Globalizing the democratic community

Abstract

This article discusses the problem of global democracy, and why democratic legitimacy seems so difficult to attain at the global level. I start by arguing that the difficulties we experience when we try to widen the scope of democratic governance beyond the boundaries of individual states have nothing to do with the characteristics of global society, but result from the underlying assumption that a political community has to be bounded and based on consent in order for democratic legitimacy to be possible. I then explore how this view came into being, arguing that the perennial paradoxes of democratic legitimacy are little but results of earlier and successful attempts to make the concept of political community coextensive with that of the nation. Finally, I argue that once we let go of the idea that political communities have to be bounded and based on consent in order to qualify as democratic ones, the paradox of democratic legitimacy will look like a category mistake rather than an inescapable obstacle to global democracy.

• cosmopolitanism
• democratic paradox
• legitimacy
• global governance

Today leading political theorists believe that globalization constitutes a threat to modern democracy by undermining its foundations: state sovereignty and national identity.1 Since most of these theorists would like to save democratic institutions and practices without sliding back into nationalist nostalgia, they have explored a variety of ways to widen the scope of democratic governance beyond the boundaries of the state. Yet these efforts have been constantly compromised by what looks like an insurmountable problem. If democratic governance presupposes a community in order to be legitimate, global governance cannot be democratically legitimate since there is no corresponding community at the global level that could bestow it with legitimacy. But this problem is neither new nor specific to the global level: In order for any political authority to be legitimate in democratic terms, it must be based on the actual or hypothetical consent of people or a community. But since the identity of that people or community is difficult to account for in terms themselves democratic, most theories of democratic legitimacy issue in paradoxes that cannot be satisfactorily resolved by modern political theory.2 As Van Roermund eloquently has put this problem, ‘self-representation never seems to capture the self that is representing itself’.3

Since the people cannot decide on their own composition, many political theorists have assumed that democratic community and its boundaries are the outcome of historical accidents and therefore cannot be judged according to any standard of legitimacy.4 While being essential to democratic legitimacy, the political community and its boundaries are themselves outside the purview of normative reasoning. This insight has led to despair among those who argue in favor of global democracy, sometimes to the point of conceding that talk of democratic legitimacy at the global level is pointless in the absence of a world government that first could reconstitute mankind into one single political community. Since such a world government presently seems to be out of reach, global democracy therefore looks equally utopian. Others have responded more optimistically to this lack of democratic legitimacy by trying to find viable substitutes for a global demos, such as a global civil society or an increased responsiveness and accountability among global governance institutions.5 But these latter solutions presuppose that the global realm display some features that could permit common norms to emerge and become institutionalized independently of what goes on at the domestic level. So, while there is no demos to be found at the global level, it is reasonable to expect a global society based on democratic values to emerge from the interplay of global political institutions and those affected by their decisions.

But why is it so hard to make coherent sense of the concept of a global demos? Answering this question is the task of the present article. But instead of trying to solve the problem of democratic legitimacy, I will try to explain how this problem came into being, and, by implication, why it might be less of a problem for global democracy. As I shall argue, the difficulties we experience when we try to widen the scope of democratic governance beyond the boundaries of individual states have nothing to do with the characteristics of global society, but result from the underlying assumption that a political community has to be bounded and based on consent in order for democratic legitimacy to be possible. From this point of view, the perennial paradoxes of democratic legitimacy look like little but residuals of earlier attempts to nationalize the concept of political community by making this concept semantically equivalent to that of the nation, and taking the hypothetical consent of its members to be the source of its legitimacy. This is to say that claims to particularity cannot be justified in universalistic terms, only in terms themselves particularistic. Accordingly, once we let go of the idea that political communities have to be bounded and based on consent in order to qualify as democratic ones, the paradox of democratic legitimacy will look like a category mistake rather than an inescapable obstacle to democratic governance.

Pursuing this argument, I will proceed in three steps. First, I will take a critical look at some contemporary attempts to widen the scope of democratic governance beyond the boundaries of individual states. Second, I will briefly describe how democratic legitimacy became a problem of modern political theory, and why the conventional solutions to this problem issue in paradoxes that have resisted resolution. Third, I will suggest a way of redefining political community that makes it possible to dissolve the paradoxes of democratic legitimacy by suggesting that the only prima facie legitimate demos must be coextensive with mankind as a whole.

