demos at that time excluded women, and so women formed an alternative, that is, 105  non-electoral, constituency when they mobilized around Anthony’s claim, in hopes of securing the vote and expanding the inclusion of the American demos. A third objection to the affected interests principle is that interests are subjective and so they are always potentially boundless. Dahl puts the objection as follows: Is it to my interest to lower the high rate of infant mortality among residents of inner cities? reduce poverty in Appalachia? Certainly so, for my “interests” are determined by my beliefs and values about the well-being of others, like the groups just named. Notice, however, how this subjective approach enlarges one’s interests and hence the possibility of being affected by the decisions of others. (Dahl 1970, 51) I think this problem is at least partly a problem with the language of “interest,” which can be quite broad and encompassing, as Dahl suggests. Having an “interest” in an issue can simply mean that I am concerned about that issue, though it can also mean that I have a stake in the issue. To address this concern of the subjectivity of interests, I follow Iris Young’s formulation and conceive of “interests” as “what affects or is important to the life prospects of individuals” (2000, 134).7 Interests are those means or conditions required to achieve desired ends, such as conditions of self-development and self-determination. Self-development is facilitated when “just social institutions provide conditions for all persons to learn and use satisfying and expansive skills in socially recognized settings, and enable them to play and communicate with others or express their feelings and perspectives on social life in contexts where others can 7  Young offers useful distinctions between interests, opinions, and perspectives. By opinion, she  means “any judgment or belief about how things are or ought to be, and the political judgments that follow from these judgments or beliefs” (Young 2000, 135). By perspective, Young refers to structural social positions that “produce particular location-relative experience and a specific knowledge of social processes and consequences” (2000, 136). 106  listen” (Young 2000, 31-32). Self-determination is facilitated when a person is able to participate politically (Young 2000, 32). When we conceive of “affected interests,” then, we conceive of those relevant effects that enable or constrain self-development and/or selfdetermination. If self-appointed representatives represent those whose interests are affected by an issue, this means that they are responsive to those effects that enable self-development and self-determination. Stated negatively, some self-appointed representatives may seek to reduce those effects that produce oppression, defined as institutional constraints on self-development, and domination, defined as institutional constraint on self-determination (Young 2000, 31). The “affected,” then, are those whose ability to enjoy the basic conditions required to make life choices has somehow been inhibited. Linking interests with self-development and self-determination removes at least some of the subjectivity of the language of “affected interests” and shows that there can be objective standards by which I am or am not affected by an issue. Ordinary use of language would suggest that I am “affected” by poverty and hunger in the world, but I want to strip that understanding of the word from its meaning in this context. On this formulation, I think we can see that my basic interests (as a relatively privileged Canadian citizen) are not affected by infant mortality in inner cities nor by poverty in Appalachia – I am concerned by these issues, certainly, and concerned enough to donate my time and/or money to self-appointed representatives acting on their behalf, but they do not affect my conditions of self-development and/or self-determination. To return to Dahl’s statement, then, his interests are not determined by his beliefs and values about the wellbeing of others. Rather, his beliefs and values define the ends he desires (ending poverty in Appalachia). His interests, however, define the means to those ends (Young 2000, 134). On this formula, Dahl has the resources – the conditions of self-development and self-determination – to pursue the ends he desires. Let me provide what I think is a more difficult example – one that occurs within formal 107  contexts. Let us say that the distribution of wealth in a given nation-state is drastically uneven, such that 42 percent of the population is living below the poverty line of $1.25 each day. If the government chooses to tax the wealthiest citizens at some marginally higher rate in order to redistribute wealth to the poorest, certainly the wealthy might claim that their interests are affected by that decision. But we could ask, are the conditions of self-development and selfdetermination of the wealthy affected by a marginally higher rate of tax? Conditions of selfdevelopment and self-determination involve the basic conditions required to make life choices, including the distribution of resources and the organization of power, status, and communication. The wealthy could certainly register their displeasure, expressing their feelings and perspectives, and they could still pursue their lives in their own way. For these reasons, I do not think the wealthy could claim that their interests are affected – or at least, not in the same way or to the same extent that the least well off could. It is unlikely that the life choices of someone with $200,000 in yearly income would be much affected by, say, an extra $3000 in taxes, representing 1.5 percent of their income. But that same $3000 would represent a doubling of income for the average person in El Salvador, or a quadrupling of income for the average person in Haiti. While an increased tax rate affects the interests of the wealthy person, the relative affectedness for the poor person is many times greater, which, normatively speaking, should produce a relatively greater claim to representation – based on the relative threats and opportunities for life choices. All of this is to suggest that we can identify those collectivities that have the right to self-determination on the basis of their affected interests, which is, potentially, a very concrete concept. As Michael Walzer (1983) suggests, the courts have developed ways of determining and measuring who is most affected in reference to class-action lawsuits, in which multiple groups of people receive differing amounts of monetary compensation. Of course, even with the understanding of affectedness as defined by conditions of self108  development and self-determination, who is “affected” may still be multiple and contested, pitting some kinds of affected interests against others. With Young, I would like to point out, however, that not all competing interests are necessarily structured as a zero-sum relationship (2000, 134). That said, making decisions among competing interests is often the difficulty of both politics and representation, whether in the formal, informal, or self-appointed realm. Indeed, the point of conceiving of representation in terms of the affected interests principle is not to pre-judge political outcomes, but rather to suggest a principle that would expand boundaries of representativeness to include those who have claims to influence, but whose claims are organized out of political processes by residence-based constituencies and/or other social structures or relations of power. In other words, the principle should support equal opportunities to connect to a representative claimant, while acknowledging greater and lesser claims to representation depending upon the degree of affectedness. A fourth objection to the affected interests principle speaks to related difficulties: even if we accept the principle as an ideal standard for inclusion, it is endlessly expansionary (Goodin 2007; Miller 2009; Song 2010). Goodin seeks to limit the principle by making a distinction between “all actually affected interests” and “all possibly affected interests.” However, he argues that it is the latter that is the only coherent interpretation of the principle, because “it is incoherent to try to determine who should get to vote by asking whose interests are actually affected by the course of action actually decided upon. It is like the winning lottery ticket being pulled out of the hat by whomever has won that selfsame lottery” (Goodin 2007, 53). The implication of such an inclusive principle would be to give “virtually everyone everywhere a vote on virtually everything decided anywhere” (Goodin 2007, 68). I have two responses to this problem. The first is to repeat that when interests are carefully defined with the criteria of self-development and self-determination, as described above, the affected interests principle can be aggressively limiting rather than radically 109  expansionary. But it is true that there will be issues, such as global warming or the spread of AIDS, that suggest it is perfectly appropriate that the unit is even larger than the 600-700,000 people that, on average, form an electoral constituency in the United States. Further to this point, the term “radically expansionary” is likely used in reference to existing standards that are based on highly exclusionary electoral constituencies. As Rehfeld articulates, “political representation enables modern democracies to exclude virtually everyone from the institutions that govern them” (2005, xi). From this perspective, there will be issues that are “radically expansionary” but appropriately inclusive as compared to the exclusionary institutions of formal representation. The second response is that this line of argumentation supports the prima facie importance of self-appointed representatives because it demonstrates that our institutions are not living up to this democratic principle of inclusion. The affected interests principle is easier to achieve within non-institutionalized forms of politics – public sphere politics in particular – than it is in the formal realm in which votes in elections are required by territorially bound constituencies. Influence in non-institutionalized politics is often discursive in nature, a form of influence that is inherently more open to inclusion. In contrast, although it is true that Afghanis, for example, are affected by the election of the American president, and as true as it is that they likely should get a vote in that election, our institutions are structured in such a way as to make this difficult, if not impossible. The fact that self-appointed representatives are required to fill these gaps should tell us that our institutions are not doing all of the work that they should or can, and so highlight the importance of these alternative forms of representation for the purposes of democracy. A final objection to the affected interests principle is related to the temporal coherence of the demos. As Canovan notes, “A number of theorists have argued that in contemporary circumstances, only the ties of nationhood are likely to generate a people with the kind of long110  term political solidarity that is needed to sustain self-rule” (2006, 353-54). The objection to the affected interests principle emphasizes the dangers of a malleable demos, and the concern is two-fold: one, that without a state, rights might be undermined; and two, that trust and solidarity require repeat players (Miller 2009; Song 2010). With respect to the political functions of selfappointed representatives, the first concern is easily dismissed because they supplement rather than replace existing structures of government. The affected interests principle in the domain of the self-appointed representatives does not, in principle, undermine the sovereignty of states, because self-appointed representation is a supplement to electoral representation, where it exists, and an important form of representation where it does not. The second concern, that the stability of the demos is important because of its implications for trust and solidarity, is perhaps more relevant. But it is by no means clear that the peoples formed by self-appointed representatives need to generate “long-term political solidarity” – this is not necessarily their aim, and there are other features of governments and societies that are better equipped to fill this role. Instead, the comparative advantage of self-appointed representatives is that they have the potential to achieve representation outside of the formal institutions of the state, which may be nonexistent, or structurally incapable of providing representation due to the scope of the issue at stake. As Canovan suggests, ‘The people’ cannot be restricted to a group with definite characteristics, boundaries, structure or permanence, though it is quite capable of carrying these senses. Furthermore, its lack of spatial definition is compounded by discontinuity over time. We have seen that ‘the people’ often seems to refer to occasional mobilizations that appear unpredictably and fade away again. (2005, 140) In the context of the self-appointed representative, the demos need not necessarily be the stable, enduring demos imagined necessary for “nationhood.” This speaks to a broader point: that the affected interests principle is less problematic in its practical application when 111  considered in relation to self-appointed representation than it is in relation to formal government. The affected interests principle, though “the best general principle of inclusion you are likely to find” (Dahl 1970, 49), has significant consequences for inclusion in relation to government: its application may very well be radically inclusionary and may mean spending more time on such decisions than we do currently (Goodin 2007, 68). In the context of formal government, the affected interests principle remains a norm by which to critique alternative methods of inclusion, and cannot serve as an actual method of inclusion. In the context of the self-appointed representative, however, the affected interests principle can act both as a norm and also as a method of inclusion (considering a constituency’s potential affected interests), because it does not have the same institutional effects.  The Self-Appointed Representative and Democratic Constituency Formation The democratic potential of the self-appointed representative resides most fundamentally, then, at the level of constituency formation. And constituency formation is fundamental: it determines the initial boundaries of inclusion and exclusion. Moreover, that inclusion and exclusion can be based on the democratic norm of affected interests. If the selfappointed representative functions to enable inclusion on the basis of the affected interests principle, s/he serves to increase democracy. For those parts of the world where there is an absence of electoral representation, or within electoral systems that fail to reflect legitimate claims to representation, the self-appointed representative may provide an important relationship of representation to those who are excluded from formal representation. Democratic nationstates have, of course, electoral boundaries (often tied to territorial units) that limit who shall be included as members. As Iris Young argues, however, when the scope of social and economic interactions does not match the scope of political jurisdiction, the nation-state system can enact  112  and legitimate profound exclusions (2000, 9). The economy operates on a scale far beyond the reach of any individual nation-state, and a number of contemporary issues also highlight the inadequacies of the territorial basis of representation. For example, Bono’s organization, One, has suggested that climate talks should focus on Africa’s poor as it is they who will suffer the earliest and the most (One 2009). In this way, the self-appointed representative encourages what Shapiro calls “the structure of decision rules” to “follow the contours of power relations, not those of memberships” (1999, 37-38) – where “power relations” comprise a fundamental feature of affected interests. Thus, the self-appointed representative does not simply include individuals in an already-existing group, such as an electorate; rather, the representative claims of the selfappointed representative have the capacities to constitute alternative, issue-defined, serial, overlapping peoples. The key to these capacities is that the self-appointed representative defines a people in relation to an issue through a representative claim. If the claim has merit, it organizes that collection of people with respect to some issue that they have in common. Once the claim is made, the affected may coalesce as a “people”; that is, it is through the naming and narrating of problems, feelings, and silent experiences of injustice, that individuals may become aware of themselves as a people. One reason the poor and powerless face such obstacles in making themselves heard is because they lack knowledge even about themselves, about the group they belong to. For example, the millions who carry the AIDS virus in Africa are each suffering singly; they have no way of saying “We, the 30 million AIDS victims…” because most of them don’t know it’s a scourge of mammoth proportions striking many countries, and consequently they have no way to demand care, medicine and reparations…Being able to say “we” reflects genuine power, and “we” in the global justice movement should respect it, use it and try to extend it to others. (George 2004, 187-88) 113  To put a finer point on it, the constitutive element of self-appointed representation often transforms a latent constituency into a self-conscious and effective one. Put yet another way, there is a relationship between a projected/imagined people and the claim of the self-appointed representative that, if authorized, then becomes an alternative people or public. This is a fundamentally different way of thinking about membership in a democracy. The potential of the self-appointed representative to create a constituency offers, potentially, one kind of answer to the challenge of how to constitute the demos in a democratic way. Whelan, like most democratic theorists, argues that “democracy, which is a method for group decision making or self-governance, cannot be brought to bear on the logically prior matter of the constitution of the group, the existence of which it presupposes” (1983, 40). In the context of the self-appointed representative, however, it is theoretically possible for individuals to decide whether or not they want to be part of this and/or that people – and not only once, at the time of election, but serially and reflexively, on an ongoing basis. That is, when individuals authorize or reject putative representatives who have appointed themselves for the role of representative, they constitutive themselves as a demos in relation to the representative claim, thus choosing the boundaries of their own demos. This conceptualization of representation posits democracy as an ideal of selfdetermination in which citizens self-constitute. Importantly, it is not only the citizens of a particular nation-state who can authorize, or not, the claims of a self-appointed representative. In part, because self-appointed representation is a form of non-electoral representation that functions primarily in civil society and the public sphere, it can span, supplement, deepen, and resist formal political authority, whether democratic or non-democratic. And it can do so locally, nationally, or globally, in a way that is fluidly and flexibly responsive to peoples constituted by multiple and overlapping forms of affectedness.  114  Thus, self-appointed representation complicates the issue of where democratic representation occurs – it does not occur solely in the formal realm of electoral representation – and contributes to a profusion of representative entities. In theory, it need not compete with electoral representation; that is, self-appointed representation does not, in principle, undermine the representation provided by the state (if any). Rather, the particularity or issue-specificity of the self-appointed representative can provide representation in dimensions that would otherwise be ignored because they fall outside of formal, electoral constituencies, or because they are subject to oppressive conditions of power. Self-appointed representatives do not exist to represent the complete set of interests of an entire electoral constituency; they do not compete for votes and can therefore afford to be more overtly single-minded and agile than a state or political party, partial to particular peoples and particular issues. I may, for example, vote for a particular party or representative because I feel that the platform best approximates my concerns with education and health care. Because of the claims of self-appointed representatives, however, I may become both aware of and concerned by the destruction of ancient forests both in my own country and elsewhere. My chosen electoral representative may not (or cannot) address that concern, tied as she is to a party platform, or having to trade off attention to multiple issues. The existence of self-appointed representatives expands my representative options. As a result, I may choose to donate time and/or money to Greenpeace, which includes me in a new constituency, and makes me a part of a different, issue-defined “people.” The self-appointed representative thus identifies an affected constituency, which may have the potential to transform the representative claim into a relationship of democratic representation by retroactively authorizing the claim and holding the representative claim maker accountable for representation. I investigate these features of self-appointed representation in the next chapter. Here, however, I want to emphasize the prior moment in democratic representation: there can be no demos without representation, and the constitution of an 115  appropriate demos – an affected demos – is a condition for democracy. As Laclau nicely states, “The represented depends on the representative for the constitution of his or her own identity” (Laclau 2007, 158). I am also suggesting that this very identity constitution identifies, potentially, an appropriate demos while simultaneously empowering individuals to accept or reject any particular version of the demos. The self-appointed representative offers a claim that constitutes the constituency, and the constituency then decides whether or not to authorize that claim. In this way, the self-appointed representative serves to make effective the Kantian idea that “to be an autonomous agent is to act on reasons you give yourself” (LaVaque-Manty 2006, 365). Democracies can and should include a capacity for society to represent itself and to generate vanguard publics, which may very likely become the precursors for institutional innovations or reforms with democratic effects. Institutional conservatism is sometimes key for constituency stability but at the cost of slow or non-existent responsiveness. When the electorate and the politician are institutionally conservative and reactive, potentially, self-appointed representatives can balance these elements of formal democratic systems, while filling in where democratic systems do reach, providing highly specific, fluid, and forward-looking kinds of responsiveness.  Non-Democratic Constituency Formation All this said, self-appointed representation does not necessarily constitute democratic constituencies. In this section, I theorize several alternative kinds of constituency formation, which I shall call surrogate constituencies, skewed constituencies, and failed constituencies. I return to these categories in Chapter Six, completing them in the form of a typology of kinds of self-appointed representation, only some of which prove to be “democratic.”  