1 I

The idea that globalization constitutes a threat to modern democracy can be formulated in at least two ways. First, if we take globalization to bring a virtually unrestricted flow of capital and the reign of market forces at a transnational level, it becomes tempting to focus on the corrosive effects of this on state autonomy. As Held states, ‘[m]odern democratic theory and practice was constructed upon Westphalian foundations. National communities, and theories of national communities, were based on the presupposition that political communities could, in principle, control their destinies.’6 When domestic politicians seek to regain control over national economies, they do so by ceding at least some autonomy to supranational institutions lest they want to loose out completely to the corporate world. Yet such ceding of autonomy comes at a price, since they then effectively move formal authority as well as control over outcomes outside the scope of domestic democratic institutions. What once was within the purview of due democratic deliberation is now more a matter of multilateral agreements between government officials at different levels.7 Deprived of any real power, domestic democratic institutions become increasingly hollow. From this follows two strategic options for the democratically minded: either to increase the independence both from global forces and supranational institutions, or, to opt for democratization of those supranational institutions in order to tame these forces and restore some consensual legitimacy to their decisions. Otherwise nobody is in charge and no one is accountable, and we will have no way left to influence our destiny as citizens.8

Second, if we take globalization to bring a virtually unrestricted flow of information and people at the transnational level, it becomes tempting to focus on the corrosive effects on the identity of national political communities. Transnational flows might compromise the cultural homogeneity of a people, and since it takes a people to constitute the demos necessary for democratic institutions to be legitimate, those transnational flows might subvert the foundations of democracy by pushing cultural plurality to an intolerable limit. In order for a people to govern itself, its members need to know who they are: a people and not just a multitude of strangers.9 The democratically minded again have two options at their disposal: they can either move in a nationalist direction by trying to preserve the uniqueness of their own community against the onslaught of global mobility, or, they can move in a cosmopolitan direction by trying to extend the scope of democracy beyond the boundaries of particular political communities while trying to become as tolerant as possible within each of them.10

Let us disregard the nationalistic option for a moment, and focus on current attempts to globalize democratic governance. Theories of cosmopolitan democracy usually buy into some version of the above diagnosis, and then proceed by rethinking political community in light of these challenges.11 They frequently begin their argument by pointing out that democracy—in the shape we know it—has been closely associated with the nation state. They then argue that if the nation state indeed is about to lose its status as the predominant locus of political authority and community, then the only way to save democracy is by redefining political community in such a way that it can include people irrespective of their citizenship in particular communities. Instead of several mutually exclusive demoi, we need to create one inclusive demos to cater to the demands of popular sovereignty in an increasingly globalized world.12 As Buchanan and Keohane have argued, ‘the most obvious difficulty with this view is that the social and political conditions for democracy are not met at the global level.’13 This being so, since there is ‘no worldwide political community constituted by a broad consensus recognizing a common domain as the proper subject of global collective decision-making’. But nevertheless, in the absence of any kind of community at the global level, the very aspiration to democratic legitimacy would be rather pointless.

But how, then, can a global demos be constructed and justified? Two main ways of constructing a global demos compete in the literature. First, we find the idea that a global demos ought to include all human beings. Each human being should have an equal voice since each serious political concern is likely to be of global scope.14 Second, we find the idea that those who are affected by a particular decision should be included in the demos, so what constitutes the scope of the demos in question will vary with the nature of the decision. Each issue should therefore be settled by those affected by the outcome in each particular case, not by mankind as a whole.15 But as both Näsström and Wendt have shown, justifying these solutions is very difficult, since the transition from our present situation in which political communities are bounded to an unbounded global community of all mankind has to take the present situation into consideration: in order for this new community to enjoy democratic legitimacy, it has to be considered legitimate by its prospective citizens.16 That is, it must be democratically constituted, rather than forced upon them by some global political authority. But this merely begs the question who these citizens are, a decision that itself cannot be settled by any democratic process, since that process then would presuppose exactly what it is supposed to yield.

The second solution is equally problematic, since we then have to face the question of how to determine who is affected and who is not by a particular decision, and this might of course lead to divergent interpretations in each individual case. But if democratic legitimacy is wanted, who is affected and who is not should be settled in ways themselves democratic, that is, by those affected. Ergo: who is affected should be decided by those affected. Thus, both ways of justifying a global democratic community in terms themselves democratic presuppose the prior existence of that community, trapping these attempts to construct a global demos in a vicious circularity.

In response to these difficulties, some authors have tried to find routes to global democracy that rely on other sources of democratic legitimacy, such as deliberation and contestation. One of these goes through transnational institutions. The relocation of political authority to the transnational level might yield decentralized forms of authority that eventually will chime well with a world of territorially unbounded communities. Hopefully, the collective allegiance to the procedures of deliberative democracy would then generate overlapping and constantly fluctuating demoi, each being relative to the issue area at hand.17 From this perspective, we would not need any demos to keep democracy alive at the global level, only a proper differentiation into different spheres of social and political action, and the maintenance of democratic conduct within each of them from the bottom up.18 Another solution would be to accept the existence of a multiplicity of different demoi, and opt for the gradual democratization of the relations between them by strengthening the transnational public sphere and its institutions.19 Yet in both cases, it is difficult to see how and why the allegiance to democratic values and procedures could be safeguarded through the transnational dispersion of political authority, since the warrant of democratic deliberation seems to be some normative authority prior to the structure of authority legitimized by means of the same set of procedures. So rather than finding ourselves lost when it comes to justifying a global demos, we are lost when it comes to justifying the authority necessary to uphold standards of democratic deliberation within as well as across different demoi in democratic terms.

The second route goes through negotiating the paradox of democratic legitimacy. Even grated that not all communities are the outcome of popular consent, the democratic paradox nevertheless becomes inescapable whenever we want to justify these communities and their boundaries in democratic terms. To Benhabib, the way to negotiate the resulting paradoxes is by means of iterations of democratic practice which could allow a given demos to redefine itself in the face of ongoing ‘political contestation in which the meaning of rights and other fundamental principles are reposited, resignified, and reappropriated by new and excluded groups.’20 To Honig, the paradoxes of democratic legitimacy are integral to the possibility of democratic governance and productive of its widening scope beyond the boundaries of individual communities. To her, ‘democracy is always about living with strangers under a law that is therefore alien [and] about being mobilized into action periodically with and on behalf of people who are surely opaque to us and often unknown to us’.21 Thus, the paradox of democratic legitimacy is ‘the condition in which we find ourselves when we think and act politically’.22 But if peoples and political communities owe their existence to the contingencies of history, why should we bother justifying them at all? Worse still, why should democratic practices then be confined to bounded communities thus constituted? Answering these questions will force us to take a closer look at the problem of democratic legitimacy and the paradoxes its solutions give rise to.

2 II

How and why did we end up with the problem of democratic legitimacy? Before answering this question, I think it is important to note that this problem presupposes that democratic communities have to be bounded and based on consent. In a world without boundaries, the boundary problem would not be a problem. In a community without consent, legitimacy would have to be derived from other sources. Thus, if we want to understand why democratic legitimacy is a problem, we should start by asking how the concepts of boundaries and consent once were married in political theory. As I would like to suggest, this particular union was the outcome of a broader trend to nationalize socio-political concepts that went hand in hand with efforts to justify the modern sovereign state. This nationalization implied that the range of reference of socio-political concepts was brought to coincide with the boundaries of the sovereign state, and that their meaningful usage was equally restricted by the imagined necessity of such boundaries: all I can offer is a brief sketch of how this happened in political thought.

To the ancients, democratic legitimacy had been less of a problem. They could largely take the existence of the polis for granted, and if ever in doubt, they could point to the founding authority of a Solon or a Lycurgus, or appeal to the conventions embodied in ancient customs and institutions.23 When democratic forms of government later fell into disrepute, this was largely because of the intrinsic difficulty in determining the scope of the relevant demos without thereby inviting its corruption in a world in which boundless and universal forms of community constituted the given starting point for most attempts to justify political authority. But when democratic ideals started to resurface during the Enlightenment, however, these ancient roads to legitimacy had been blocked by the secular and revolutionary aspirations of that age, and the pitfalls of democratic government well forgotten.24 Without the city–state as the given point of reference and with a universalistic framework still in operation, it was also hard to come up with reasons why democratic governance should be restricted to particular communities, rather than applied to mankind as a whole, irrespective of its division into distinct communities. Hence, to writers like Diderot, Raynal, Paine and Condorcet, the global spread of popular sovereignty was seen as a way of overcoming what they saw as a tragic division of mankind into distinct communities of unequal standing, and the immoral impact this had on their intercourse. As Diderot argued, the general will is universal and ‘forms the rule binding the conduct of an individual towards another in the same society, together with the conduct of an individual towards the whole society to which he belongs, and of that society itself towards other societies … submission to the general will is the bond which holds all societies together’.25 But as Robert Wokler has shown in admirable detail, the paradox of democratic legitimacy arises the moment Rousseau tries to restrict the scope of this general will to a particular community by demanding that the community in question ought to be based on the consent of its members.26 Taking such consent as his starting point, Rousseau discovered that the boundaries of a democratic community cannot be justified in terms themselves democratic, since the people cannot constitute itself ex nihilo. If sovereignty has to derive from the people, and if the people by definition cannot be defined by itself—that is, democratically—then how is it possible for a people to be both ruler and ruled all at once? As he stated the resulting paradox:

For a young people to be able to relish sound principles of political theory and follow the fundamental rules of statecraft, the effect would have to become the cause; the social spirit, which should be created by those institutions, would have to preside over their very foundation; and men would have to be before the law what they should become by means of law. The legislator therefore, being unable to appeal to either force or reason, must have recourse to an authority of a different order, capable of constraining without violence and persuading without convincing.27

While the city republic continued to evoke nostalgic pangs in his imagination, Rousseau had to make his case for democracy from scratch.28 In order for democracy to be possible, there has to be a people united by means of common laws, yet these laws would have to derive their legitimacy from the same people. But how could the people ever be constituted in the absence of a founding authority, and how could the proper boundaries of the political community be drawn without thereby presupposing the existence of that people? Since the above problem could not be solved by logical means, it quickly became a matter of finding a pragmatic solution that catered to the political agenda of the Revolution while concealing its paradoxical character. What Emmanuel de Sieyès did in this respect may seem self-evident to us who have been accustomed to take parts of his solution for granted: he introduced the concept of the nation in order to define the proper boundaries of the political community, thereby also justifying the exercise of popular sovereignty within it. As he asked rhetorically:

[h]ow can one believe that a constituted body may itself decide on its own constitution (…) Power belongs only to the whole … From this it follows that the constitution of a country would cease to exist at the slightest difficulty arising between its component parts, if it were not that the nation existed independently of any rule or any constitutional form.29

To Sieyès, the nation is logically prior both to sovereign authority and to the corresponding demos. As he explains, ‘[t]he nation is prior to everything. It is the source of everything. Its will is always legal; indeed, it is the law itself’.30 By conceptualizing the nation as the original source of political authority, Sieyès could brush the paradox of democratic legitimacy under the carpet. As Näsström has summarized this move, it was a matter of placing the nation rather than the individual in an imaginary state of nature, and spell out the consequences for the inner workings of the political community.31 And as Wokler remarked on the end result, ‘in addition to superimposing undivided rule upon its subjects, the genuinely modern state further requires that those who fall under its authority be united themselves—that they form one people, one nation, morally bound together by a common identity … the modern state generally requires that the represented be a moral person as well, national unity going hand in hand with the political unity of the state.’32 In the French context, it was then left to the next generation to bring the nation into existence through an array of clever propagandistic measures.33

But in the guise it first emerged during the Revolution, the concept of the nation did not presuppose cultural homogeneity or a common identity. To Sieyès, as to many liberals after him, what makes it possible for the people to appear as a unity is not the sameness of the citizens, but rather the fact that the nation is something more than the sum of its parts, irrespective of the individual characteristics of the citizens.34 Not only did this way of defining the political community circumvent the problem of democratic legitimacy as it had been posed by Rousseau, but it had obvious practical advantages compared to competing definitions, since it made it possible to articulate a view of popular sovereignty based on representation rather than on the participation of all citizens.35 Later, in those times and places where the legitimacy of a political community was in doubt, the link between state and nation could be reinforced by appealing to a common culture or the common historical memory of a people.36 Consequently, in many cases, ethnos and demos have become inseparable expressions of the same quest for popular sovereignty and democratic legitimacy.37