116  A surrogate constituency is constituted when the self-appointed representative is responsive to a constituency of affected interests but receives authorization from a sympathetic, proxy constituency. In this way, “the people” may include not only the affected but also the concerned: those who have mobilized around the representative claim, who have been constituted, therefore, as part of the people, but who are not themselves affected. For example, Bono’s claims constitute two constituencies. Africans negatively affected by AIDS and policy around debt and trade are constituted as the affected constituency. Peoples in the first world, whose interests are not affected but who are concerned for the affected constituency, are constructed into a surrogate constituency by Bono’s claim that we in the developed world have a responsibility to act. Due to the authority provided by the audience (leaders of the G8 countries to whom he appeals for medication, debt reduction, and policy change), Bono represents those Africans who are negatively affected by AIDS, debt and trade. In addition to the authority granted to Bono by the audience, his claim has also been authorized by those who are concerned about living in a world where 300,000 Africans die every month from poverty and disease – a surrogate constituency that is constituted as a demos when it authorizes Bono’s claim. Because of the authority provided to him by both the audience and the surrogate constituency, Bono represents poor Africans, even though they have not authorized him. However, this is a constitutively legitimate representation, because it is responsive to the affected constituency. It is not, however, democratically legitimate because it does not involve the empowered inclusion of the affected constituency, but rather of the surrogate constituency. Ideally, although not necessarily, surrogate representation is a category of self-appointed representation that anticipates a moment when the affected will be active and agentic – a hopefully pre-democratic form of representation in the sense that surrogate representation may help to enable democratic constituency formation. Although the surrogate self-appointed representative does not enable self-determination of the affected, that is, participation in political 117  processes (Young 2000, 32), he does not interfere with the possibility that this might be achieved, and may help to make it possible. The act of the representation claim anticipates a moment when the affected constituency may speak for themselves, and so underwrites democracy or democratization. The surrogate actor may begin a process of constituency formation which in its first moments may be proxy, or anticipatory.8 Bono does respond to selfdevelopment by encouraging the reduction in price of medication for AIDS/HIV, so that an HIV diagnosis is no longer a death sentence. The surrogate self-appointed representative, then, might still achieve an important good for the affected constituency – there is a constitutive legitimacy to that representation, but it is not (yet) democratic legitimacy. In contrast, what I call a skewed constituency is formed when the self-appointed representative is responsive to interests of the kind that, if their power were to be enhanced through representation, would result in a less equal distribution of the conditions for selfdevelopment and self-determination of those affected. A constituency is “skewed” when the interests of those affected are not held in view, and so serves to undermine the affected constituency’s conditions of self-development and/or self-determination. An example of a skewed constituency is the Moral Majority, who authorized Jerry Falwell’s claim to exclude gays and lesbians by encouraging states to prevent recognition and acceptance of homosexuality. Once Falwell offered this claim, two constituencies can be theorized – those who authorize the claim (the Moral Majority), and those affected by the claim (lesbian, gay, 8  If/when the constituency does begin to make demands, and the surrogate claim maker  promotes the empowered inclusion of the affected, there is an evolution to a democratic author rather than one claiming to act for the public good. Conversely, if the surrogate representative were to interfere with their self-determination in this way, then there is a devolution to a failed representative.  118  bisexual, and transgendered peoples). With respect to their claim about homosexuality, the Moral Majority can be recognized as an authorized relationship of representation between it and Jerry Falwell; the existential aspect of the claim (that Jerry Falwell is the representative of the Moral Majority) is made possible by their authorization. The constitutive legitimacy of this claim, however, can be challenged on the basis of the exclusion of those affected by the performative aspect of this claim (gays and lesbians). Now, every claim of representation will likely result in some degree of exclusion of another group or groups, and the Moral Majority may also claim that their interests are affected if, for example, the meaning of marriage is expanded. However, the claims of the Moral Majority are relatively weak when compared to the impact that the lack of civil rights would have on the self-development and self-determination of LGBT peoples. As such, there are non-democratic consequences to this type of self-appointed representation. A failed constituency may be recognized when the constituency fails to recognize shared interests and a chance to have them represented. For example, David Suzuki, a self-appointed representative of environmental issues, offers many warnings about global warming, constituting affected constituencies with respect to environmental disaster and degradation. For the most part, it seems constituents have failed to recognize their interests with respect to this issue, even though it will affect conditions of self-development and self-determination, and so the constituency fails. It is also the case that a constituency may refuse the claim of representation altogether. For example, in the language I use here, we could say that “third wave” feminists refused the representative claim of “second wave” feminists. The women’s movement aimed to eliminate sexism for the benefit of all women, but third wave feminists argued that the movement was unresponsive to issues that affected women of colour, for example. In this case, the refusal of the representative claim made by second wave feminists constituted third wave feminists as an affected constituency with respect to issues that are 119  particularly harmful to women of colour, women of low income, and lesbians, that is, women who suffer from inequality in more than one dimension. In short, it is not only the authorization of a representative claim but also the rejection of the claim that constitutes a constituency. Labelling the kinds of constituencies that are formed as a result of the claims of those who represent by self-appointment as affected, surrogate, skewed, or failed, helps to clarify the activity of self-appointed representatives – what it is they are doing when they claim to represent others, and the kinds of relationships that are formed. Conceptualizing these relationships in this way helps to determine who is being represented in this non-electoral realm and for what purposes.  Conclusion The phenomenon of the self-appointed representative is intrinsically related to the generic problem of constituency in democratic theory. It turns out that self-appointed representation provides an answer to the question of how to democratically constitute the demos, or demoi, as it were, by forming constituencies on the basis of the affected interests principle. When the affected constituents can authorize their claims, or not, thus choosing the boundaries of their own demos, a democratic constituency is formed. This conceptualization of constituency formation posits each individual as potentially involved in a variety of “citizenships,” depending upon the relationships of “affectedness” that are salient to him/her. Understanding the concept of constituency in this fluid way is a necessary step in developing a democratic theory that is appropriate to the complex, globalizing, pluralistic, and highly differentiated societies within which we now live. In fact, self-appointed representation enables self-determination of constituencies, whether or not the constituency is based on the affected interest principle, when they are able to  120  authorize the self-appointed representative. Even skewed and surrogate constituencies are able to participate in their own self-determination by authorizing or rejecting the claims of the selfappointed representative, as well as holding them accountable for their claims and actions. So the question for the next chapter is: what might non-electoral authorization and accountability look like in the context of the self-appointed representative?  121  Chapter Five: ‘Who Elected Oxfam?’ Non-Electoral Authorization and Accountability According to the standard account of representation, elections function both to authorize representatives and to hold them to account. Winning an election authorizes actors to represent the constituencies that elect them. Subsequent elections support accountability: anticipating reelection, representatives provide accounts of their actions to their constituents, who then either re-elect representatives to reward their behaviour, or remove them from office to sanction their behaviour. In contrast, self-appointed representatives are not subject to formalized electoral authorization and accountability: they propose themselves as representatives of constituencies which, for their part, neither authorize nor hold these kinds of representative claimants accountable through formal elections. Indeed, as I develop the concept here, the non-electoral context is a defining element of “self-appointed representation.” That self-appointed representatives are not subject to electoral authorization and accountability is commonly noted by critics of the most high-profile kinds of self-appointed representatives, non-governmental organizations (NGOs), which often claim to represent the poor, the hungry, sick, under-educated, or otherwise marginalized constituencies, though they also claim to represent many other interests as well. These critics conclude that, without elections, there is something deeply illegitimate with the representative status that NGOs claim for themselves. These critics note that NGOs “never have to face voters or bear any sort of accountability” (Rabkin 1999, 37), that “NGOs are not elected, not accountable to any body politic” (Rivken and Carey 2000/01, 37) and that “NGOs are not very often connected, in any direct way, to masses of ‘people’” (Anderson 2000, 117). As The Economist asked, The increasing clout of NGOs, respectable and not so respectable, raises an important question: who elected Oxfam…? ... They may claim to be acting in the interests of the 121  people – but then so do the objects of their criticism, governments and the despised international institutions. In the West, governments and their agencies are, in the end, accountable to voters. Who holds the activists accountable? (2000, 86) In principle, these criticisms apply to all kinds of self-appointed representatives – not just NGOs – and they do so precisely because electoral authorization and accountability are missing. This chapter seeks to answer the question “Who elected Oxfam?” by making two observations. First, while it is true that no one elects Oxfam, the claims of such self-appointed representatives are often subject to other kinds of authorization and accountability mechanisms – membership entrance and exit, and public agreement, as examples – which I shall discuss below. Further, these mechanisms of authorization and accountability should be appropriate with respect to the powers and functions of self-appointed representatives. Self-appointed representatives, such as NGOs, are primarily discursive and voice representatives, wielding powers that fall short of those exercised by states. As discussed in Chapter Three, discursive representatives are powerful and influential to the degree that they represent positions and ideas that would not otherwise be included (Keck 2003, 45) and they are legitimate according to my theory to the extent that they further the self-development and self-determination of those who have been dominated and/or oppressed. As voice representatives, these actors represent these positions and ideas with the authorization of constituencies mobilized by their claims. Nevertheless, even with such authorization, self-appointed representatives “may well not have a mandate to make binding agreements on behalf of [those] in whose name they speak” (Keck 2003, 51). In contrast, formally elected representatives have influence over the deployment of the state’s instruments of force, as well as the state’s capacities to involuntarily extract and deploy economic resources. As such, the thresholds of formal accountability are, and should be, different than those to which self-appointed representatives are subject. Second, I address the problem of the “who” in “Who elected Oxfam?” On whose 122  authority does an actor like Oxfam claim to represent those who are so poor they cannot feed themselves? I suggested in previous chapters that the authority to act as a representative might come from the relevant audience, as Rehfeld (2006) argues, but that the constitutive legitimacy of the representative comes from their responsiveness to an affected constituency. To be considered a democratically legitimate representative, the self-appointed representative must be responsive to and promote the empowered inclusion of an affected constituency. If it exists, empowered inclusion will be expressed in non-electoral mechanisms of authorization and accountability. The challenge I take up in this chapter is to identify the many kinds of authorization and accountability that may be at work outside of elections. There is, in other words, a more serious and complex answer to the question “Who elected Oxfam?” than The Economist’s rhetorical formulation suggests. In order to answer the question, I first outline the understanding of authorization and accountability within the standard account of representation and indicate its limitations as a tool for conceptualizing and judging self-appointed representatives. Second, I suggest that non-electoral mechanisms of authorization and accountability may perform as well as, and perhaps better than, elections on many measures, including legitimacy. Third, I conceptualize the sources and mechanisms of nonelectoral authorization, and discuss some differences between electoral authorization and selfappointed authorization. Fourth, I conceptualize sources of accountability and the role that claim making may play in securing accountability. I conclude by arguing that self-appointed representatives have the potential to provide democratic representation when affected constituencies are empowered to utilize non-electoral mechanisms of authorization and accountability.  123  Electoral Authorization and Accountability Standard accounts of political representation focus mainly on the formal procedures of authorization and accountability within nation-state-based representative democracies. Most theoretical accounts of these procedures suffer from the kinds of formalism to which Pitkin objects, on the grounds that they are purely procedural, and are unable to identify and assess the actual quality of representation (Pitkin 1967, 39). For those theories that follow Hobbes in defining a representative as the person who has been authorized, the fundamental quality of representation is “the delegation or granting of authority” (Tussman 1947, 117-118). According to this account, any activity that takes place “after the right kind of authorization and within its limits is by definition representing” (Pitkin 1967, 39). In response to authorization theories of representation, others argue that it is accountability that will ensure the responsiveness of the representative to his/her constituents (Pitkin 1967, 57). However, as Pitkin suggests, neither authorization theorists nor accountability theorists provide standards for assessing the quality of representation. While those who emphasize authorization see representation generated in a particular way, those who emphasize accountability see representation as terminated in a particular way such that a “representative who acted in a completely selfish and irresponsible manner could not be criticized as long as he let himself be removed from office at the end of his term” (Pitkin 1967, 58). In short, these theorists of representation as authorization or accountability largely ignore what the activity of representation involves. In an effort to focus on what occurs during representation, Pitkin defines political representation as “acting in the interests of the represented, in a manner responsive to them” (Pitkin 1967, 209). Importantly, Pitkin’s formulations of representation as responsiveness are generic. As Saward notes, “Hanna Pitkin’s preferred definition of representation—as a ‘substantive acting for others’—does not in principle require election” (2008, 4), suggesting that  124  there are mechanisms of responsiveness not exhausted by electoral mechanisms. Pitkin does not pursue this implication of her work. Instead, she theorizes democratic representation only in terms of electoral representation. She suggests that electoral institutions are required to ensure that responsiveness to the constituency is systematic, and not subject to the caprice of the representative (Pitkin 1967, 239).  Beyond Electoral Authorization and Accountability This traditional understanding of electoral authorization and accountability and its importance for democratic representation leaves little room for conceptualizing or judging the democratic potentials of representatives outside of electoral processes. But as our political landscape has changed, post-national, transnational, and non-governmental actors, including the United Nations, the European Union, and civil society organizations, increasingly claim to act as representatives for citizens, and many probably do so in ways that are highly responsive. This changed landscape points to a key conceptual inadequacy in the standard account of democratic representation: the legitimacy of representatives is conflated with the fact of election (Rehfeld 2006, 3). Conflating legitimacy with elections is problematic not only for the purposes of conceptualizing and assessing representation beyond the state, but also for assessing representation at the nation-state level. According to the standard account of representation, elections are what make a representative democracy. That is, the “democratic” in “democratic representative” simply denotes the authorization of an existential claim – “I am the representative of this constituency” by virtue of a procedure – a fair election – rather than emphasizing the performative, or what occurs during representation. On this view, no matter  125  what the performance of the representative for the duration of her time in office, she may still claim to be a democratic representative because she was elected. Rather than fetishize electoral authorization and accountability, we should consider the possibility that non-electoral mechanisms of authorization and accountability may provide legitimacy, particularly if affected constituencies wield them. In one of the most important recent contributions to the literature on the accountability of actors at the global level, Ruth Grant and Robert Keohane (2005) offer some direction. They argue that by focusing accountability exclusively on elections, we narrow accountability in ways unsuitable for global politics. In the language I employ here, the electoral model of accountability fails to encompass affected constituencies, and fails quite obviously at the global level. Grant and Keohane offer seven non-electoral mechanisms of accountability, which operate at the global level. I survey these mechanisms below, but I also want to build upon this list, using it as a basis for a democratic theory of non-electoral relationship of representation by asking whether, and if so how, these mechanisms can be wielded not only by organized agents at the global level, but also by affected constituencies, as I have conceptualized them. Answers to these questions will allow us to assess whether or not non-electoral relationships of representation are democratically legitimate because they enable us to identify whether or not the affected constituency is the source of authorization and accountability. Grant and Keohane outline four mechanisms of non-electoral accountability (hierarchical, supervisory, fiscal, and legal) that are examples of a “delegation model of accountability,” and three (market, peer, and public reputational) that are examples of a “participation model of accountability.” Our two basic models differ fundamentally in their answer to the question: “Who is entitled to hold the powerful accountable?” In the participation model, the performance of power-wielders is evaluated by those who are affected by their actions. In the 126  delegation model, by contrast, performance is evaluated by those entrusting them with powers. (Grant and Keohane 2005, 31) More specifically, Grant and Keohane identify the following mechanisms. Leaders of organizations wield hierarchical accountability over their subordinate officials: subordinates can have their tasks constrained, and/or have financial compensation adjusted, or can be removed from office altogether. Hierarchical accountability “applies to relationships within organizations, including multilateral organizations such as the United Nations or the World Bank” (Grant and Keohane 2005, 36). Similarly, states wield supervisory accountability over multilateral organizations. Supervisory accountability “refers to relations between organizations where one organization acts as principal with respect to specified agents,” as when, for example, the World Bank is held accountable by states (Grant and Keohane 2005, 36). Funding agencies wield fiscal accountability over the agencies they fund. Fiscal accountability “describes mechanisms through which funding agencies can demand reports from, and ultimately sanction, agencies that are recipients of funding,” such as the United Nations (Grant and Keohane 2005, 36). And courts wield legal accountability over an individual official or agency. Legal accountability “refers to the requirement that agents abide by formal rules and be prepared to justify their actions in those terms, in courts or quasi- judicial arenas” (Grant and Keohane 2005, 36). These four mechanisms of accountability are examples of what Grant and Keohane call a “delegation model of accountability,” in which “performance is evaluated by those entrusting them with powers” (2005, 31). That is, these mechanisms are not primarily wielded by those affected. The remaining three mechanisms – market, peer, and reputational accountability – are considered by Grant and Keohane to be examples of a participation model of accountability, in that they “involve forms of participation, although the participants in each of these forms of accountability are different” (2005, 36). Stockholders or consumers wield market accountability 127  over firms or governments by, for example, withholding investment from countries with undesirable policies. Organizations wield peer accountability over their peers, as when NGOs, for example, evaluate their peers on issues such as quality of information and ease of cooperation. And public reputational accountability can be exercised by any one and every one: “Superiors, supervisory boards, courts, fiscal watchdogs, markets, and peers all take the reputations of agents into account” (Grant and Keohane 2005, 37). The participation model of accountability is intended to describe a situation in which “the performance of power-wielders is evaluated by those who are affected by their actions” (Grant and Keohane 2005, 31). But, for the purposes of a theory of representation, Grant and Keohane’s model is under-theorized. In particular, they fail to distinguish between the affected and those who are concerned on behalf of the affected. In their example, market accountability is exercised by consumers who “refuse to buy products from companies with bad reputations for labor standards or other practices,” suggesting that it is the consumers who are affected (Grant and Keohane 2005, 37). It is not, however, the consumers who are suffering under poor labour standards, but the employees (who may or may not also be consumers). Now it may be that the consumers are holding the company accountable on behalf of those affected, acting as a sympathetic, proxy constituency for the affected one (as I discussed in Chapter Four when conceptualizing surrogate constituencies). Certainly, these accountability mechanisms are useful for the purposes of restraining power-wielders at the global level, and I shall draw on them below. Grant and Keohane do not, however, show how they might provide a direct relationship of accountability between an affected constituency and a representative – admittedly, not their key purpose. For my purposes, Grant and Keohane still leave us with work to do if we are to assess how non-electoral relationships of representation are legitimate, in democratic terms or otherwise. Jane Mansbridge (2003) also offers some direction toward conceptualizing non-electoral 128  authorization and accountability in two ways. First, she describes forms of representation that fit within the electoral model of representation that are not usually conceptualized. One of these forms of representation is “surrogate representation,” which occurs when representatives act for citizens outside of their electoral district. The concept of surrogate representation suggests that electoral constituencies are not always appropriately formed for the purposes of addressing those (most) affected by decisions and actions. Mansbridge does not pursue this logic since her purpose is primarily to identify untheorized kinds of representative responsiveness within electoral democracy. But I shall, since my purpose in this chapter is to identify non-electoral mechanisms of authorization and accountability that may connect non-electoral representatives with non-electoral constituencies, and particularly affected constituencies. Second, Mansbridge recognizes that although the forms of representation she is describing do not meet the criteria for democratic accountability developed for the standard model of representation, each generates its own set of normative criteria by which it can be judged as legitimate. The standard criteria for democratic accountability, as Mansbridge constructs them, are the normative duties of the representative to keep promises made in the election, and the power of the voters to sanction that representative by voting him/her out of office at the next election (Mansbridge 2003, 516). Mansbridge calls this “promissory representation,” which entails the keeping of campaign promises (2003, 515). Promissory representation conceives of representation in its simplest form: the voter as principle, with preferences firmly established, exercises power over the representative as agent. Put another way, “By exacting a promise, the voter at Time 1 (the election) exercises power, or tries to exercise power, over the representative at Time 2 (the governing period): VT1 → RT2” (Mansbridge 2003, 516). Mansbridge endorses three forms of representation alternative to promissory representation that do not fit the criteria for democratic accountability attached to the standard model of representation. Anticipatory representation occurs when representatives try to please 129  future voters. This kind of representation “shifts normative scrutiny from the process of accountability to the quality of deliberation throughout the representative’s term in office” (Mansbridge 2003, 517). This shift in focus occurs because, unlike promissory representation, anticipatory representation assumes a voter with unstable preferences. If preferences were stable over time, then there would be no difference between the voter at T1 (the time of election), T2 (the period of service), and T3 (the next election) (Mansbridge 2003, 517; see also Miller and Stokes 1963; Nagel 1975). The instability of preferences poses an information problem, and so “prompts attention to public opinion polls, focus groups, and gossip about the ‘mood of the nation’” (Mansbridge 2003, 517). Moreover, anticipatory representation emphasizes underlying interests rather than present preferences, the former referring to fundamental needs and values rather than what is ephemeral or weakly held. Anticipatory representation encourages us to think of voters as educable and perhaps manipulable (Mansbridge 2003, 517-18). Gyroscopic representation “shifts normative scrutiny from traditional accountability to the quality of deliberation in the authorizing election” (Mansbridge 2003, 522). This shift occurs because “voters select representatives who can be expected to act in ways the voter approves without external incentives” (Mansbridge 2003, 520). As Mansbridge explains, gyroscopic representation does not emphasize representation as a substantive acting for others. Instead, the representative is selected on the basis of “deep predictability”; the voter assumes the representative’s future behaviour on the basis of past behaviour, as well as other cues such as descriptive characteristics (2003, 520-22). And surrogate representation, which occurs when representatives act for citizens outside of their electoral district, “shifts normative scrutiny from constituent-oriented accountability to systemic inequities in representation” (Mansbridge 2003, 525). Barney Frank is an example of a surrogate representative because he acts as a representative for gays and lesbians with whom he has no electoral relationship. Surrogate representation “plays the normatively critical role of 130  providing representation to voters who lose in their own district” (Mansbridge 2003, 523), and so is a form of non-territorial representation, capturing substantive interests that would otherwise go unrepresented because of the traditional emphasis on territorial representation and, in this case, because of homophobia or particular kinds of power differentials that result in the over-representation of some groups relative to others whose conditions of self-development and/or self-determination are (differentially or unequally) harmed. As Mansbridge explains, “The situation has changed from the time when territorial representation captured many of a voter’s most significant interests, but in the United States the representational system has not changed with it” (2003, 525). Mansbridge argues that these are all legitimate forms of representation, though they do not fit the criteria for democratic accountability developed for promissory representation. The appropriate normative criteria for judging these more recently identified forms of representation are systemic, in contrast to the dyadic criteria appropriate for promissory representation. The criteria are almost all deliberative rather than aggregative. And, in keeping with the conclusion that there is more than one way to be represented legitimately in a democracy, the criteria are plural rather than singular. (Mansbridge 2003, 515) Like Mansbridge, who recognizes that democratic representation is multiplicitous with each expression of it grounded in its own norms, I suggest that the authorization and accountability mechanisms of self-appointed representatives should be appropriate to their particular discursive powers and functions. In the next sections, I follow the strategy of Grant, Keohane, and Mansbridge to explore functional equivalents to electoral authorization and accountability. I suggest how these mechanisms might operate between the affected constituency and their representative: when such mechanisms function to promote the empowered inclusion of the affected constituency, they can in fact be democratic mechanisms of accountability. 131  Non-Electoral Sources of Authorization According to standard accounts of democratic representation, the self-appointed representative falls short of two conditions which are at the heart of democratic representation: authorization and accountability. In fact, as I argued in Chapters Three and Four, the lack of electoral authorization and accountability is both the defining feature of a self-appointed representative and also necessary to its potential democratic function. It is precisely because they are not restricted to the limits of electoral constituencies that they are able to represent affected constituencies. By distinguishing between the authority that a representative receives from a relevant audience and/or constituency and the legitimacy that a representative receives from his/her responsiveness to an affected constituency, we can begin to conceptualize some important differences between electoral authorization and democratic self-appointed authorization. First, there is a distinction to be made between the source of authority and the source of legitimacy of representation. I have suggested that authority is derived from a relevant audience (and sometimes, a constituency) – “the relevant parties before whom the representative claims to stand in for the represented and act as defined by the function of the representative” (Rehfeld 2006, 6) – and constitutive legitimacy is derived from responsiveness to an affected constituency. Electoral representation masks the difference between authority and constitutive legitimacy because it assumes the electorate as the affected constituency. In the context of selfappointed representatives, however, a constituency is not assumed, providing the space to conceptualize the possibility that when an audience authorizes a claim of representation, even if the performative element of the claim is antithetical to the interests of an affected constituency, that authorization provides the authority to act as a representative, though it is not constitutively  132  legitimate. In any case of representation, the representative’s claims to political authority depend upon the representative having been authorized to speak and act on behalf of a constituency. The authority of an elected representative is relatively clear: it follows from election to a representative office. The political authority of self-appointed representatives, however, is not pre-given, as it were, by institutional function. Rather, an audience may provide the selfappointed representative with political authority on the basis of whatever rules that audience uses to recognize a claimant as a representative (Rehfeld 2006). A constituency may also provide the self-appointed representative with political authority. In these cases, typically, political authority is earned. It is earned on the basis of the initial representative claim, which establishes a promissory moment of authority, and is offered to a proposed (or claimed) constituency. Once identified, authority is generated by the relationship with the constituency: if the claim is accepted, political authority is generated – in much the same way that winning an election generates authority for an elected representative. The claim of representation is made prior to authorization and is then affirmed or refused. That is, any claim of representation, whether formal, informal, or self-appointed, receives authorization retrospectively. Even formal representatives offer their claim – that they can and perhaps should be our representative – and then receive a vote, or not. Theorists of representation often refer to this as a prospective vote: authorization is made on the basis of promises made (Mansbridge 2003). A promise refers to an expected future event and so is prospective in that sense: the promise has not yet been fulfilled. Yet, when one considers representation as a process of claim making, the standpoint changes: the claim is made and is then affirmed or refused, and so receives authorization, or not, retrospective to the claim making. We can thus think of political authority as generated, at least in principle, within the representative relationship between the representative claim maker and the people, and it is held on behalf of the people by the representative. However, such political 133  authority is only constitutively legitimate when the claim is responsive to an affected constituency. Bono derives his authority to act as a representative from leaders of the G8 countries who are willing to meet with him and discuss these issues (the audience) and also from those concerned peoples in the developed world who support his claim (the surrogate constituency). His political authority is constitutively legitimate because his claims are responsive to an affected constituency. But his political authority is not democratically legitimate because it is not derived from the affected constituency. As we shall see in Chapter Six, there is not, in Bono’s case, a clear relationship of authorization by, and accountability to, the affected, which suggests that the empowered constituency is a surrogate one. That is, democratic political authority should be conceived as generated in the representative relationship between the representative claim maker and the affected constituency (responsiveness), with the affected constituency empowered to employ mechanisms of authorization and accountability (empowered inclusion). Put another way, the authority to act as a representative is derived from the existential element of the representative claim, and the (constitutive) legitimacy to act as a representative is derived from the responsiveness of the performative element of the claim to an affected constituency. I suggested in Chapter Three that the representative claim is not only an existential statement (e.g. “I, Jerry Falwell, represent the Moral Majority), but also a performative one (e.g. “We oppose state recognition and acceptance of gays and lesbians”). When a relevant audience (state leaders) and/or constituency (Moral Majority) authorizes the existential claim, even if the content of the performative claim is antithetical to the interests of the affected, that authorization provides the authority to act as a representative. However, as I have been arguing, the authority received from an audience and/or constituency is not legitimate unless it is responsive to an affected constituency. When a relevant audience and/or constituency authorizes the existential claim, and the performative claim is in the interests of the affected, that is a constitutively 134  legitimate relationship of representation. This suggests that Bono’s activities as a representative may be constitutively legitimate on the basis of their responsiveness to an affected constituency, though they are not democratic because the affected constituency is not empowered. Second, there are at least two possible sources for legitimate authorization of selfappointed representatives: one from the surrogate constituency and another from the affected constituency. It is the latter that is democratically legitimate. When a constituency (not the affected) authorizes a claim of representation, and the performative element of the claim is in the interests of an affected constituency, that authorization supports a constitutively legitimate, though not democratic, relationship of representation. In these kinds of cases, we have a responsive, but surrogate form of representation. Of course, the problem of judging responsiveness to the interests of the affected in these kinds of cases is difficult because, unlike democratic representation, this form of representation cannot rely on the judgments by the affected of their own interests. This is a danger with all surrogate forms of representation (see Alcoff 1991), as evidenced by the case of Bono. At the same time, the example of Bono is revealing: his claims on behalf of poor Africans enters into common discourse as well as helps to constitute a constituency of the affected. The former effect helps to generate discursive accountability; the latter effect helps to create the conditions under which the affected might begin to speak for themselves, authorizing, perhaps, Bono to speak on their behalf, and so provides constitutive legitimacy. The authorization processes in the case of self-appointed representatives are typically ambiguous because they lack formalized processes. But they are also highly substantive, and hence potentially more responsive than formal representation. Indeed, it is not the responsiveness of electoral representation that speaks in its favour, but its lack of ambiguity. By virtue of the process of election, elected representatives – no matter their performance in office – can always claim to be democratic representatives. They are elected and, unless term-limited, 135  they are subject to re-election; this has been considered sufficient criteria to be considered authorized as a democratic representative. Because self-appointed representatives are not elected, their democratic legitimacy is, relatively speaking, more difficult to achieve, but once earned, is more closely connected to their substantive performance. That is because a selfappointed representative requires not only the existential element but also the performative element of the claim to be accepted by an affected constituency to be considered a democratic representative. Third, and following from these points, electoral and self-appointed representations’ modes of authorization differ. Once the affected constituency is identified with the help of the criteria of self-development and self-determination (those whose interests are or would be institutionally oppressed and/or dominated), as discussed in Chapter Four, the norm of democracy requires that this constituency have the capacity to influence those decisions that (potentially) affect them. In formal electoral politics, that participation is facilitated through voting. If self-appointed representatives are to contribute to democracy, we should be able to conceptualize and identify other mechanisms that allow for the empowered inclusion of the affected. More specifically, if a self-appointed representative is to fulfil democratic functions with respect to constituencies defined by the affected interests principle, we need to identify mechanisms of authorization and accountability that are not exhausted by electoral mechanisms. While the self-appointed representative does not receive electoral authorization, other mechanisms determine their constitutive or democratic legitimacy. Operating outside of the formal authorization granted to elected representatives by the electorate, the kinds of authorization granted to self-appointed representatives fall into two categories. One is discursive authorization, which connects the self-appointed representative to the public opinion formed in the public sphere. Discursive authorization might take the form of public agreement expressed by affected constituencies, as well as the peers of the self-appointed representative and the 136  media. The other kind of authorization granted to a self-appointed representative is organizational, which might take one or more of the following forms: (1) the self-appointed representative acquires a following or membership based on a convergence of interests as expressed in a mission statement; (2) the self-appointed representative receives financial contributions from donors to pursue their work; and/or (3) voting within organizational selfappointed representatives can act as signals. For example, boards of trustees can vote on issues of policy such as grant-making, geographic focus, spending, investment, and management; and members can sometimes elect representatives to the Board of Directors. To step back, then, I am suggesting that there are three possible classes of authorization: one, of course, the formal authorization granted to elected representatives by the people as traditionally conceived (the electorate); two, discursive authorization granted to formal representatives, informal representatives, and self-appointed representatives by an audience and/or constituency; and three, organizational authorization granted to formal representatives, informal representatives, and self-appointed representatives by the audience and/or constituency. The medium of formal authorization is the vote, the medium of discursive authorization is public agreement (communicated through discursive or deliberative means, such as letter-writing, dialogue, etc.), and the medium of organizational authorization may be membership, voting within organizations, and/or donations. With these types and their media distinguished, we can now see that elected representatives draw on formal, discursive, and organizational forms of authorization. In addition to vote-based authorization, discursive authorization is received through approval ratings and opinion polls: they attempt to (accurately) measure public support for a candidate or an incumbent, and gauge (and perhaps influence) public opinion. Donations are an example of organizational authorization. Barack Obama’s fund-raising record was unrivalled and perhaps served as an indicator for him that he was doing something right. (It was noted in the media that 137  donations to Obama’s campaign spiked the day after Governor Sarah Palin of Alaska was named to the Republican ticket). Discourse-based and organization-based authorizations operate on different planes than the power-based authorization of the vote; but once representation is viewed as a generic set of relationships, we can see that even elected representatives partake of non-electoral forms of authorization. And the self-appointed representative leverages these two less powerful authorization mechanisms, for less powerful, though often highly influential, purposes. A fourth and final difference between elected authorization and self-appointed authorization is that an elected representative is primarily authorized at the time of election. In contrast, the self-appointed representative is typically subject to a process of authorization that is ongoing and recursive (Urbinati 2006a, 53).9 Public agreement can quickly and easily become public disagreement, and donations and memberships can be renewed – or not – yearly or even monthly. The representative voice of the self-appointed representative grows with these serial and incremental authorizations. As Nadia Urbinati argues, Authorization is the key to a democratic view of representation provided we do not reduce it to an isolated act (the election) as liberal-elitists or Schumpeterians do, but see it as “a relationship between the constituency and the representative” that creates us (our constituency) as a political unity. Representation has inspired the democratic turn that can be described as the process of authorization versus authorization as a simple act of decision. (2006a: 53-54) 9  Though formally elected representatives, particularly in the United States, are often subject to  various forms of discursive and organizational authorization, which can be ongoing, I think it is true that self-appointed representatives usually have to engage even more in a series of authorizations. 138  The self-appointed representative achieves discursive and/or organizational authorization from their constituency as a result of the persuasiveness of the representation claim. If a constituency is persuaded by the representation claim, and continues to be persuaded, that constituency will repeatedly authorize the claims of the self-appointed representative by, for example, renewing their membership fees.10 But the constitutive legitimacy of the claim is intimately connected to its responsiveness to the affected constituency; that is, the performative act of the representation claim must demonstrate a primary concern for the interests of those affected. And democratic legitimacy requires that it is the affected constituency who is empowered to authorize their representatives.  Sources of Accountability Democratic representation requires not only authorization by, but also accountability to, an affected constituency. “A crucial feature of representative democracy,” argue Grant and Keohane, “is that those who govern are held accountable to the governed” (2005, 29). In what sense are self-appointed representatives accountable? In the literature, this topic has been addressed in the context of non-governmental organizations. It is frequently noted that NGOs lack the formal accountability mechanisms of re-election or removal from office and yet continue to serve as representatives of the poor and the marginalized. Unless they are termlimited, constituents can hold representatives accountable by the possibility of re-election. Representatives are required to provide an account of their actions to their constituents, who  10  In this way, the processes of authorization and accountability are intimately connected; as  Keohane suggests, “accountability relationships are created by acts of authorization” (2002, 9). Consent must continually be tested and reaffirmed. 