At this point, it is common to point out that this transition was greatly facilitated by the fact that writers like Bodin and Hobbes already had justified the principle of indivisible sovereignty and that the territorial framework of its exercise already had been established a long time ago. All that Rousseau and Sieyès had to do was to replace the King with the people as the ultimate source and locus of that indivisible sovereign authority within an already territorially demarcated community. But how was this transition from kings to people carried out within political thought? I think important clues can be found in the ways the concept of a general will was defined and used before Rousseau made it the touchstone of popular sovereignty. For when he distinguishes between a general will and the will of all, he does so by identifying the former with the formal sovereignty of the people as a whole, and the latter with an aggregate expression of individual wills. The general will is never wrong, since ‘the people is never corrupted, but it is often deceived’, while particular will often easily become misguided.38 Now these different wills can only be brought to coincide if individual wills are considered in their individuality, that is, in strict isolation from any other association than the state itself, since such ‘partial societies’ are potent sources of corruption. As Rousseau rephrases Machiavelli's warning against factionalism, ‘if groups, sectional associations are formed at the expense of the larger association, the will of each of these groups will become general in relation to its own members and private in relation to the state’.39 Thus, a viable political community requires that the people consist of nothing but individuals, each standing in an equal relationship to the indivisible authority of their totality. Only then can the differences between individual wills be cancelled out and ultimately be reconciled with the general will through representation or deliberation. Thus the very concept of a general will presupposes that the people is categorically distinct from the individuals that compose it, and hence also that the people thus constituted can act wholly independently of their individual wills, however combined. Now this assumption is hard to reconcile with the idea that the people itself is constituted by a prior contract between its individual members to enter as free and equal members into the political community before they can consent to any sovereign authority, even granted that this authority emanates from their wills at the very same moment they enter into the political community. The tension between the general will and the will of all therefore remains unresolved within the contractual framework as long as the latter is supposed to be a precondition of the former.

But what if the general will actually is a precondition of the will of all? As Foucault has reminded us, before the triumph of modern democracy, there was an art of government that took its object of governance to be a population, and which regarded the happiness of its members as a means to the survival and smooth functioning of the state.40 If we step outside the contractual framework for a moment and unpack some of its underlying assumptions, I think that some clues to how this transition was carried out can be found in the theory of political will which Rousseau borrows from his absolutist predecessor d'Argenson. In fact, d'Argenson furnishes the missing link between the concept of population as an object of governance and the idea of a people able to govern itself. By breaking down the people into individual wills, d'Argenson is able to argue that there is no basic difference between the will of the sovereign and the will of the people, only a numerical distinction between different wills that only can be handled through the use of political arithmetic by the sovereign. Through this investigation, writes the Marquis solemnly, ‘I hope to show that popular administration can be exercised under the authority of the Sovereign, without diminishing the public power which it instead increases, and that this is the source of happiness of the people’.41 In order to bring about this outcome, the sovereign must learn how to control the manifestations of particular wills at different levels of society, this royal control sometimes includes giving people latitude to deliberate and act independently whenever it is suitable from the perspective of the sovereign. As a consequence, the sovereign will strengthen his power, benefit the community, as well as get an edge in international affairs.42 So perhaps we are forced to admit that modern democracy is a manifestation of a prior will to govern, a will that first constitutes the people as an object of government and then turns it into a subject of government in order to legitimize itself.

Now this foray into the prehistory of modern democratic theory does nothing to solve the problem of democratic legitimacy, but it does helps us understand a few things better. First, it makes us aware that the problems faced by democratic communities today cannot be blamed on globalization, but rather are to be found at the very origin of modern democratic theory. The paradox of democratic legitimacy has been around since democracy was nationalized, and the paradigmatic way of handling this problem since then has been to use the concept of the nation—however defined—in order to square the circle and brush the question of what makes nations legitimate under the carpet. The revolutionary concept of the nation was created precisely in order to furnish democratic governance with legitimacy, to the same extent as popular rule itself was necessary in order to justify the existence of indivisible sovereign authority within bounded political communities. Second, the above account helps us understand why emancipation from this state of affairs today is perceived as so urgent by so many. If the revolutionary solution consisted in substituting the nation for the individual in an imaginary state of nature, this move had the inevitable side effect of actually realizing that nasty state of affairs among political communities. In order to enjoy the benefits of democracy domestically, we have had to accept that sovereign authority ultimately is justified with reference to the state of exception prevailing in the international realm precisely as a consequence of democratic legitimacy. Therefore, it seems as if the revolutionary solution to the problem of democratic legitimacy has backfired, since the cash value of the above arrangement is that mankind now oppresses itself—in a perversely democratic way—by consenting to remain confined into particular communities whose bounded character also is the very condition of possible warfare between them. Hence, democratic governance at the domestic level is an obstacle to be overcome if we want to emancipate ourselves from the costly illusion of being human by virtue of being members of a ‘people’, as well as from the corollary and even costlier reality of being stuck in an international state of war.