139  either re-elect representatives to reward their actions, or remove representatives from office to sanction them. So while it is true that NGOs are not elected, it does not necessarily follow that they are not, or cannot be, held accountable for their representative claims by their constituents. They will be held accountable in a manner different than that applied to states and legislators, and for different powers and purposes. Self-appointed representatives, such as NGOs, primarily have discursive power – the power to make audible those voices that have been excluded or marginalized. This form of power is primarily based in persuasion. Self-appointed representatives do not have power over anyone, in the way that our political leaders have influence over the deployment of state-based coercive power (Wrong 2002). As such, the thresholds of accountability for the self-appointed representative will be, and should be different than for those representatives who have coercive power, that is, control over the instruments of force. If a representative can, for example, send people to war, citizens require empowerments that are institutionalized: not just votes, but also a package of rules and rights. As Grant and Keohane explain, NGOs largely depend on reputation and funding, and lack the coercive force and large material resources over which formal representatives have influence. As such, the decisions and actions of NGOs are less likely to lead to serious abuses of power (Grant and Keohane 2005, 38). Self-appointed representatives operate mostly within civil society and the public sphere, and lack the decision-making powers of legislators and administrators. Thus, elections are not required in the sense that the potential dangers of state power in relation to which elections provide a protective function do not exist. But appropriate, non-electoral accountability mechanisms are required, because self-appointed representatives claim legitimacy, and exercise influence when they are able to generate legitimacy. It is also the case that, unlike state powers, discursive power is not subject to monopoly control. Self-appointed representatives do not have monopolies over certain publics or issues: 140  more than one actor can represent the same constituency, interest, or issue. Were monopoly powers to exist, constituencies would lack choice. It is because constituents can typically choose among representative claims that they are able to induce responsiveness and accountability through their choices. Indeed, this structural fact about self-appointed representatives suggests that at least one important form of accountability would be market-based, in the sense that accountability to constituencies is increased when constituents can choose from among several “suppliers” of representative claims. Finally, in contrast to state-based powers, discursive power works only through agreement or consent. Because a self-appointed representative makes claims for which she seeks to gain authorization, she relies on the response of the public to her claims. Implied in the concept of the representative claim, then, is the role of publicity: the decisions and actions of the representative must be made public in order to be judged. Publicity not only enables authorization – its acceptance by those to whom the claim is addressed – but also implies accountability – that the claimant should discursively justify the claim if/when it is challenged. In other words, inherent in the concept of the representative claim is a supposition of accountability in the sense of giving an account to those the representative claims to represent. The deliberative element of accountability requires that the representative explain and justify their behaviour to others: “By whatever means, accountability is a form of public reasoning, the presupposition of which is that when accountability is discharged there is an intelligible connection between the action for which an account is being rendered and the putative reasons that are offered to explain and justify that course of action” (Savage and Weal 2009, 69). Being held accountable by others involves both deliberation (to give an account, to explain and justify one’s behaviour to others), and the possibility of control (to be held accountable, to be sanctioned in some way) (Savage and Weal 2009, 69). As Dunsire argues, “It is the coupling of information with its evaluation and application of sanctions that gives 141  ‘accountability’ or ‘answerability’ or ‘responsibility’ their full sense in ordinary usage” (1978, 41). Traditionally, such public reasoning occurs between elected representatives and their constituents, in which they “seek to justify to one another their preferences for alternative forms of public policy” (Weale and Savage 2009, 70). This simple registering of preferences, however, does not fully capture the creative elements of the discursive activity of the self-appointed representative. Because constituency formation is reflexive, so too is accountability. Through public reasoning, justification, and explanation, affected constituencies may be formed, transforming latent constituencies into selfconscious and effective ones. Thus, when considered as a part of constituency formation, discursive accountability is in itself constitutive. As Ranson puts it, “Accountability in this view, as discursive reason, is the very expression rather than denial of our reflective agency” (2003, 461). Accountability is not simply or primarily “a summons of compliance but rather provides understanding, as Shotter (1989) argues, of how we constitute the sense we have of ourselves (our identities) as well as shared ways of constructing the meanings that inform our social orders” (Ranson 2003, 461). Viewed through the lens of representation, discursive accountability is constitutive, the medium through which latent constituencies become – to borrow loosely from Marx – constituencies in and for themselves. What, then, might accountability look like in the realm of self-appointed representation? Although authorization and accountability are conceptually distinct, a particular mechanism sometimes serves both. The vote, for example, is both a mechanism of authorization (election) and a means of accountability (re-election or removal from office). Similarly, membership will serve both authorization and accountability: through joining (authorization) and through exit or anticipation of exit (accountability). Summarizing and elaborating these points from a schematic perspective, we can now identify at least eight distinct accountability mechanisms that might operate, singly or in 142  combination, in the domain of self-appointed representatives, and by whom they can be wielded: •  Voice  •  Organizational exit  •  Discursive exit  •  Public reputational accountability  •  Peer to peer accountability  •  Hierarchical accountability  •  Fiscal accountability  •  Legal accountability Following Hirschman, voice is “any attempt at all to change, rather than to escape from,  an objectionable state of affairs” (1970, 30). According to Hirschman, a firm’s customers or an organization’s members employ voice accountability when they express their dissatisfaction to those in charge who, in turn, “search for the causes and possible cures of customers’ and members’ dissatisfaction” (1970, 4).11  11  Interestingly, voice accountability may be considered its own kind of representation:  companies operate under the assumption that every one customer who voices a grievance represents ten people who have the same grievance, but have not taken the time to lodge a complaint. NGOs operate by the same principle: when one person signs a petition for Amnesty International, for example, both Amnesty and the government operate under the assumption that every one petitioner stands for ten others. This may be a different phenomenon than that discussed in this dissertation. I am specifically addressing representation with a claim, but the example of the petitioners may be representation without a claim. 143  The mechanism of voice accountability can also be used more generally. An affected constituency, for example, may also exercise voice accountability, even if they are not members of an organization. The poor and the marginalized may be unable to pay the dues required of member organizations, and may be unable to escape from an objectionable state of affairs (which is likely why they require representation). They can often, however, utilize the voice mechanism, that is, articulate their opinion in an attempt to change their circumstance, when a self-appointed representative actually speaks to those people the self-appointed representative hopes to affect – and of course, not only speaks to them, but listens them and responds to them. For example, an organization called Fancy Stitch – a partner organization supported by Oxfam Australia – supports women living in the Ingwavuma area of KwaZulu-Natal, South Africa, a community devastated by HIV and poverty. This organization brings together women in this area to make handicrafts, artworks, and textiles, providing them with an opportunity to both share and express their stories and also earn them money through the sale of their work. These women are not members of Oxfam, but they utilize voice accountability with the founder of Fancy Stitch, Maryna Hess, letting her know, and through her, Oxfam, that their lives have improved as a result of joining Fancy Stitch, allowing them to build their own homes and support their families (Oxfam 2010). Exit accountability is a market-based consequence of joining and leaving organizations. The self-appointed representative is made aware of the approval of his/her actions/goals in terms of membership levels and/or donations: high approval is expressed via a stable and perhaps growing membership and/or donations, and low approval is expressed by the exit of members from the organization, or the withdrawal of donations. For example, Oxfam UK reported in 2001 that it had a half million regular givers, when eight years prior it had only 70,000 regular givers (Oxfam UK 2010). Some will argue that “exit” is less democratic than “voice,” but the two are often structurally related; as Hirschman argues, “the effectiveness of the voice 144  mechanism is strengthened by the possibility of exit” (1970, 83). This point is especially important for, say, NGOs, since they occupy a rather competitive market for both funding and members. Such organizations tend to be attentive to its members because of the threat of exit. Exit accountability has three more desirable attributes from the standpoint of constituents: first, the cost of exit is low. Second, entrance and exit is voluntary, unlike, say, citizenship. And third, provided that there are other choices available, exit can occur at any time the member chooses, unlike periodic elections. In addition to this traditional understanding of exit within organizations, discursive exit occurs when a member, or members, of the affected constituency refuses the grounds of the representation claim. Unlike voice, in which one articulates their opinion in an attempt to change their circumstance, discursive exit is a refusal of the representation claim altogether. If “women,” for example, do not accept Gloria Steinem as their representative – which they might express through public disagreement – the alternative to her representation claim may be the claim of another self-appointed representative, such as Anita Bryant. Similarly, some Black Americans exercised discursive exit when they refused Dr. King’s claims and instead supported the claims of Malcolm X (and vice versa). Discursive exit may communicate particular information, such as the reason for exit, and so has the potential to inform the broader public about the reasons for exit. Public reputational accountability overlaps with other forms of accountability, given that “superiors, supervisory boards, courts, fiscal watchdogs, markets, and peers all take the reputations of agents into account. Indeed, reputation is a form of ‘soft power,’ defined as ‘the ability to shape the preferences of others’” (Grant and Keohane 2005, 37, citing Nye 2004, 5). Self-appointed representatives will often make efforts to develop their reputations as representatives by investing in a cause, manifesting consistency of purpose, and attending to the continuity of their “brand” as standing for those affected by the cause. Al Gore, for example, has 145  established a reputation as an environmental activist and has arguably played an important role in the public’s emerging concern about global warming. He now stands for those concerned about global warming and has, presumably, an interest in monitoring the integrity of that representative relationship, which in turn provides a mechanism of accountability. Public reputational accountability is an especially important mechanism of accountability for an individual self-appointed representative, who is unlikely to be subject to other means of accountability, such as market or hierarchical accountability. As Grant and Keohane note, the “category of public reputational accountability is meant to apply to situations in which reputation, widely and publicly known, provides a mechanism for accountability even in the absence of other mechanisms as well as in conjunction with them” (2005, 37). Other mechanisms of accountability do not promote the empowered inclusion of the affected, but may contribute to the trustworthiness, transparency, or integrity of a self-appointed representative, which may in turn underwrite other, more democratic forms of accountability. For example, because partnerships and networking are critical to the functioning of NGOs, peer accountability can be vital for these types of organizations. Grant and Keohane point out that “Organizations that are poorly rated by their peers are likely to have difficulty in persuading them to cooperate and, therefore, to have trouble achieving their own purposes” (2005, 37). The American Institute of Philanthropy and Charity Navigator, as examples, serve the function of charity watchdogs, providing information on the financial efficiency, governance, and fundraising practices of charities, for the benefit of potential members and donors. Thus, peer accountability “arises as the result of mutual evaluation of organizations by their counterparts” (Grant and Keohane 2005, 37). For example, in 2003, Oxfam Belgium came under fire for producing posters that discouraged the purchase of fruits and vegetables grown in Israel, and so were considered anti-Israeli. Following some negative publicity and pressure from Israeli NGOs, Oxfam removed the poster from their website and its Chairman issued an apology (NGO 146  Monitor 2010). Peer accountability may also encourage professional codes of conduct, as evidenced by the signing of the International Non Governmental Organizations’ Accountability Charter, which Oxfam signed in 2006 (Oxfam 2010) Hierarchical accountability refers to relationships within organizations, where “superiors can remove subordinates from office, constrain their tasks and room for discretion, and adjust their financial compensation” (Grant and Keohane 2005, 36). Often, boards of trustees act in this regard – they may be invited in as outsiders, in a sense, to act as evaluators and reviewers, voting on policies, positions, and operational budgets, thereby adding a degree of accountability. Some organizations, such as Oxfam International, provide their members with the ability to elect representatives to their Boards of Directors, which adds an element of hierarchical accountability. Fiscal accountability “describes mechanisms through which funding agencies can demand reports from, and ultimately sanction, agencies that are recipients of funding” (Grant and Keohane 2005, 36). Finally, those entities that are neither states nor international organizations are subject to the laws of the states possessing jurisdiction (Grant and Keohane 2005, 36). Legal accountability, then, ensures the lawful behaviour of the self-appointed representative – funds will not be siphoned off, for example. This list of mechanisms is not necessarily exhaustive. But it does begin to identify the places in which we might look for mechanisms of non-electoral accountability that are appropriate with respect to the discursive power of self-appointed representatives (see table 5.1). Discursive and organizational mechanisms of authorization and accountability accurately reflect the kinds of political spaces that self-appointed representatives occupy and the kinds of powers they can deploy.  147  Table 5.1 – Electoral and Self-Appointed Authorization and Accountability  Electoral  Sources of Authorization  Mechanism of Authorization  Constituency  Mechanism of Accountability (Retrospective) Vote  Constituency  Formal  (Prospective) Vote  District-based  Electorate  Informal  Discursive Authorization  Affected, Electorate  (Retrospective) Vote  Electorate  Discursive Authorization  Voice (Public Agreement)  Affected, Audience  Voice (Public Disagreement)  Affected, Audience  Discursive Exit Public Reputational  Self-Appointed  Organizational Authorization  1. Membership  Affected, Audience  Exit  Affected, Audience  Voice (Public Disagreement) Public Reputational Peer 2. Donations  Audience  Fiscal  (Affected), Audience  Legal Public Reputational Peer 3. Votes  Audience (Board of Trustees)  Hierarchical  (Affected), Audience (Board of Trustees)  148  Conclusion Oxfam clams that it is “dedicated to fighting poverty and related injustice around the world” (Oxfam 2010). In this way, Oxfam claims to represent those subject to poverty and injustice. We can now see that The Economist’s challenge to this claim, “Who elected Oxfam” and “Who holds the activists accountable?” is entirely answerable. While Oxfam has not been elected, it does not necessarily follow that they have not been authorized, or that they cannot be held accountable. Oxfam receives organizational authorization in the form of donations and memberships, primarily from those relatively privileged in the developed world. Oxfam is also subject to discursive authorization and accountability, as exemplified by the challenge issued by The Economist. That said, it does not follow that Oxfam is democratically legitimate. The types of authorization and accountability to which Oxfam is subject does not produce democratic legitimacy if they are not received from those affected. As I shall argue in the next chapter, organizations such as Oxfam function as surrogate self-appointed representatives: although they do act in response to affected constituencies, they do so without their authorization and accountability. Instead, Oxfam relies on a sympathetic, proxy constituency – the relatively privileged – for their authorization and accountability. It is the relatively privileged in the developed world who have authorized Oxfam to pursue their work, and who can and do hold them accountable, via non-electoral mechanisms of authorization and accountability such as membership, donations, exit, and public reputation, as well as peer, hierarchical, and fiscal. As a consequence, Oxfam has constitutive legitimacy, of a kind related to the responsiveness of its mission, and as a consequence of the good that it does for its claimed constituencies. But for Oxfam to be considered a democratic representative, it would have to work directly with the  149  affected; instead, Oxfam works primarily with local organizations who themselves have these direct relationships of authorization and accountability with the affected. Perhaps Oxfam trades off democratic legitimacy for the purposes of efficiency and reach of assistance. Whatever Oxfam’s motivation, identifying the representative claim, the constituencies involved, and nonelectoral mechanisms of authorization and accountability, provide the criteria by which we can conceptualize and judge the legitimacy of such organizations. In the next chapter, I bring these considerations to bear on a typology of self-appointed representation, so we might see more precisely where the democratic legitimacy of self-appointed representation resides.  150  Chapter Six: Self-Appointed Representation in Practice As I have argued, in order to be considered a democratic representative, the selfappointed representative must be both responsive to and promote the empowered inclusion of an affected constituency by employing mechanisms of authorization and accountability, such as voice, exit, discursive exit, and public reputational accountability. I have reproduced the table first introduced in Chapter Three to remind us of how these dimensions work to produce four ideal-types of self-appointed representatives.  Table 6.1 – Dimensions of Self-Appointed Representation Empowered Inclusion Responsiveness to Affected Interests +  -  +  Failed Representation  Skewed Representation  Surrogate Representation  Democratic Representation  If the self-appointed representative acts on only the first of these dimensions, that is, if the selfappointed representative is responsive to, but does not promote the empowered inclusion of, an affected constituency, I refer to this kind of representation as surrogate. If the self-appointed is unresponsive to the affected constituency and promotes the empowered inclusion of another group with the effect of undermining the self-development and self-determination of the affected constituency, I refer to skewed representation. Finally, if the actions of a self-appointed representative fail on both dimensions of responsiveness and empowerment, that is, the selfappointed representative claims to represent an affected constituency but is neither responsive to  151  them nor promotes their empowered inclusion, I term this kind of representation as, simply, failed. In this chapter I illustrate these kinds of self-appointed representatives with a number of examples. The examples I offer here are illustrative rather than exhaustive. The aim of this chapter is to show that the theory I have developed allows us to not only identify instances of self-appointed representation, but also to critically assess their potential contributions to democracy. The judgments I offer here focus on the relationship between the representative claim maker and those affected by the issues identified by the claim in both dimensions indicated above: responsiveness to the affected, and empowered inclusion of the affected. The examples I discuss include a variety of self-appointed representatives: Dr. Martin Luther King, Jr., the Rwanda Women’s Network, the American Civil Liberties Union, Bono, Amnesty International, Human Rights Watch, Jerry Falwell, Anita Bryant, National Organization for Women (NOW), and the National Association for the Advancement of Colored People (NAACP). When we look at each example from the perspective of the relationships between the representative claim and the constituency of the affected, we will see that we can judge them to be democratic, surrogate, skewed, or failed. On the basis of these judgments, we can also judge their relative contributions to better and worse politics.  Democratic Self-Appointed Representatives Dr. Martin Luther King, Jr. is exemplary of the phenomenon of democratic selfappointed representation. Dr. King is a democratic self-appointed representative because he offers a claim of representation for the benefit of a constituency whose basic interests were affected by so-called “Jim Crow” laws that was then able to authorize those claims and hold him  152  accountable for his activities.12 That is, he demonstrates both responsiveness to, and the empowered inclusion of, an affected constituency. Dr. King worked for the civil rights of African-Americans in his capacities as a preacher, a member of the executive committee of the National Association for the Advancement of Coloured People (NAACP), and as the elected president of the Southern Christian Leadership Conference (SCLC), an organization formed for the purpose of providing leadership for the civil rights movement. Dr. King’s claims for the freedom and citizenship rights of African Americans are an example of responsiveness to a constituency whose self-development and self-determination were being inhibited. The Jim Crow laws that existed throughout the southern United States mandated racial segregation in all public facilities, including public schools, public transportation, restaurants, and restrooms. The laws were justified by a “separate but equal” doctrine, though, of course, accommodations for black Americans were not only inferior, but carried the stigma of inequality. Inspired by Mahatma Gandhi and through direct and non-violent means such as boycotts, marches, protests, and sit-ins – what he referred to as “massive non-cooperation” (King 1998) – Dr. King worked to overturn Jim Crow laws and to secure basic civil and political rights for African-Americans (and, by extension, for all Americans). King defined his constituency through his speeches and deeds. On the basis of this constituency definition, those affected by Jim Crow laws were able to then authorize his claims and hold him accountable, 12  I have referred to Dr. King as a self-appointed representative, even though he was the elected  president of the SCLC, because he made claims outside of formal representative institutions. Moreover, the authorization was not provided by an electoral constituency, as traditionally conceived. The distinction is important because African Americans form a constituency of affected interests that do not neatly fit with a particular district, and that are non-territorial. 