3 III

But how can we escape this predicament? Ironically, our situation with regard to the problem of democratic legitimacy is not unlike that of Rousseau, insofar as his solution is as irrelevant to us as those of the ancients were to him. We no longer live in a world in which bounded communities remain the predominant loci of political authority and the ultimate sources of human belonging, and the way in which these once were justified today only makes sense as a source of nationalist nostalgia. The way in which boundaries are drawn and peoples defined cannot be justified with reference to theories of democracy, since they presuppose exactly that which stand in need of justification. This insight has led to a widespread cynicism with regard to the possibility of justifying democratic governance at any level, since it is tempting to argue that all communities ultimately owe their existence to more or less violent relocations of political power rather than to the consent of their members. If this is the case, political authority is nothing more than power having been around long enough to become taken for granted by those subjected to it, and peoples and boundaries are but outcomes of successful efforts to homogenize an arbitrary multitude of human beings into citizens. By implication, our theories of democratic legitimacy are but ideologies designed to conceal the violence of such founding acts and their consequences.43

Many people believe that this is what is happening today at the global level as well. If this pattern were to repeat itself at the global level, this would entail that questions of legitimacy only could be meaningfully posed if and when a global structure of authority has been firmly established. 44 This implies that the creation of a global demos would require a prior concentration of power at the global level in order to be possible. Since there is no global culture or common historical memory which could provide the symbolic foundations of a global political identity or citizenship, the creation of a single demos of a planetary scope would require allegiances to particular political communities to be gradually undone through a global process of homogenization. Only when this process has been completed, global political institutions can start to enjoy legitimacy by commanding consent among its members.

But such domestic analogies merely make us forget what made them possible, namely conceptual nationalization. Through these processes, the meaningful usage of the concept of democracy was restricted to bounded political communities, and democratic legitimacy within such communities was supposed to derive from the consent of their members. But would it be possible to make coherent sense of democracy in the absence of both boundaries and consent? I think an affirmative answer to this question becomes possible when we realize what makes both boundaries and consent possible. Such claims to particularity are only possible against the backdrop of characteristics that are universally shared by human beings, yet these characteristics themselves do not confer any automatic legitimacy upon such claims. That claims to particularity have to be justified in universalistic terms also make them reversible, since these claims equally well could be contested on the same universalistic grounds. The same set of reasons used to legitimize a particular people or community in terms of consent could equally well be used to dispute its legitimacy on grounds of its boundless contestability.

In fact, before the process of nationalization gained momentum, the predominant way of understanding political community in Western political thought was by regarding mankind as one immanent and universal community, by virtue of the sociability of its members. The idea that consent ought to constitute the only legitimate foundation of authority was closely connected to the ambition to nationalize the concept of political community, insofar as such consent also was essential to the identity of the political community. But the idea of consent derived from the very same sources as human sociability: the universal human capacity to form social bonds by means of the use of language and reason. But as long as human sociability provided the foundation of most attempts to legitimize political authority, it was hard to come up with reasons why political communities should be bounded or based on consent, given the fact that the capacity to form social bonds by means of language and reason are innate characteristics of all members of the species. All the way from the Stoics to Kant, such assumptions of a universal community of all mankind provided the starting point of much Western political and legal thought, as well as for subsequent critiques of despotism, imperial expansion, and colonial exploitation.45 While many of those universalistic conceptions of human community are difficult to defend in scientific and secular terms, they provided the conceptual foundations for subsequent attempts to legitimize particular peoples and communities in terms of consent, as well as for most attempts to contest the legitimacy of the same peoples and communities in universalistic terms.

Such universalistic conceptions of political community might contain some of the things we need in order to make coherent sense of democracy in the absence of a unifying global authority or a common global culture, without reducing humanity to a mere multitude of unencumbered selves. This being so, since universalistic conceptions of community understand mankind to be a unity categorically distinct from the mere sum total of its individual members, constituted not by their sameness but rather by their radical diversity. However different each people or community might be from each other, they nevertheless share the attribute of being different in common, which is the condition of their basic unity. From this point of view, communities of lesser scope are but instantiations of a shared capacity among human beings to form social bonds by means of the use of language and reason, rather than manifestations of particular principles or reason or expressions of particular linguistic communities. This entails that the basic modern requirement of democratic legitimacy—the existence of bounded communities based on the consent of their members—must be seen as the outcome of a prior differentiation of mankind that is essentially contestable since it is not based on the universal consent of all mankind but on historical contingencies alone. Thus, what has to be justified in democratic terms is not the existence of this or that particular people and the boundaries they have drawn around themselves, but the very division of mankind that has made such claims to particularity possible in the first place.