153  contributing to their empowered inclusion. Thus, those citizens of Montgomery who participated in and supported the bus boycott discursively authorized Dr. King’s claim of representation through public evaluation that registered approval of his aim to end racial segregation in the public transit system. And they registered that approval by participating in the boycott every day for 381 days until the Supreme Court of the United States declared such segregation unconstitutional. Similarly, in St. Augustine, Dr. King and the SCLC helped to organize and participate in marches, sit-ins and other forms of peaceful protest from May until July of 1964, actions supported by hundreds of black and white civil-rights supporters. Discursive authorization also came from the 25,000 people who participated in the Selma to Montgomery march in 1965. In addition to these examples of discursive authorization, members of the SCLC organizationally authorized Dr. King its president, as they contributed their time and effort to the advancement of civil rights through this umbrella organization. Every local organization that decided to become an affiliate of the SCLC authorized its claims that “civil rights are essential to democracy, that segregation must end, and that all Black people should reject segregation absolutely and non-violently” (SCLC 2010). Through exercising choices around membership, African-Americans had opportunities to authorize Dr. King’s claims (as well as the claims of the SCLC, of which Dr. King would not always be president, and is itself another relationship of democratic self-appointed representation). They also enjoyed the opportunity to exit, if they so chose. Did the affected constituency also hold Dr. King accountable? Dr. King regularly preached sermons in the churches of his affected constituencies, making himself available to those constituents he was representing. They could thereby articulate their opinion in an attempt to change their circumstance, that is, utilize voice by expressing their dissatisfaction to those in charge who, in turn, “search for the causes and possible cures of customers’ and members’ 154  dissatisfaction” (Hirschman 1970, 4). His constituents could have exercised exit accountability by choosing to not participate in the boycotts, protests, and demonstrations that Dr. King planned and organized. In fact, his constituents had two types of exit available to them, as outlined in Chapter Five. First, the exit that makes a self-appointed representative aware of the approval of his actions/goals in terms of membership levels and/or donations: high approval is expressed via a stable and perhaps growing membership and/or donations to the SCLC, and low approval could be expressed if members choose to exit the SCLC, or withdraw donations, which would send a signal to Dr. King regarding his claims and actions. The second type of exit is discursive: when a member, or members, of the affected constituency refuses the grounds of the representation claim. In fact, many African Americans did refuse Dr. King’s claims and instead supported Malcolm X, who often disagreed with Dr. King’s methods. But the widespread participation in the boycotts, marches, protests, and demonstrations that Dr. King organized, demonstrates their approval of his work; if they did not approve, they could choose to not participate, and to publicly discuss why they are not participating (distinguishing discursive exit from apathy). Public reputational accountability also played a key role. In his speeches and writings, Dr. King justified his actions to the broader public, thereby developing reputational accountability. At the same time, Dr. King’s detractors tried to utilize public reputational accountability by, for example, suggesting he had ties to Communists. His reputation waxed and waned even among his supporters. In his autobiography, Dr. King suggests that he made a mistake in Albany, choosing to protest against segregation generally rather than concentrating upon a particular issue such as integrating buses or lunch counters. Because the aim of the protest was vague, nothing was achieved, which led to a decrease in morale and support (King 1998, chapter 16). Mindful of these failures, King altered the strategy for Birmingham. Birmingham was considered a success: King’s reputation soared, and the SCLC was much in 155  demand to effect change in many Southern cities. Dr. King’s reputation suffered as a result of the failure in Albany, acting as a mechanism of accountability. To sum up, Martin Luther King, Jr. is a democratic self-appointed representative because he offers a claim that demonstrates his responsiveness to an affected constituency, which promotes the interests, defined by conditions of self-development and self-determination, of black Americans, and which promotes their empowered inclusion by providing them with the opportunity to authorize his claims and hold him accountable. A second example of a democratic self-appointed representative is the Rwanda Women’s Network (RWN). This non-member organization works toward “improvement of the socio-economic welfare of women in Rwanda through enhancing their efforts to meet their basic needs” (Rwanda Women Network 2010). The RWN works with survivors of sexual and genderbased violence across Rwanda in the recognition that women and children bore the brunt of the genocide, and remain the most vulnerable and marginalised groups within Rwandan civil society. The organization’s core programmes include educating women on issues such as their human and legal rights and sexual gender-based violence, and empowering women with their families to improve their socioeconomic status. The RWN is responsive to an affected constituency (women and children who are survivors of sexual and gender-based violence) and promotes their empowered inclusion by working directly with the women they affect. These women discursively authorize the RWN by participating in the three core programs offered: the provision of health care and support, education and awareness programs on human rights and legal procedures, and socioeconomic empowerment (Rwanda Women Network 2010). The RWN is a non-member organization and so is primarily subject to voice accountability: Rwandan women can express their dissatisfaction to the RWN. The constituents also enjoy discursive exit. The women and children with whom they work can refuse the claim of representation made by the RWN by not participating in its clinics. Finally, the affected 156  constituency can employ public reputational accountability, which overlaps with the other mechanisms, and holds the RWN accountable through its desire to monitor the integrity of their representative relationship. A third example of a democratic self-appointed representative is the American Civil Liberties Union (ACLU), which works to “defend and preserve the individual rights and liberties that the Constitution and laws of the United States guarantee everyone in this country” (ACLU 2010). The ACLU is responsive to an affected constituency (those whose rights have been denied, including people of colour, women, LGBT peoples, prisoners, and people with disabilities), by working in courts, legislatures, and communities to preserve and protect rights and liberties. The ACLU promotes their empowered inclusion by providing opportunities for the affected to both organizationally (membership) and discursively (public agreement) authorize their activities. The affected constituency also enjoys four types of accountability. The affected constituency can exercise voice by contacting their local ACLU affiliate and voicing concerns with/about the organization. The affected constituency enjoys organizational exit, where the withdrawal of membership or the threat of withdrawal acts as a mechanism of accountability. The affected constituency can exercise discursive exit by refusing the grounds of the representation claim. Finally, the ACLU is subject to public reputational accountability through its desire to attend to the continuity of its “brand” as standing for those whose rights and liberties are threatened.  Surrogate Self-Appointed Representatives The well-known case I have used throughout this dissertation, the musician Bono, is a good example of a surrogate self-appointed representative. The story of Bono as a self-  157  appointed representative begins with Bob Geldof,13 an Irish singer, songwriter, and political activist, who was once involved in famine-relief efforts in Ethiopia. In 1984, Geldof helped found Band Aid, the charity super-group that recorded the single “Do They Know It’s Christmas?” to raise money for famine-relief. Bono lent his voice to the single and the following year, he participated in Geldof’s Live Aid rock concert, which raised money for Ethiopia. Bono spent six weeks in an orphanage in Wello, Ethiopia to see the effects of the African famine (Tyrangiel 2002). It was not until 1997 that Bono committed himself to issues of debt in Africa. Bono learned from a development advocate that although Live Aid raised $200 million, Ethiopia alone paid $500 million in annual debt service to the world’s lending institutions (Tyrangiel 2005). In response, Bono signed on as a spokesman for Jubilee 2000, a “drop the debt” movement. Jubilee’s aim was to erase the debt of the world’s poorest countries so that they may spend money on health care and education rather than paying down loans that had been incurred by earlier, and sometimes corrupt, governments (Tyrangiel 2005). Bono also started flying to Washington for weekends at the World Bank and learning about debt and trade policies from economist Jeffrey Sachs. He asked Bill Clinton’s administration to make debt relief a core aim of U.S. policy to which Clinton agreed, although Congress did not. In 2002, Bono cofounded the organization DATA (Debt, AIDS, Trade, Africa, and also Democracy, Accountability, Transparency, Africa). Through its auspices, Bono acts as a nonelectoral actor, connected primarily to the public sphere and civil society, largely in the developed world, seeking an associational basis for his claims. In 2004, DATA joined with ten other anti-poverty organizations to create a new campaign for the fight against extreme poverty and preventable global disease: ONE. According to the website, the intention was to combine 13  Geldof might also be considered a self-appointed representative. 158  “DATA’s high-level global advocacy and policy depth with ONE’s grassroots mobilization expertise. DATA was created to press the governments of developed nations to do their part in the fight against extreme poverty in Africa, with a focus on debt relief, AIDS treatment and prevention and reform of unfair trade rules. DATA also advocated for increased democracy, accountability and transparency in government so civil society in poor countries had a greater say in how those resources were deployed. ONE quickly got to work, via global online actions, development of an iconic ad campaign and support for the Live 8 concerts, to push for a new initiative to fight African poverty at the 2005 G8 Summit in Gleneagles. That push, in conjunction with the Global Call to Action Against Poverty, succeeded in helping to secure a pledge by the G8 to direct an additional $25 billion in effective assistance to Africa by 2010. In less than a year, ONE signed up more than 2 million members and created a powerful, grassroots political force in support of better policies for combating poverty. (ONE 2010) When Bono aims to achieve debt-relief for Africa from the leaders of the G8 countries, then the G8 leaders are established as the relevant audience – that is his claimed function and they are the relevant audience required in order to fulfil that function. In turn, these world leaders react to this claim, choosing to meet with Bono and listen to his arguments – even if they do not always agree. As such, they provide the authority for Bono’s claim as a representative. As stated in earlier chapters, a person is, in fact, a representative, if the relevant audience accepts that person as such. In addition to the audience that provides Bono with the authority to act as a representative, Bono’s claims clearly identify an affected constituency. His organization seeks “to press the governments of developed nations to do their part in the fight against extreme poverty in Africa, with a focus on debt relief, AIDS treatment and prevention and reform of 159  unfair trade rules” (ONE 2010). The affected constituency is the Africans suffering from debt, AIDS, and trade. Bono is being responsive to an affected constituency whose self-development is being directly undermined by poverty-related issues, and indirectly undermined by large amounts of debt, which necessitates the diversion of money to debt repayment rather than to education and health care.14 This constituency is also suffering from unfair trade rules such as high tariffs on African products, which make it difficult to compete in international markets. Bono’s claims also identify a surrogate constituency on behalf of an affected constituency, and it is the relationship to this constituency, rather than the affected constituency, where the representative relationships are most immediate. In an interview Bono gave to the London Evening Standard, it was suggested to him that he represents the 18-30 year olds who listen to his music. If you examine what Bono’s constituency actually is one might think it’s a very powerful one, all those 18-30 floating voters. Politicians pay attention to him because they want a hook into contemporary culture. They imagine these are the people he represents.” (Iley 2005) Bono responds, Outside of that I represent a lot of people who have no voice at all. In the world’s order of things, they are the people that count least. They are the 6,500 people who are dying of Aids in Africa every day for no good reason. They haven’t asked me to represent them. It’s cheeky but I hope they’re glad I do... (Iley, 2005) This statement illustrates the distinctions between what I have called the audience, the affected constituency and the surrogate constituency. With the authority of the audience (leaders of the G8 countries), Bono is acting on behalf of those he considers to be affected by AIDS, and 14  Bono and ONE advocate “smart aid”: aid that is accountable to those for whom it is intended. 160  policies on debt and trade, though with the authorization of a constituency of a people who are not affected, but concerned. In this example, the surrogate constituency –the 18-30 floating voters – is the one that organizationally authorizes Bono by, for example, their membership in ONE, and may hold Bono accountable by exercising exit accountability. It is also this constituency of people that purchase RED products: merchandise that has earmarked 50% of its profits for contribution to The Global Fund, which uses “100% of this money to finance HIV health and community support programs in Africa, with a focus on women and children” (Join Red 2010). Thus, we can see that entrance and exit-based authorization and accountability may be working in these market-based contexts as well. Public reputational accountability certainly plays a role as constituencies and audiences hold Bono to account for his purpose. A surrogate representative is responsive to the needs of the affected but receives authorization from and is accountable to the surrogate constituency, which acts as a sympathetic proxy. ONE does not have a clear relationship of authorization by, and accountability to, the affected. In this way, the surrogate self-appointed representative does not enable the empowered inclusion of the affected – possibly due to circumstance, as I suggested might occur if, for example, China closes Tibet’s borders so that those outside Tibet who claim to act on behalf of Tibetans have a difficult time contacting the affected constituency. This kind of representation is not democratic, but neither does it interfere with the possibility that this might be achieved. Moreover, the surrogate self-appointed representative might still achieve important goods for the affected, which is why I use the term “surrogate,” indicating care for the affected, though it is not democratic. Such care is preferable to no representation at all, which is often the alternative. More positively, when a surrogate self-appointed representative’s claim benefits conditions of self-development and self-determination, it underwrites conditions for democracy, anticipating a moment when the affected will be active and agentic. When Bono secures funding for AIDS medication, he is helping people to live with HIV/AIDS, rather than suffer and die 161  from the disease. With respect to this claim, Bono is a surrogate self-appointed representative because he offers a claim of representation for the benefit of an affected constituency, but said constituency is not empowered to authorize those claims and hold him accountable for his activities. Instead, it is the surrogate constituency that is empowered in this way. In short, Bono demonstrates responsiveness to, but does not foster the empowered inclusion of, an affected constituency. What about Bono’s claims with respect to aid and debt relief? When we consider representation through the framework of the representative claim, we are able to parse the claim, as I suggested in Chapter Three, identifying the existential, constitutive, and performative elements, the latter of which is issue-based. Claim-based and issue-based representation implies that an individual actor can be a surrogate representative with respect to one claim or issue and perhaps a failed representative with respect to another. And in fact, this may be the case with Bono: a surrogate representative of those Africans with AIDS, but perhaps fails to represent Africans when he issues claims about aid and debt relief. There are those who argue that aid and debt relief may actually have negative effects on self-development and self-determination. For example, the journalist Andrew Mwenda says, “Unfortunately…foreign aid and debt relief can exacerbate Africa’s problems by postponing economic reforms and the emergence of a transparent and accountable government” (2006, 1). The case of Bono illustrates that the discursive power of the self-appointed representative works only through agreement or consent. As long as an audience such as the leaders of the G8 countries (and a surrogate constituency such as 18-30 year-old voters), agrees with Bono’s claims rather than Mwenda’s, Bono remains a self-appointed representative. Relatedly, the claims of the self-appointed representative are connected to the public opinion that forms in the public sphere, which secures accountability. As discussed in Chapter Five, the claims of the representative must be made public in order to be judged, and this publicity enables both authorization and also accountability. Because his 162  communication is public, Mwenda holds Bono accountable through mechanisms such as voice, by publicly disagreeing with such arguments about aid and debt relief, which has the potential to change public opinion on this issue. It is not only public opinion that might be changed through the publicity of such claims and counter-claims, but people’s understanding of interests. As Bono offers his claim about the importance of smart aid and debt relief for African self-development, Mwenda responds to this claim, disagreeing with his interpretation and offering his reasons for that disagreement. When Bono’s claim resonates with a constituency, that constituency mobilizes around his claim that aid and debt relief benefits conditions of self-development. If Mwenda’s claim resonates with a constituency, it will mobilize around his claim that aid and debt relieve undermine selfdevelopment. For now, it seems that audiences and surrogate constituencies largely take up Bono’s claim; he is a surrogate representative because he still does not have a direct relationship of authorization and accountability with those affected, but is responsive to that constituency, as we currently understand that responsiveness. But that understanding could change. The publicity of these claims helps ensure that they can and likely will be revisited at any time, and our judgments may change so that we agree with Mwenda that debt relief undermines selfdevelopment. Put another way, everything is always potentially on the agenda all the time. And the potential – and the danger – of self-appointed representatives is that they exert a form of influence, namely, the influence of argument, which can do a lot of work in shaping our understanding of our interests. A second example of a surrogate self-appointed representative is Amnesty International. Amnesty campaigns for internationally recognized human rights to be respected and protected for everyone. Amnesty and its “members and supporters exert influence on governments, political bodies, companies, and intergovernmental groups” (Amnesty 2010). This claim of representation establishes victims of human rights abuses as the affected constituency. Amnesty 163  is responsive to this constituency by working to “prevent and end grave abuses of human rights” and “demands justice for those whose rights have been violated” (Amnesty 2010). The claim establishes the governments, political bodies, companies, and intergovernmental groups as its relevant audience: that group of people who must recognize a claimant as a representative. The claim also establishes its “members and supporters” as the surrogate constituency that acts on behalf of the affected constituency. That is, Amnesty is responsive to but does not promote the empowered inclusion of the affected constituency. The surrogate constituency organizationally authorizes Amnesty by becoming members and discursively authorizes Amnesty by participating in “mass demonstrations, vigils, and direct lobbying as well as online and offline campaigning” (Amnesty 2010). Correspondingly, the surrogate constituency can hold Amnesty accountable by organizational exit, where members can allow their membership to lapse. Voice accountability is also used to express dissatisfaction with the organization as when Amnesty comes under scrutiny for being anti-Israeli, for example. This criticism also affects their public reputation as an organization that depends for its legitimacy upon its reputation as a provider of objective expertise (Price 2003). A final example of a surrogate self-appointed representative is Human Rights Watch (HRW). HRW “is dedicated to protecting the human rights of people around the world” and focuses “international attention where human rights are violated” (HRW 2010). The organization is responsive to affected constituencies when it seeks to hold abusers of human rights accountable by building public pressure for action and raising “the cost of human rights abuse.” Each year, Human Rights Watch publishes more than 100 reports and briefings on human rights conditions in some 90 countries, generating extensive coverage in local and international media. With the leverage this brings, Human Rights Watch meets with governments, the United Nations, regional groups like the African Union and the 164  European Union, financial institutions, and corporations to press for changes in policy and practice that promote human rights and justice around the world. (HRW 2010). For example, in 2009, HRW discovered that Zimbabwe’s armed forces were engaging in the forced labour of children and adults, torturing and beating local villagers in the diamond mines of Marange. HRW “called on the Kimberley Process Certification Scheme (KPCS), an international group governing the global diamond industry, to press Zimbabwe, a participant, to end the smuggling of diamonds,” which would protect Zimbabweans from abuse, and also called on consumers to be responsible in their purchase of diamonds (HRW 2010). With this claim, HRW constitutes an affected constituency (the Zimbabweans who are victims of human rights abuses), and a surrogate constituency (consumers of diamonds). However, HRW is not always allowed in to Zimbabwe, which interferes with their ability to promote the empowered inclusion of the affected constituency. When asked by the Associated Press whether Amnesty International and Human Rights Watch would be allowed to enter Zimbabwe, Robert Mugabe answered, “Ha! Ha! Ha! Ha! Ha! Ha! Let them keep out” (Penketh 2008). In this case, HRW does not always have access to the affected constituency and so, by necessity, acts as a surrogate representative. With respect to the blood diamonds, HRW also called on consumers as a sympathetic, proxy constituency, to purchase diamonds responsibly and so help stop the flow of funds to a government that perpetuates human rights abuses. This surrogate constituency can organizationally authorize HRW’s claims by its membership and/or discursively authorize its claims by respecting its request to purchase diamonds responsibly. The surrogate constituency can then hold HRW accountable by organizational exit.  Skewed Self-Appointed Representatives When the self-appointed representative enables empowered inclusion for one  165  constituency, with the effect of undermining another constituency’s self-development and selfdetermination, I call this skewed representation. From the perspective of the theory of representation I develop here, what defines skewed representation is that those affected by the claims and actions of representatives are not included in the representation, nor are their interests held in view – the feature that distinguishes skewed representation from the surrogate kind. This kind of self-appointed representative does not respond to the need for the basic conditions necessary for those they affect to make life choices, and so fails in the dimension of responsiveness to an affected constituency. Instead, this self-appointed representative advances the claims of either: a) interests that are proportionally less affected than another constituency, and/or b) vested interests that already benefit from advantages in the cultural, economic, and/or political dimensions. In the latter case, the term “skewed representation” calls attention to a problem that is common in pluralist democracies: on average, well-resourced people and groups tend to be over-represented in policy making and other kinds of processes that affect people well beyond those who are represented. It remains true, as Schattschneider (1960) famously commented, that in the pluralist heaven, the interest group chorus sings with an upper-class accent. The skewed self-appointed representative thereby contributes to the over-representation of these interests, and undermines the affected constituency by providing the privileged with yet another forum of representation. When self-appointed representatives advance the first group – interests that are proportionally less affected than another constituency – the concept of skewed representation speaks to an objection often voiced in relation to the affected interests principle, mentioned earlier: that there are often multiple groups who are affected, and that it is difficult to distinguish whether, and if so how, one group is just as affected or more affected than another. For example, if working class taxpayers are asked to bear the cost of debt relief or foreign aid, then can they claim affectedness? I would suggest that they can: it is likely that the self-development of the 166  poor or working class taxpayer would be harmed by even a marginally higher rate of tax; but as suggested earlier, the wealthy taxpayer likely cannot claim that their self-development would be harmed by a marginally higher rate of tax or, at least, not to the same extent. All of this is to suggest that an understanding of proportionality should supplement the affected interests principle, and to acknowledge differential affectedness that will have to be worked out in the public sphere. Jerry Falwell and the Moral Majority provide an example of such differential affectedness that is, in some places, still being worked out. In the 1980s in the United States, Jerry Falwell provided representation for the “Moral Majority,” a group of largely fundamentalist evangelical Christians that he cofounded in 1979. Jerry Falwell was “a Baptist Bible Fellowship pastor who had built the Thomas Road Baptist Church in Lynchburg, Viginia, from an initial gathering of thirty-five adults into a megachurch with more than 15,000 members” (Wilcox 2000, 36). According to its website, this organization was launched “with a mission of organizing evangelical leaders who will boldly engage the culture. The Moral Majority kicks off with a pro-life, pro-traditional family, pro-national defense and pro-Israel platform” (Moral Majority 2008). Issues included, but were not limited to, “opposition to abortion, to civil rights protection for gays and lesbians, and to the ERA and support for school prayer and tuition tax credits for religious schools” (Wilcox 2000, 37). Falwell’s claim to oppose state recognition and acceptance of homosexuality posited two constituencies: those who authorized the claim (the Moral Majority), and those affected (or most affected) by the claim (lesbian, gay, bisexual, and transgendered peoples). With respect to his claim about homosexuality, Falwell received authorization from the Moral Majority; it is they who guided, informed, and sanctioned his behaviour. Its legitimacy, however, can be challenged on the basis of its unresponsiveness to those (most) affected by the representative claim. From the perspective of democratic theory, self-appointed representatives are constitutively legitimate when they are responsive to those effects that result in oppression, defined as institutional 167  constraints on self-development, and to those effects that result in domination, defined as institutional constraint on self-determination (Young 2000, 31). Falwell’s claim is an example of one group that would claim that their interests are affected by civil rights for LGBT peoples – a group that is arguably less affected than the LGBT peoples whose self-development and selfdetermination would certainly be undermined by the success of their claim.15 Those who authorized Falwell were able to hold him accountable through mechanisms of exit and public reputational accountability. Indeed, we can see evidence of these kinds of accountability in the fact that the financial support of Christian Right groups began to decline in the mid-1980s. According to Clyde Wilcox, this decline occurred for two reasons. One, Reagan was re-elected, which satisfied conservative Christians. And two, there were a number of scandals involving televangelists, which made people reluctant to donate (Wilcox 2000 37). Moreover, the Moral Majority often came under fire for misleading their members. The Moral Majority frequently mobilized its members to bury Congress in mail and phone calls. Often these appeals were misleading. While Congress was considering legislation to reverse Supreme Court decisions on civil rights, Falwell mobilized his followers by telling them that the bill would classify sin as a handicap and then force churches to hire as youth counsellors “active homosexuals, transvestites, alcoholics, and drug addicts, among others. (O’Hara 1989, 13, cited in Wilcox 2000, 91) In other words, those who authorized Falwell eventually held him accountable through exit. By 1988, the Moral Majority was strapped for cash and the organization soon disbanded. “Falwell claimed he quit because he had accomplished his goal, but the key issue agenda of the Moral 15  This claim continues to be worked out in the public sphere and, notably, Reverend Falwell, in  his later years, seems to have changed his stance on this issue, claiming that civil rights are an American value.  168  Majority remained unrealized” (Wilcox 2000, 38). Falwell was “immensely unpopular and his organization was bankrupt” (Wilcox 2000, 135). In addition to exit, then, Falwell was also held accountable through public reputational accountability. Jerry Falwell is an example of skewed self-appointed representative because he offered a claim of representation for the benefit of a constituency that would result in the exclusion and marginalization of another constituency. Those (most) affected by his claims are not empowered to authorize those claims and cannot hold him accountable for his activities. Instead, it is the skewed constituency who is empowered in this way.  Failed Self-Appointed Representatives Finally, if a self-appointed representative claims to represent an affected constituency, but is both unresponsive to the affected constituency and also exclusive of the affected constituency, the attempt to represent fails. This effort to represent fails because, like skewed representation, it disregards the self-development and self-determination of an affected constituency. However, unlike skewed representatives, this kind of self-appointed representative claims to represent the interests of a given constituency, but still fails in the dimensions of responsiveness to, and empowerment of, the affected constituency. As is the case with other categories, what counts as “failed representation” depends upon these two relationships: to a claimed constituency and an affected constituency. Indeed, some of the most clear-cut examples have less to do with representative claimants that fail outright, but rather those that over-reach – in effect claiming constituencies they do not have, and which do not recognize them, and fail to represent issues on behalf of these constituencies. For example, organizations such as the National Association for the Advancement of Colored People (NAACP), or the National Organization for Women (NOW), claim to represent broad-based  169  constituencies – “all ethnic minorities” and “all women,” respectively. The breadth of these claims include what Dara Strolovitch (2007) calls “disadvantaged subgroups” within that broad constituency – those who are poor and/or gay and lesbian, as examples, and so face political, cultural, and economic challenges that others within that broader constituency of “African American” and “women” do not. Strolovitch finds that such organizations do not often do a very good job of representing the interests of disadvantaged subgroups and, in fact, privilege advantaged subgroups of those constituencies. In her book on Affirmative Advocacy, Strolovitch argues that “it does not suffice to distinguish only between advantaged and disadvantaged groups” (2007, 7). To understand the priorities and activities of advocacy organizations, we must distinguish among four types of issues affecting four differently situated constituencies: universal issues, which, at least in theory, affect the population as a whole, regardless of race, gender, sexual orientation, disability, class, or any other identity; majority issues, which affect an organization’s members or constituents relatively equally; disadvantaged-subgroup issues, which affect an organization’s constituents who are disadvantaged economically, socially, or politically compared to the broader constituency; and advantaged-subgroup issues, which also affect a subgroup of an organization’s constituents but one that is relatively advantaged compared to the broader constituency. (Strolovitch 2007, 7) Strolovitch finds that issues affecting advantaged subgroups are given disproportionately high levels of attention by these organizations, as compared to issues affecting disadvantaged subgroups, and also majority issues. She provides the example of the attention paid to affirmative action in higher education, which affects a subgroup of relatively advantaged women, versus the relative lack of attention paid to welfare reform, which affects intersectionally disadvantaged women: “welfare reform has a disproportionately high chance of 170  affecting specific subgroups of women – in particular low-income women and women of color, that is, intersectionally disadvantaged subgroups of all women” (Strolovitch 2007, 30). According to Strolovitch, such organizations do a fine job of representing their broad constituencies and their advantaged-subgroups but they do not perform as well with respect to representing the least well off among their constituents/members. Strolovitch is evaluating such actors by the extent to which their work benefits the least well off among their constituents/members. This is an approach with which I agree, and I suggest something similar when I define affected constituencies as those whose interests are or would be undermined. Issue-based, claim-based representation allows us to parse the claim so that it is possible for an individual actor or organization to be a democratic representative with respect to one claim and a failed representative with respect to another. So the NAACP and NOW need not be judged entirely as failed representatives; rather, at a particular time in their organization’s history, they failed in their claim to represent disadvantaged-subgroups with respect to certain issues, but provided skewed and perhaps democratic representation for their broad constituencies with respect to other issues. To their credit, once made aware of these problems, the National Organization for Women was responsive to their disadvantaged subgroups and now counts economic justice and lesbian rights among their issues (NOW 2010). All of this is to suggest that this theory provides criteria by which to analyze such representative relationships, rather than to make judgments about an organization’s identity or functions apart from these relationships. NOW, like most organizations, are complicated entities that have identities that stretch over time and space, and that have and will continue to evolve. In an ideal world, they would evolve in response to representative claims that fail, to be replaced by representative claims that are democratically legitimate. Another case of a failed representative claim occurs in the women’s movement, exemplified by third-wave feminists’ accusations that second-wave feminists failed to represent 171  their interests. In personal correspondence, Jane Mansbridge spoke of the possibility of being a self-appointed representative, and discussed her participation in the occupation of a Harvard University building for 10 days. [W]hen I was with a group of women who occupied a Harvard building on 888 Memorial Drive just across the river, making in some sense a “representative claim” to represent “women,” did we have a required responsibility to engage in extensive and widespread deliberation with those we claimed to represent before we acted? In some sense, yes, we did have that responsibility. In the case of the 888 occupation, the planners in our group were quite sophisticated veterans of SDS, so they were careful to consult – clandestinely of course – with leaders in the predominantly African American community in the area around the building we planned to occupy, and got their support by broadening our demands from demands aimed primarily at women to include demands for housing support for the local community. Now, of course these were consultations and deliberations among two groups of self-appointed representatives, but it was far better than the usual lack of such consultation. We did not, however, consult with all “women” – or any kind of representative sample of women – before making claims, which I think could be called self-appointed representative claims, for all women. Our name for the movement – the “women’s movement” – I believe embodied such representative claims. As it turned out, our self-appointed claims to speak for “women” infuriated many women who considered themselves seriously misrepresented by the movement. Mansbridge is referring here to the anti-feminist movement, led by Phyllis Schlafly, who successfully challenged the Equal Rights Amendment. But in response to this women’s movement there eventually arose a “third wave” of feminism that demanded that attention be paid to, to use Strolovitch’s term, disadvantaged subgroups, e.g. women of colour, rather than 172  only to upper middle class white women or “advantaged sub-groups.” In a 2005 interview, Judy Rebick, a Canadian journalist, political activist, and former president of the National Action Committee on the Status of Women (NAC), says, “I’ll never forget, the first year I was president of NAC…I went to a meeting of NOW…and there were fewer women of colour in NOW than there were in NAC. And with the history of Black activism in the States, I was quite shocked by that” (Rebick 2005). Second- and Third-wave feminism have many dimensions, of course, but the point here is to suggest that individual and collective actors can achieve democratic representation with respect to one claim but fail in that representation with respect to another: the women’s movement aimed for the elimination of sexism in all areas of life, and had legal successes. But the second-wave has been accused of being a white middle-class movement, and so may have failed in their representation of women of colour on issues that affected them particularly.  Conclusion In this chapter, I have shown that identifying the criteria I have developed – responsiveness to affected constituencies and their empowered inclusion – can help to determine the types of self-appointed representatives at work, and also how they accomplish this work. That is, whether or not a self-appointed representative serves to enable the empowered inclusion of people whose interests are potentially affected by collective decisions in influencing those decisions. The criteria are demanding: they require not only that a representative is responsive to this affected constituency, but that it is linked to the constituency through relationships of authorization and accountability. Of course, most self-appointed representatives do not meet this threshold – often necessarily, because affected constituencies exist beyond borders, or live within dictatorships, or exist under conditions that stunt their basic conditions of development.  173  However, I have suggested that self-appointed representatives may be beneficial to these constituencies even if relationships of authorization and accountability are weak or non-existent – if the representative is responsive to the affected constituency. This is a kind of surrogate representation that is not yet democratic, but it may also provide the conditions the enable democratic development. That said, it is by no means clear that, on average, self-appointed representatives enhance democracy (democratic representation) or underwrite its development (surrogate representation). That is why I have identified two other kinds of self-appointed representatives, which mirror, theoretically, democratic and surrogate representation: skewed and failed representation. Skewed representation biases democracy, enhancing the voices of those who are less affected than others, and/or who are well resourced and powerful, at the expense of the affected, with anti-democratic consequences. Skewed representation is not, however, exceptional: it is as much a part of our polities as other forms. Failed representation is also common – indeed, probably more so in the marketplace of representative claims and claimants, and is very likely part of any deliberative form of democracy. This theory of self-appointed representation cannot change these facts; what it can do is point out that self-appointed representation is integral in making democracies work, by providing voice for those who would otherwise go without, because the issues that affect the life chances of people – that is, conditions of self-development and self-determination – fall outside of democracy, cross jurisdictions, land between borders, or because, even within democracy’s jurisdictions and borders, peoples are historically oppressed and/or dominated. This theory also identifies, from the standpoint of democracy, better and worse forms of self-appointed representation, providing criteria to judge these actors who claim to represent, and often claim to legitimately represent, others, or who are received as legitimate representatives without grounds.  174  Chapter Seven: Conclusion In this dissertation I ask how should we theorize and normatively assess those individuals and organizations that claim to represent others for political purposes, but do so without the electoral authorization and accountability usually thought to be at the heart of democratic representation? Standard accounts of democratic representation involve both the authorization of a representative by election, and the accountability of elected officials to their constituents for their performance in office. Yet self-appointed representatives, such as nongovernmental organizations, are increasingly included in the deliberation processes of institutions such as the European Union, the United Nations, and the World Trade Organization. There is a need for a theory of representation that will both identify and assess the types of selfappointed representatives that, although unelected, comprise growing and important parts of our political landscape. In this dissertation, I have developed a theory of representation that goes beyond the standard identification of democratic representation with electoral representation. The reasons we need to develop new theory are both empirical and normative: the standard account does little to identify, much less assess, the many new kinds of representative claims made by selfappointed actors in both formal and non-formal settings. Further, the standard account conflates the institutional identification of the representative, primarily through fair election, with its normative/democratic legitimacy, and thus lacks both analytic and normative relevance to nonelectoral forms of representation. The theory of democratic representation that I develop here has at least four advantages over the standard account. First, it moves us towards an account of representation that can identify and assess forms of democratic representation that do not reduce to, or even depend upon, electoral representation. Second, it helps us to conceptualize democratic potentials of  175  representation beyond the nation-state, in the global arena. Third, it theorizes representation for peoples across borders and across electoral memberships, rather than assuming historicallycontingent electorates as exhaustive of democratic constituencies. And finally, it helps us to think normatively about what are, in fact, very messy representative relationships that comprise contemporary political systems. The argument, put simply, is this: Self-appointed representatives are potentially important parts of political systems because they can, potentially, give voice to the peoples and interests that are not taken up by existing institutions. But if this is so, then democratic theorists need ways of assessing these representative claims so that we can judge them to be more or less legitimate. To develop the theory required to address these possibilities, my strategy has been to separate the concept of democratic representation from its connection to electoral representation, which then enables us to ask more generic questions about •  the responsiveness of a representative to an affected constituency; and  •  the capacities of that constituency to authorize the representative and hold him/her accountable (empowered inclusion)  By analyzing these two functions and the relationships embedded within them, we can both theorize kinds of self-appointed representation, and judge their legitimacy. In developing this argument, I have followed Michael Saward’s critical approach to representative claim-making: I agree with Saward that we should see “representation in terms of claims to be representative by a variety of political actors, rather than (as is normally the case) seeing it as an achieved, or potentially achievable, state of affairs as a result of election” (2006, 5). Building on Saward’s critical approach, I have suggested that the representative claim consists of the following three kinds of actions: •  The existential self-identification as a representative (e.g. “I am a representative of this group”); 176  •  The constitutive identification of a constituency (e.g. “this group” is composed of “the poor,” or “Africans suffering from AIDS,” or “victims of human rights abuses,” or “evangelical Christians”);  •  The performative identification of a good/purpose/issue (e.g. “The poor need food and water; give it to them”).  