So instead of asking under what conditions globalization might bring about a transition to global democracy, and how the desired end result of such a transition might be justified, I think we should reverse the thrust of the entire argument. Such a reversal might help us understand why the existence of particular communities and their boundaries has been so hard to justify in democratic terms, once we realize that the members of different demoi share some characteristics in common that are essential for the formation of any political community of whatever scope and size. If we are willing to admit that mankind as a whole ought to be considered the ultimate source of legitimacy by virtue of these shared capacities for social life, the burden of proof would no longer rest with those who argue in favor of considering mankind as a single demos. Rather it would rest with those who argue that any people or community could enjoy legitimacy independently of the rest of mankind, not only since each member of the former necessarily also is a member of the latter, but also since these capacities themselves are universal.

Thus it also becomes plain why democratic governance must be global in scope in order for democratic legitimacy to be possible, and why the paradox of democratic legitimacy is a category mistake rather than a genuine logical paradox. For democratic governance to be legitimate in terms themselves democratic, all claims to particularity must be open to contestation at the global level before democratic communities of lesser scope and size can be considered democratically legitimate. Those issues that must be settled either by or with reference to mankind as a whole if the outcome is to be legitimate in democratic terms thus concern whether this or that particular people or community are legitimate sources of political authority and hence also entitled to sovereignty. All such claims would ultimately therefore have to be settled with reference to the contestability of the community in question. For a community and its boundaries to be contestable in practice means that barriers – legal as well as cultural – to entry and exit are low or non-existent. Thus, the easier it is for members to exit and non-members to enter and remain within a given community, the more democratically legitimate it is, as well as conversely. This implies that the existence of a global demos at least has to be assumed before claims to legitimacy by any people or community can be conclusively settled in terms themselves democratic. And this leads to the conclusion that all particular claims to democratic legitimacy must be considered invalid in principle until they have been opened to such contestation. Until then the legitimacy of each particular people or community will remain wholly contingent on the historical accidents that brought them into being, and they will therefore remain wholly provisional sources of democratic legitimacy.

Now most of those who have wrestled with the paradox of democratic legitimacy have resisted this obvious conclusion. The logical difficulties we encounter when we try to justify particular claims to democratic legitimacy indicate that these claims simply are invalid on their own terms, and are based on lingering nationalist intuitions rather than on logical analysis. This is not to say that all particular peoples or communities necessarily lack democratic legitimacy, only to say that such claims will have to be evaluated against a framework that takes mankind as a whole into consideration, since a global demos is the only demos that could enjoy prima facie democratic legitimacy. Nor is this to say that all boundaries between communities necessarily are illegitimate, only to say that the question of their legitimacy can only be settled democratically with reference to the wider community of all mankind. Hence, to put it simply, democracy has first to become global before it can be a democratically legitimate form of governance at any other level.

Acknowledgements

I would like to thank Nicholas Greenwood Onuf, Sofia Näsström, Anne-Kathrine Nyborg, Eva Erman, Hans Agné, Ulrika Mörth and Anders Uhlin for their valuable comments on earlier drafts of this article.

Notes

¹James Tully (2002) The unfreedom of the moderns in comparison to their ideals of constitutional democracy, Modern Law Review, 65 (2), 204–228; Pierre Rosanvallon (2006) Democracy past and future. New York, Columbia University Press.

²Marc G. Doucet (2005) The democratic paradox and cosmopolitan democracy, Millennium: Journal of International Studies, 34 (1), 137–155.

³Bert van Roermund (2003) Sovereignty: unpopular and popular, in: Neil Walker (Ed.), Sovereignty in transition. Oxford, Hart, 41.

⁴Sofia Näsström (2007) The legitimacy of the people, Political Theory, 35 (5), 624–658.

⁵See for example John Keane (2003) Global civil society? Cambridge, Cambridge University Press; Andrew Kuper (2004) Democracy beyond borders: justice and representation in global institutions. Oxford, Oxford University Press; Allen Buchanan & Robert O. Keohane (2006) The legitimacy of global governance institutions, Ethics and International Affairs, 20 (4), 405–437.

⁶David Held (1999) The transformation of political community: rethinking democracy in the context of globalization, in: Ian Shapiro & Casiano Hacker-Cordón (Eds), Democracy's edges. Cambridge, Cambridge University Press, 86–97, 84–111.