Through these three kinds of actions, the self-appointed representative constitutes the people she represents through the claim of representation, by identifying – offering a definition of – a people to be represented who, under ideal circumstances, may then accept or deny the claim. If that constituency is composed of those affected by the claim, then this is, in principle, where the democratic features of self-appointed representation are to be found: in its constitutive dimension, that is, in its ability to identify and mobilize affected constituencies around claims of representation. Understanding representation in this way provides a solution to what has long been considered a paradox in democratic theory. Democratic theory traditionally holds that the formation of the demos cannot be democratic because not everyone is included in the decisionmaking. As Whelan points out, democracy, “a method for group decision making or selfgovernance, cannot be brought to bear on the logically prior matter of the constitution of the group, the existence of which it presupposes” (1983, 40). My theory of representation sketches a logic that flips this assumption on its head. I have suggested that through processes of authorizing and rejecting self-appointed claims of representation, it is theoretically possible for everyone to decide whether or not they want to be part of this and/or that people. Self-appointed representatives expand people’s representative options: confronted with multiple representative claims, individuals can choose which concern(s) is a priority and requires representation, in addition to the representation one receives as a member of an electorate, if available. In this way, self-appointed representatives engage in a highly responsive process of constituency 177  formation, constituting alternative, issue-defined, overlapping peoples. The self-appointed representative thus enables an understanding of democracy as an ideal of self-determination in which, as Bohman describes, “the terms and boundaries of democracy are made by citizens themselves and not others” (2007, 2). That is, when individuals authorize or reject claims of representation, they choose the boundaries of their own demos. It is for this reason that the traditional paradox does not apply. Such a concept of constituency formation enables an alternative conception of democratic representation, one that begins with the problem of who needs representation, namely, those affected or potentially affected by collective decisions. The affected interests principle suggests that those whose interests are affected by collective decisions should have influence over those decisions. I have followed Young and conceived of “affected interests” as those relevant effects that enable or constrain self-development and self-determination – those basic conditions required to make life choices, including both the distribution of resources and also “the institutional organization of power, status, and communication” (Young 2000, 32). As Young explains, there are systematic institutional processes “which inhibit people’s ability to play and communicate with others or to express their feelings and perspective on social life in contexts where others can listen” (2000, 156). The “affected,” then, are those whose interests are (or would be) inhibited by a structural, systemic domination and/or oppression. Those who make representative claims outside of electoral processes – those who “self-appoint” – function democratically, I argue, only to the extent that non-electoral mechanisms of authorization and accountability connect the self-appointed representative to an affected constituency. In this way, self-appointed representation can be justified through its contributions to the basic democratic norm that those interests potentially affected by collective decisions should have influence over those decisions. This understanding of constituency formation contributes to a plurality of “citizenships,” and is a necessary step in developing a democratic theory that is appropriate to 178  the complex, globalizing, pluralistic, and highly differentiated societies within which we now live. The constitutive legitimacy of self-appointed representation, as I have constructed the concept, is determined by the responsiveness of that actor to an affected constituency. If the self-appointed representative is responsive to an affected constituency, that is, works to improve conditions of self-determination and/or self-development, though without their empowered inclusion, then that self-appointed representative is a constitutively legitimate representative. If a constitutively legitimate self-appointed representative, that is, one who is responsive to an affected constituency, is also authorized by and held accountable by that affected constituency, then she is a democratically legitimate representative. If, however, the self-appointed representative is unresponsive to an affected constituency, that is, undermines self-determination and/or self-development, then that self-appointed representative is an illegitimate representative. The theory I have developed here addresses the vexing issues of authorization and accountability beyond electoral mechanisms by conceptualizing functional equivalents to standard electoral authorization and accountability mechanisms. I theorize non-electoral mechanisms by which the self-appointed representative might receive approval from, and also be sanctioned by, a constituency. For example, organizational authorization might take the form of membership: the self-appointed representative acquires a following or membership based on a convergence of interests as expressed in a mission statement. Although authorization and accountability are conceptually distinct, sometimes the same mechanism serves both: the vote, for example, is both a mechanism of authorization (election) and a means of accountability (reelection or removal from office). Similarly, membership will serve both authorization and accountability: through joining (authorization) and through exit or anticipation of exit (accountability). Discursive authorization might take the form of public agreement (authorization) or disagreement (accountability), expressed through protests, boycotts, letter 179  writing, and/or petitions, among others. Thus, self-appointed representatives, though unelected, can be authorized and held accountable for their representative claims. It is simply that they will be held accountable differently than states and legislators. In fact, the reason we want to accept these different kinds of mechanisms is due, in part, to the fact that the powers assumed by self-appointed representatives are different than those of representatives elected as part of state-organized systems. If a representative can, for example, send people to war, citizens require institutionalized empowerments: not just votes, but also the rules and rights of the kinds identified, for example, by Dahl’s well-known criteria of democracy (Dahl 1989, chap. 4). The self-appointed representative primarily has discursive power – the power to make audible those voices that have been excluded or marginalized – but lacks the decision-making power of states and administrators. Accordingly, the democratic legitimacy thresholds for the self-appointed representative will be, and should be, less than for those representatives who guide the deployment of the powers of state. But they will also be different: democratic legitimacy, typically, will be won through the means of claim-making and public discourse, broadly understood, and supplemented by organizational mechanisms and indicators such as memberships, followings, and discursive mechanisms such as public reputational accountability, and all of this directed towards, and exercised by, affected constituencies. To summarize, in order to function as a democratic representative, the self-appointed representative must be responsive to an affected constituency and provide mechanisms by which the affected constituency may offer discursive and/or organizational authorization of the representative claim, and hold the representative accountable via one or more non-electoral mechanisms. Put another way, to assess the relationship between the self-appointed representative and the affected constituency as “democratic,” requires the self-appointed representative to be both responsive to the affected constituency, seeking to improve conditions 180  of self-development and self-determination, and also to promote the empowered inclusion of the affected constituency in order that they may guide, inform, and sanction the representative. If the self-appointed representative acts on only the first of these dimensions, that is, the selfappointed representative is responsive to, but does not promote the empowered inclusion of, the affected constituency, I refer to the representation as surrogate – a constitutively legitimate form of representation, particularly if the alternative is that constituencies have no representation at all. If the self-appointed representative is unresponsive to an affected constituency, and promotes the empowered inclusion of those who are less affected and/or already privileged, I refer to the representation as skewed. Finally, if the actions of a self-appointed representative fail on both dimensions of responsiveness and empowerment, that is, the self-appointed representative is both unresponsive to and exclusive of an affected constituency, perhaps acting only on the authority of the audience, then representation simply fails. Conceptualizing these relationships in this manner clarifies where self-appointed representatives work to democratic effect and in democratic ways, and also where they do not work to democratic effect and in nondemocratic ways. Self-appointed representation speaks to a condition in which the institutions of representative democracy are underdeveloped in most of the world, and increasingly inadequate in the countries with functioning democratic institutions. It may look unfamiliar, as it differs from our traditional understanding of democratic representation, which is based on formal authorization and accountability. However, the potential of self-appointed representation to be responsive and promote empowerment serves an important function for the many groups, interests, and discourses that are unrepresented in formal political institutions. Self-appointed representation is therefore not only normatively desirable, but is almost certainly essential to the deepening and development of democracy in a complex, post-national world.  181  The Future of Democratic Theories of Representation In developing this approach to democratic representation outside of electoral democracy, I have focused primarily on the democratic theory of representation. Yet the implications of my approach go well beyond those I have been able to develop here. This view of democratic representation has important implications for how we think about challenges for democratic theory and practice such as civil society organizations, citizen representatives, the idea of tacit consent, representing future generations, and assessing systems of representation.  Formal Political Roles for Civil Society Organizations While this project has been limited to developing a theoretical framework for analyzing self-appointed representatives in informal contexts, the framework I develop here may illuminate the emerging complexities of formal democratic representation. Consider, for example, how the European Union (EU) involves civil society organizations to buttress democratic legitimacy. Faced with a mismatch between the powers of the EU and the ability of citizens to exert control, the EU has been criticized for its “democratic deficit”: the European Parliament is the only directly elected parliamentary institution in the EU and, as Magnette explains, [T]his only very indirectly makes EU institutions accountable. The Council of Ministers and the European Council are not affected by European elections, and the composition of the Commission is only slightly dependent on the results of these transnational elections…Because it does not correspond to the mechanisms of participation and accountability with which citizens are familiar, the EU is often said to suffer from a congenial ‘democratic deficit’, and its constitutional reform is the object of a permanent debate. (2003, 144)  182  In response to its perceived “democratic deficit,” the EU has increasingly included selfappointed representatives – civil society organizations (CSOs) and non-governmental organizations (NGOs) – in formal state-like processes: “EU institutions openly call for civil society participation in decision-making and policy implementation to strengthen the political system’s democratic character” (Saurugger 2007, 385). The White Paper on European Governance argues for “opening up the policy-making process to get more people and organizations involved in shaping and delivering EU policy” (Commission of the European Communities 2001, 3). The White Paper claims that CSOs “play an important role in giving voice to the concerns of citizens,” and that they “mobilize people and support, for instance, those suffering from exclusion or discrimination” (Commission of the European Communities 2001, 14). NGOs, in particular, “play an important role at global level in development policy” often acting “as an early warning system for the direction of political debate” (Commission of the European Communities 2001, 14). The concern is that the inclusion of CSOs and NGOs might skew representation in the direction of groups that are already empowered. Self-appointed representatives have political standing that sometimes parallels that of elected representatives, but certainly privileges their “voice” in consultation processes and deliberative forums. How might we judge their democratic bona fides? Similarly, in countries in Latin America, civil society actors are increasingly included in participatory “councils” that affect public policy formation and implementation on issues of health care, social assistance, and the environment. In Brazil, these “hybrid” institutions, in which civil society and state actors share decision-making, are formally recognized as representatives of specific categories of civil society, and have the formal authority to make policy affecting health care, social assistance, and the environment at the municipal, river basin, state, and national levels (Coelho and Marcos 2004; Abers and Keck 2009).  183  The theory I have developed here suggests that the inclusion of self-appointed representatives in formal political processes only remedies democratic deficits insofar as the following conditions are met: the self-appointed representative offers a claim of representation that is responsive to an affected constituency: those whose interests, defined by conditions of self-development and/or self-determination, are or would be undermined. Further, the selfappointed must promote the empowered inclusion of the affected constituency, providing them with opportunities to authorize their claims and hold them accountable. And these mechanisms of authorization and accountability must be appropriate with respect to the powers and functions of the self-appointed representative. There is also a critical criterion: the inclusion of selfappointed representatives should not provide more representation for those who are already well represented by representative institutions, thereby producing skewed representation. By extending my general theory of democratic representation into domains where the standard mechanisms of representation function poorly, or do not exist at all, including the postnational level, my theory has the potential to interpret and assess global actors from the standpoint of their representative functions. The theory stems from a new understanding of representation that moves away from the idea that representatives who are elected are the sole repositories of democratic legitimacy. Instead, by viewing the process of representative claim making as an active process of constituency formation, we can frame the problem of “democratic representation” broadly enough to capture and assess these emerging political forms. These forms have important implications for how we conceive of the relationship between geographical borders and democracy, between electoral processes and democracy, and between global actors and democracy. From an empirical perspective, comparative political scientists are under pressure to empirically demonstrate the legitimacy of these representatives who make policies and share in formal decision-making. My approach provides normative criteria by which to judge the contributions of such representatives to democracy, and also has 184  the potential to provide empirical indicators that point to how citizens recognize the right of civil society organizations to provide representation – indicators such as membership and exit.  Citizen Representatives Another emerging political phenomenon that is likely to prove important to the evolution of democracy and democratization might be called “citizen representatives” (Warren 2008). The term refers to a relatively new class of political devices in which bodies comprised of ordinary citizens provide representative functions for other citizens. An example would be a “minipublic” (Goodin and Dryzek 2006): a body that is typically constituted as a random or stratified sample, and typically charged with learning about deliberating a particular issue in order to make a recommendation to policy makers and/or to the citizenry. For example, the purpose of the British Columbia Citizen’s Assembly was to assess the province’s electoral system and, if deemed necessary, propose an alternative that would be put to the popular vote. As the name suggests, the assembly was a collection of citizens: 160 citizens, one male and female, from each of the 79 ridings in British Columbia, plus two aboriginal citizens. These citizens met every other week for one year, learning about different voting systems, deliberating, and ultimately providing a recommendation to the province that it reconsider its current single member plurality system and instead adopt a single-transferable vote system (Warren 2008). Increasingly, these bodies are constituted as “representative” bodies, even though their members are neither elected nor (typically) appointed by duly elected officials. Warren argues that the citizen’s assembly is not so much a contribution to participatory democracy, as is sometimes suggested, but a new kind of representative institution. The “notion of participatory democracy suggests that most citizens should participate in self-rule” (Warren 2008, 3), but only 160 citizens participated in the citizen’s assembly – “a fraction of BC’s 4.2 million people”  185  (Warren 2008, 3). Indeed, the assembly was constituted as a “representative sample” of the people of British Columbia – clearly a descriptive representative purpose, although as it developed, the assembly quite self-consciously assumed other kinds of representative roles (Pearse 2008). There are also other, less formalized cases in which citizens act as representatives of other citizens, even though there exist no explicit representative claims. For example, Amnesty International regards every person who signs a petition as representing ten people who have not done so, and citizens who attend town meetings informally represent those who do not attend. In the latter example, the attendees make no claim to represent those who are not there, and perhaps would not welcome a suggestion that they have a responsibility to speak not only for themselves but for the absent townspeople as well. Nevertheless, in a certain way they do stand and act for the absent townspeople, as their voice may, for example, change town by-laws and so affect the absent townspeople. Non-electoral representation, then, seems to require political theorists to consider the representative system as a whole, and to think of the active citizenry as informal representatives of less active citizens. But how well do such active citizens do their “job” of representing the inactive? Under what conditions can such informal and often unintentional representation be rendered legitimate?  Tacit Consent The line of inquiry I develop here has another important implication: the challenges of determining the legitimacy of non-electoral representation often depends quite heavily on tacit consent: that is, on the represented people not actively repudiating non-elected acts of representation. For example, Barney Frank, a member of Congress from Massachusetts, acts as a surrogate representative for gay and lesbian citizens throughout the United States. I argue that  186  he receives “informal authorization” to do this work – if his constituents did not approve, they could choose to hold him accountable by voting him out of office. Because his constituents have not voted him out of office, they may have informally authorized this form of representation of others. Now, some of the represented may eventually validate Frank’s claims (or not) at a reasonable future date, but for some unelected representatives, such as the more ephemeral Bono, the time for such validation may pass quickly. Certainly, this would require taking a hard look at the conditions that make this plausible, but I believe this is a fertile field for investigation, and may open the way for newly relevant reconstructions of John Locke’s concept of tacit consent.  Representing Future Generations The theory of representation I have developed may also help to illuminate the conditions under which future generations are legitimately represented by existing generations. Dennis Thompson argues that political theorists and public officials have not addressed the problems of “representing the unrepresentable – those who are bound by laws but cannot have a voice in making them” (2010, 18). The problem for democratic theory is not only one of temporality – representing future generations – but also one of boundary – representing those who are bound by laws but who do not have a voice in making them. In this dissertation, I have addressed the problem of boundary by arguing for the importance of forming constituencies on the basis of the affected interests principle, where interests reflects conditions of self-development and selfdetermination. If present generations are making decisions that will affect the self-development and/or self-determination of future generations – as they do, for example, whenever decisions are made that involve environmental damage or the sustainability of social security – then that future generation is constituted as the affected constituency with respect to these issues.  187  Following the affected interests principle, decisions should be made that are responsive to those affected future generations, although, of necessity, those generations cannot be included in the decision-making process. Extending the framework I develop here begins to provide some guidance – at least in framing the problem. I would argue that present generations may legitimately represent future generations on the same grounds that surrogate self-appointed representatives provide representation for affected constituencies: their responsiveness to affected constituencies. Surrogate representation describes a representative relationship in which an actor is responsive to the affected constituency but receives authorization from and is accountable to a different constituency – in this case, to present generations, who may act as a proxy, sympathetic constituency for affected future generations. The surrogate self-appointed representative might still achieve an important good for the affected, and is therefore preferable to no representation – but, of necessity, it cannot be democratic in the sense of empowered inclusion of future generations. Logically, there cannot be a direct relationship of authorization by and accountability to the affected constituency, and so the representative cannot enable the empowered inclusion of the affected. However, as I explained in Chapter Six, this kind of selfappointed representative does not interfere with the possibility that this might be achieved. This is a category of self-appointed representation that anticipates a moment when the affected will be active and agentic – a pre-democratic representation, provided the claim benefits conditions of self-development and/or self-determination. That is, the surrogate representative may underwrite democracy for future generations by preserving their capacities to choose.  188  Representative Systems The contemporary political landscape is complex – so complex that no single kind of representative system is adequate to democracy. This is, in part, why I have suggested that we need to look at whether, and if so how, self-appointed representation complements other forms of representation – particularly electoral representation. If I am right that self-appointed representation can complement electoral representation, then we need to assess not only particular representative relationships and mechanisms within self-appointed or electoral representation, but also to ask how they work together to form a system in which people have a chance to influence the decisions that affect them. A complete theory of democratic representation will need to formulate this question: Does the system provide responsiveness and empowerments in ways that enhance or undermine democratic norms? This question moves beyond this dissertation, though the theory I have developed here points in the direction of assessing representative systems. With the help of the affected interests principle, we can begin to identify those who are affected by collective decisions and then ask more detailed questions about whether or not they are represented in proportion to their “affectedness” – the standard that is implied if we interpret effects on selfdetermination and self-development as the relevant sense of “affectedness.” That is, by focusing on relationships established by representative claim, we can assess whether there is overrepresentation of some groups, while others remain under- or unrepresented. As I have argued in this dissertation, it is possible for complementary or supplementary forms of representation to enhance democracy – in particular when informal representatives such as Barney Frank, and self-appointed representatives such as the Rwanda Women’s Network, function to represent constituencies of people who need representation but who are not currently well served by existing structures and relationships of power. But as I have also  189  argued, it is not necessarily the case that supplementary forms of representation will augment democracy. It is also possible that new kinds of representation will overlay electoral representation, serving to enhance the powers of those who are already well represented, biasing or skewing representation in ways that make the system less rather than more democratic. Indeed, the category of self-appointed representation that I term “skewed representation” anticipates the question of how representation may be layered such that a group that is already well served by existing structures of power is provided with yet another avenue of representation through self-appointed representatives. This category illustrates the necessity of understanding how the various kinds of representative institutions and relationships work as a system – but this kind of analysis will require a more complete theory of democratic representation than the one I develop here.  Concluding Remarks We live in an era of enormous democratic hope and promise. At the same time, it is also a period of democratic crisis with continuing deep, if not deepening, inequalities and exclusions among affected groups. It is often noted that as democracy is spreading and is largely unchallenged, it is also suffering from apathetic citizens, corrupt and untrustworthy leaders, and institutions that are far removed from citizen control. Electoral democracy, though it will remain fundamental to democracy, seems not only inadequate to the challenges of representing those who should have influence in collective decisions, but often coexists with clientelism, corruption, and even – in some countries – effective authoritarianism. Democratic equality requires that those whose basic interests are affected by decisions should have some capacity to influence those decisions. Some democracies and democratic institutions are better than others with respect to democratic equality. Many democracies are  190  considering alternative ways to provide representation, considering changes to electoral systems in order to better reflect the composition of the electorate. However, many citizens remain disaffected and powerless. Further, even when individual democracies adopt electoral systems that better represent affected interests within their boundaries, the nation-state may still be unable to ensure the democratic equality of those who are affected by their decisions but who are not citizens. The nation-state can no longer remain the sole focus of democratic theory or of representation theory. The scope of issues that affect citizens is increasingly global and complex, challenging the sovereignty of nation-states. Regional and transnational institutions have scaled up alongside these issues, but their legitimacy is in question. The political landscape is changing rapidly, but our institutions are creaky – they are often comprised of institutions and mechanisms that were imagined decades and even centuries ago, for societies that were far less complex. Many of these institutions were designed to contain democracy, not broaden its reach (Macpherson 1977). Democratic theory tends to follow history, but its role cannot be to simply justify inherited institutions. It should and can respond to emerging challenges and opportunities with new acts of imagination. These acts of imagination should do what all good political theory must do: imagine what ought to be within the potentialities of what exists. What exists now is extraordinarily rich in possibilities, and it is an exciting time for democratic theory. What I have sought to achieve here is to identify a small but growing and enormously important set of possibilities. In doing so, I hope to add to a new democratic theory – one that is attentive to the present, attuned to realities, but imagines a future in which democracy will be broader, deeper, and better.  191  References Abers, Rebecca, and Margaret Keck. 2009. Mobilizing the state: The erratic partner in Brazil’s participatory water policy. Politics & Society 37: 289-314. Alcoff, Linda. 1991. The problem of speaking for others. Cultural Critique 20: 5-32. Amnesty International. http://www.amnesty.org/en/how-you-can-help (accessed June 18, 2010) Amnesty International. http://www.amnesty.org/en/who-we-are/about-amnesty-international (accessed June 18, 2010) Amy, Douglas. 1996. Real choices/new voices: The case for proportional elections in the United States. New York: Columbia University Press. Anderson, Kenneth. 2000. The Ottawa convention banning landmines. European Journal of International Law 11: 91-120. Andeweg, Ruby B. 2003. Beyond representativeness? Trends in political representation. European Review 11: 147-61. Ankersmit, Frank. 2002. Political representation. Stanford: Stanford University Press. Barber, Benjamin. 1984. Strong democracy. Los Angeles: University of California Press. Barber, Kathleen. 2001. A right to representation: Proportional election systems for the 21st century. Columbia: Ohio University Press. Bartels, Larry M. 1996. Uninformed votes: Information effects in presidential elections. American Journal of Political Science 40: 194-230. Bohman, James. 2007. Democracy across borders: From demos to demoi. Cambridge: MIT Press. Brown, Mark B. 2006. Survey article: Citizen panels and the concept of representation.” Journal of Political Philosophy 14: 203-25. Burke, Edmund. 1792. A letter to Sir Hercules Langrishe. www.ourcivilisation.com/ smartboard/shop/burkee/extracts/chap18.htm (accessed March 27, 2010). Burke, Edmund. 1949 [1774]. Burke’s politics: Selected writings and speeches on reform, revolution, and war. Ed. Ross J.S. Hoffman and Paul Levack. New York: Alfred A. Knopf, Inc. Campbell, Albert Angus, Philip E. Converse, Warren Edward Miller, and Donald E. Stokes. 1960. The American voter. Chicago: University of Chicago Press.  192  Canovan, Margaret. 2005. The people. Cambridge: Polity Press. Canovan, Margaret. 2006. The people. In The Oxford handbook of political theory, ed. John S. Dryzek, Bonnie Honig and Anne Phillips, 349-62. Oxford: Oxford University Press. Christiano, Thomas. 1996. The rule of the many. Boulder: Westview Press. Coehlo, V.S.R.P., and Nobre, Marcos, 2004. Participação e deliberação: Teoria democrática e experiências institucionais no Brasil contemporâneo. São Paulo: Editora. Cohen, Carl. 1971. Democracy. Athens: University of Georgia Press. Cohen, Jean L. and Andrew Arato. 1992. Civil society and political theory. Cambridge: MIT Press. Commission of the European Communities. 2001. European governance: A white paper. Brussels. Dahl, Robert A. 1970. After the revolution? Authority in a good society. New Haven: Yale University Press. Dahl, Robert A. 1989. Democracy and its critics. New Haven: Yale University. Dahl, Robert A. 2000. On democracy. New Haven: Yale University. DATA. http://www.data.org/whydata/ (accessed May 28, 2007). Delli Carpini, Michael X. and Scott Keeter. 1996. What Americans know about politics and why it matters. New Haven: Yale University Press. Disch, Lisa. 2009. Beyond mandate democracy. Paper presented at the conference, “Political representation: New forms of measuring and old challenges,” November 12-13, in Berne, Switzerland. Disch, Lisa. 2010. Rethinking responsiveness. Paper presented at the Western Political Science Association Annual Conference, April 1-3, in San Francisco, CA. Dovi, Suzanne. 2002. Preferable descriptive representatives: Or will just any woman, black, or latino do? American Political Science Review 96: 745-54. Dovi, Suzanne. 2007. The good representative. Malden and Oxford: Blackwell Publishing. Dryzek, John S., and Simon Niemeyer. 2008. Discursive representation. American Political Science Review 102: 481-93. Dunsire Andrew. 1978. Control in a bureaucracy. Oxford: Oxford University Press. Economist, The. 2000. Anti-capitalist protests: Angry and effective. September 21: 85-92.  193  Edwards, Michael. 2009. Civil Society. Cambridge: Polity Press. Eulau, Heinz, and Paul D. Karps. 1977. The puzzle of representation: Specifying components of responsiveness. Legislative Studies Quarterly 2: 233-54. Fraser, Nancy. 1990. Rethinking the public sphere: A contribution to the critique of actually existing democracy. Social Text 25/26: 56-80. Fung, Archon. 2009. The principle of affected interests: An interpretation and defense. Unpublished manuscript. George, Susan. 2004. Another world is possible if…London: Verso Press. Goodin, Robert E. and John Dryzek. 2006. Deliberative impacts: The macro-political uptake of mini-publics. Politics & Society 34: 219-44. Goodin, Robert E. 2007. Enfranchising all affected interests, and its alternatives. Philosophy & Public Affairs 35: 40-68. Government of Ontario. 2005. Smog alert response guide. Ontario Ministry of the Environment: Queen’s Printer for Ontario. Grant, Ruth and Robert O. Keohane. 2005. Accountability and abuses of power in world politics. American Political Science Review 99: 29-44. Greenpeace International. Frequently asked questions. http://www.greenpeace.org/international/en/about/faq/ (accessed July 8, 2010). Guinier, Lani. 1994. The tyranny of the majority: Fundamental fairness in representative democracy. New York: Free Press. Habermas, Jürgen. 1996. Between facts and norms: Contributions to a discourse theory of law and democracy. Cambridge: MIT Press. Held, David. 1996. Models of democracy. Stanford: Stanford University Press. Hirschman, Albert O. 1970. Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Cambridge, Massachusetts: Harvard University Press. Hobbes, Thomas. 2002 [1651]. Leviathan. Ed. A.P. Martinich. Broadview Literary Texts. Honig, Bonnie. 2007. Between decision and deliberation: political paradox in democratic theory. American Political Science Review 101: 1-17. Human Rights Watch. About us. http://www.hrw.org/en/about (accessed June 18, 2010). Human Rights Watch. 2009. Diamonds in the rough. http://www.hrw.org/en/reports/2009/06/26/diamonds-rough-0 (accessed June 18, 2010). 194  Human Rights Watch. What they say about Human Rights Watch. http://www.hrw.org/en/whatthey-say-about-us (accessed June 15, 2010). Human Rights Watch. Who we are. http://www.hrw.org/en/node/75136 (accessed June 18, 2010). Huntington, Samuel. 1991. The third wave: Democratization in the late twentieth century. Norman: University of Oklahoma Press. Iley, Chrissy. 2005. Why Africa needs U2. London Evening Standard, June 28. International Committee of the Red Cross ICRC in action. http://www.icrc.org/web/eng/siteeng0.nsf/html/section_discover_the_icrc!Open (accessed June 14, 2010) Join Red. The (red) idea. http://www.joinred.com/aboutred (accessed June 15, 2010). Keck, Margaret E., and Kathryn Sikkink. 1998. Activists beyond borders: Advocacy networks in international politics. Ithaca: Cornell University Press. Keck, Margaret E. 2003. Governance regimes and the politics of discursive representation. In Transnational activism in Asia: Problems of power and democracy. Ed Nicola Piper and Anders Uhlin. Routledge. Keohane, Robert O. 2002. Political accountability. Paper prepared for the Conference on Delegation to International Organizations, May 3-4, in Park City, Utah. King, Jr., Martin Luther. 1998. The autobiography of Martin Luther King, Jr. Ed. Clayborne Carson. New York: Warner Books, Inc. Laclau, Ernesto. 2007. On populist reason. London: Verso. LaVaque-Manty, Mika. 2006. Kant’s children. Social Theory & Practice 32: 365-88. Lijphart, Arend. 1990. The political consequences of electoral laws, 1945-85. American Political Science Review 84: 481-96. Lijphart, Arend. 1994. Democracies: Forms, performance, and constitutional engineering. European Journal of Political Research 25: 1-17. Lindblom, Charles E. 2002. The market system: What it is, how it works, and what to make of it. New Haven & London: Yale University Press. Linz, Juan J. 1978. The breakdown of democratic regimes: Crisis, breakdown, and reequilibration. Baltimore: Johns Hopkins University Press. Lipset, Seymour Martin. 1959. Some social requisites of democracy: Economic development and political legitimacy. American Political Science Review 53: 69-105. 195  Lipset, Seymour Martin. 1981. Political man: The social bases of politics. Baltimore: Johns Hopkins University Press. Lupia, Arthur, and Mathew D. McCubbins. 1998. The democratic dilemma: Can citizens learn what they need to know? Cambridge University Press. Luskin, Robert C. 1990. Explaining political sophistication. Political Behavior 12: 331-61. Machiavelli, Niccolò. 1994 [1532]. The prince. In Selected Political Writings, ed. David Wootton, 5-80. Indianapolis/Cambridge: Hackett Publishing Company, Inc. Macpherson, C.B. 1977. The life and times of liberal democracy. Oxford: Oxford University Press. Magnette. Paul. 2003. European governance and civic participation: Beyond elitist citizenship? Political Studies 51: 144-60. Mansbridge, Jane. 1999. Should blacks represent blacks and women represent women? A contingent 'yes'. The Journal of Politics 61: 628-57. Mansbridge, Jane. 2003. Rethinking representation. American Political Science Review 97: 51528. Mansbridge, Jane. 2004. Representation revisited: Introduction to the case against electoral accountability. Democracy and Society 2: 1, 12-13. Manin, Bernard. 1997. The principles of representative government. Cambridge: Cambridge University Press. Manza, Jeff and Fay Lomax Cook. 2002. A democratic polity? Three views of responsiveness to public opinion in the United States. American Politics Research 30: 630-67. Mill, John Stuart. 1991 [1861]. Considerations on representative government. In John Stuart Mill: On Liberty and Other Essays, ed. John Gray, 205-467. Oxford: Oxford University Press. Miller, David. 2009. Democracy’s domain. Philosophy & Public Affairs 37: 201-28. Miller, Warren E., and Donald E. Stokes. 1963. Constituency influence in congress. American Political Science Review 57: 45-56. Moral Majority. Moral Majority timeline. http://web.archive.org/web/20080405051544/www.moralmajority.us/index.php?option=co m_content&task=view&id=5&Itemid=29 (accessed July 2 2010). Mwenda, Andrew. 2006. Foreign aid and the weakening of democratic accountability in Uganda. Foreign Policy Briefing: Cato Institute. Nagel, Jack H. 1975. The descriptive analysis of power. New Haven: Yale University Press. 196  National Gay and Lesbian Task Force. http://www.thetaskforce.org/activist_center/take_action (accessed May 18, 2010). NGO Monitor. 2003. Oxfam’s apology. http://www.ngo-monitor.org/article/oxfam_s_apology (accessed July 2, 2010). Nye, Joseph S. 2004. Soft power: The means to success in world politics. New York: Public Affairs Press. O’Hara, Thomas J. 1989. The civil rights restoration act: The role of religious lobbies. Paper presented at the annual meeting of the American Political Science Association, in Atlanta, Georgia. ONE International. Africa and the global climate deal. http://www.one.org/c/international/policybrief/3143/ (accessed September 28, 2009). ONE International. ONE history. http://www.one.org/c/international/about/944/ (accessed July 2, 2010). Oxfam Australia. South Africa: Working together gives support to women affected by HIV. http://www.oxfam.org/en/development/women_affected_hiv (accessed June 14, 2010). Oxfam Canada. http://www.oxfam.ca/what-we-do/campaigns (accessed June 3, 2010). Oxfam International. 2006. NGOs lead by example: World’s international NGOs endorse accountability charter. http://www.oxfam.org/en/news/pressreleases2006/pr060606_ngo_charter (accessed April 1, 2007). Oxfam International. Mission statement. http://www.oxfam.org/en/about/what/mission (accessed July 2, 2010). Oxfam UK. 2000/2001. Oxfam annual reports. http://www.oxfam.org.uk/resources/downloads/reports/report2001.pdf (accessed June 29, 2010). Pateman, Carole. 1974. Participation and democratic theory. Cambridge University Press. Pearse, Hilary. 2008. Institutional design and citizen deliberation.” In Designing deliberative democracy: The British Columbia citizens’ assembly, ed. Mark E. Warren and Hilary Pearse. 70-84. Cambridge: Cambridge University Press. Penketh, Anne. 2008. Unrepentant Mugabe addresses the U.N. The Independent, September 26. Pennock, J. Roland. 1979. Democratic political theory. Princeton: Princeton University Press. Pettit, Philip. 1997. Republicanism. Oxford: Oxford University Press.  197  Phillips, Anne. 1995. Politics of presence. Oxford: Clarendon Press. Phillips, Anne. 1998. Democracy and representation: Or, why should it matter who our representatives are? Feminism and politics, ed. Anne Phillips, 224-240. Oxford: Oxford University Press. Pitkin, Hanna Fenichel. 1967. The concept of representation. Berkeley: University of California. Plotke, David. 1997. Representation is democracy. Constellations 4: 19-34. Przeworski, Adam. 1991. Democracy and the market: Political and economic reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press. Przeworski, Adam, Susan C. Stokes, and Bernard Manin, ed. 1999. Democracy, accountability, and representation. Cambridge: Cambridge University Press. Price, Richard. 2003. Transnational civil society and advocacy in world politics. World Politics 55: 579-606. Rabinder James, Michael. 2004. Deliberative democracy and the plural polity. Lawrence: University Press of Kansas. Rabkin, Jeremy. 1999. International law vs. the American constitution – something’s got to give. The National Interest, Spring: 30-41. RainbowPUSH coalition. http://www.rainbowpush.org/pages/issues (accessed May 18, 2010). Ranson, Stewart. 2003. Public accountability in the age of neo-liberal governance. Journal of Education Policy 18: 455-80. Rebick, Judy. 2005. Seven questions. Seven Oaks Magazine, March 15. Rehfeld, Andrew. 2005. The concept of constituency: Political representation, democratic legitimacy and institutional design. Cambridge: Cambridge University Press. Rehfeld, Andrew. 2006. Towards a general theory of political representation. The Journal of Politics 68: 1-21. Rehfeld, Andrew. 2009. Representation rethought: On trustees, delegates, and gyroscopes in the study of political representation and democracy. American Political Science Review 103: 214-29. Rivken, Jr., David B., and Lee A. Carey. 2000/01. The rocky shoals of international law. The National Interest, Winter: 35-45. Rousseau, Jean Jacques. 1978 [1762]. The social contract. Trans. Judith Masters and Roger Masters. New York: St. Martins Press.  198  Rwanda Women’s Network. About RWN. http://www.rwandawomennetwork.org/about_us.html (accessed June 18, 2010). Rwanda Women’s Network. Polyclinic of hope. http://www.rwandawomennetwork.org/clinic.html (accessed June 18, 2010). Saurugger, Sabine. 2007. Democratic ‘misfit’? Conceptions of civil society participation in France and the European Union. Political Theory 55: 384-404. Savage, Deborah, and Albert Weale. 2009. Political representation and the normative logic of two-level games. European Political Science Review 1: 63-81. Saward, Michael. 2006. The representative claim. Contemporary Political Theory 5: 297-318. Saward, Michael. 2008. Authorisation and authenticity: Representation and the unelected. Journal of Political Philosophy 17: 1-22. Schattschneider, E.E. 1960. The Semisovereign People. New York: Holt, Rinehart and Winston. Schmitter, Philippe. 2009. Re-presenting representation. Government and Opposition 44: 47690. Schumpeter, Joseph. 1947. Capitalism, socialism, and democracy. New York: Harper. Sen, Amartya. 1990. Justice: Means versus freedoms. Philosophy and Public Affairs 19: 111-21. Shapiro, Ian. 1999. Democratic justice. New Haven: Yale University Press. Shotter John. 1989. Social accountability and the social construction of ‘you.’ In Texts of identity, ed. John Shotter and K. Green, 133-51. London: Sage. Skinner, Quentin. 1999. Hobbes and the purely artificial person of the state. Journal of Political Philosophy 7: 1-29. Slim, Hugo. 2002. By what authority? The legitimacy and accountability of non-governmental organisations. Paper presented at the “International Meeting on Global Trends and Human Rights – Before and after September 11,” January 10-12, in Geneva. Sniderman, Paul M., Richard A. Brody, and Philip Tetlock. 1991. Reasoning and choice: Explorations in political psychology. Cambridge University Press. Song, Sarah. 2010. Democracy’s boundaries. Paper presented at the Western Political Science Association Annual Conference. April 2-5, in San Francisco. Southern Christian Leadership Conference. Our history. http://www.sclcnational.org/core/item/page.aspx?s=25461.0.0.2607 (accessed June 15, 2010). Spiro, Peter J. 2002. Accounting for NGOs. Chicago Journal of International Law 3: 161-69. 199  Strolovitch, Dara. 2007. Affirmative advocacy: Race, class, and gender in interest group politics. Chicago: The University of Chicago Press. Thompson, Dennis. 2010. Representing future generations: Political presentism and democratic trusteeship. Political Review of International Social and Political Philosophy 13: 17-37. Tussman, Joseph. 1947. The political theory of Thomas Hobbes. PhD diss., University of California, Berkeley. Tyrangiel, Josh. 2002. Bono’s mission. Time Magazine, March 4. Tyrangiel, Josh. 2005. The constant charmer. Time Magazine. December 19. Urbinati, Nadia, and Mark Warren. 2008. The concept of representation in democratic theory. The Annual Review of Political Science 11: 387-412. Urbinati, Nadia. 2000. Representation as advocacy: A study of democratic deliberation. Political Theory 28:758-86. Urbinati, Nadia. 2006a. Representative democracy: Principles and genealogy. Chicago: University of Chicago Press. Urbinati, Nadia. 2006b. On the value and meaning of political indirectness. Paper presented to the Yale Political Theory Workshop, November 9, at Yale University. Walzer, Michael. 1983. Spheres of justice: A defense of pluralism and equality. New York: Basic Books Inc. Warren, Mark E. 2001. Democracy and association. Princeton: Princeton University Press. Warren, Mark E. 2003. A second transformation of democracy? In Democracy transformed? Expanding political opportunities in advanced industrial democracies, ed. Bruce E. Cain, Russell J. Dalton, and Susan E. Scarrow, 223-49. Oxford: Oxford University Press. Warren, Mark E. 2006. Democracy and the state. In The Oxford handbook of political theory, ed. John Dryzek, Bonnie Honig and Anne Phillips, 382-99. Oxford: Oxford University Press. Warren, Mark E. 2008. Citizen representatives. In Designing deliberative democracy: The British Columbia citizens’ assembly, ed. Mark E. Warren and Hilary Pearse, 50-69. Cambridge: Cambridge University Press. Whelan, Frederick, G. 1983. Prologue: Democratic theory and the boundary problem. In Liberal democracy, ed. J.R. Pennock and J.W. Chapman, 13-47. New York: New York University Press. Wilcox, Clyde. 2000. Onward Christian soldiers? The religious right in American politics. Boulder: Westview Press. 200  Williams, Melissa. 1998. Voice, trust, and memory: Marginalized groups and the failings of liberal representation. Princeton: Princeton University. Wrong, Dennis. 2002. Power: Its forms, bases, and uses. New York: Harper and Row. Young, Iris Marion. 1990. Justice and the politics of difference. Princeton: Princeton University Young, Iris Marion. 1999. Justice, inclusion, and deliberative democracy. In Deliberative politics: Essays on democracy and disagreement, ed. Stephen Macedo, 151-58. Oxford: Oxford University Press. Young, Iris Marion. 2000. Inclusion and democracy. Oxford: Oxford University Press. Young, Iris Marion. 2001. Activist challenges to deliberative democracy. Political Theory 29: 670-90.  201  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0071339/manifest

Comment

Related Items