⁷Jan Aart Scholte (2000) Globalization. A critical introduction. London, MacMillan, 132–158; Kjell Goldmann (2001) Transforming the European nation-state: dynamics of internationalization. London, Sage, 74–106; Saskia Sassen (1996) Losing control? Sovereignty in an age of globalization. New York, Columbia University Press, 51–58.

⁸Daniele Archibugi (2002) Demos and cosmopolis, New Left Review, 13, 24–38; David Held (1998) Democracy and globalization, in: Daniele Archibugi, David Held, & Martin Köhler (Eds), Re-imagining political community. Oxford, Polity Press, 11–27; David Held (2003) Cosmopolitanism: globalization tamed? Review of International Studies, 29, 465–480.

⁹Jan Aart Scholte, Globalization. A critical introduction, 159–183; Kjell Goldmann, Transforming the European nation-state, 107–125; Zygmunt Bauman (1998), Globalization: the human consequences. Cambridge, Polity Press, 55–76.

¹⁰See for example Veit Bader (1997) The cultural conditions of transnational citizenship: on the interpenetration of political and ethnic cultures, Political Theory, 25 (6), 771–813; Jonathan Seglow (1998) Universals and particulars: the case of liberal cultural nationalism, Political Studies, 46 (5), 963–977; Pheng Cheah (1998) Given culture: rethinking cosmopolitical freedom in transnationalism, in: Pheng Cheah & Bruce Robbins (Eds), Cosmopolitics: thinking and feeling beyond the nation. Minneapolis, University of Minnesota Press, 290–328; Bonnie Honig (1998) Ruth, the model émigré: mourning and the symbolic politics of immigration, in: Pheng Cheah & Bruce Robbins (Eds), Cosmopolitics. Thinking and feeling beyond the nation. Minneapolis, The University of Minnesota Press, 192–215; David A. Hollinger (2001) Not universalists, not pluralists: the new cosmopolitans find their own way, Constellations, 8 (2), 236–248; Stuart Hall (2002) Political belonging in a world of multiple identities, in: Steven Vertovec & Robin Cohen (Eds), Conceiving cosmopolitanism: theory, context, practice. Oxford, Oxford University Press, 25–31; Seyla Benhabib (2002) The claims of culture: equality and diversity in the global era. Princeton, Princeton University Press; Seyla Benhabib (2004) The rights of others: aliens, residents, and citizens. Cambridge, Cambridge University Press.

¹¹Andrew Linklater (1998) The transformation of political community: ethical foundations of the post-westphalian era. Cambridge, Polity Press; Richard Bellamy & Dario Castiglione (1998) Between cosmopolis and community: three models of rights and democracy within the european union, in: Daniele Archibugi, David Held, & Martin Köhler (Eds), Re-imagining political community. Oxford, Polity Press, 152–178; Rainer Bauböck (2002) Political community beyond the sovereign state, supranational federalism, and transnational minorities, in: Steven Vertovec & Robin Cohen (Eds), Conceiving cosmopolitanism: theory, context, practice. Oxford, Oxford University Press, 110–138.

¹²See for example David Held (1995), Democracy and the global order: from the modern state to cosmopolitan governance. Cambridge, Polity Press; David Held (1998) Democracy and globalization, in: Daniele Archibugi, David Held, & Martin Köhler (Eds), Re-imagining political community. Oxford, Polity Press, 11–27; David Held (2005) Principles of cosmopolitan order, in: Gillian Brock & Harry Brighouse (Eds), The political philosophy of cosmopolitanism. Cambridge, Cambridge University Press, 10–27.

¹³Buchanan & Keohane, The legitimacy of global governance institutions, 416.

¹⁴Daniele Archibugi & David Held (1995) Cosmopolitan democracy: an agenda for a new world order. Cambridge, Polity Press; Linklater, The transformation of political community, 193–212; Daniele Archibugi (2004) Cosmopolitan democracy and its critics: a review, European Journal of International Relations, 10 (3), 437–473.

¹⁵Daniele Archibugi (1998) Principles of cosmopolitan democracy, in: Daniele Archibugi, David Held & Martin Köhler (Eds), Re-imagining political community. Oxford, Polity Press, 198–228.

¹⁶Sofia Näsström (2003) What globalization overshadows, Political Theory, 31 (6), 808–834; Alexander Wendt (1999) A comment on held's cosmopolitanism, in: Ian Shapiro & Casiano Hacker-Cordón (Eds), Democracy's edges. Cambridge, Cambridge University Press, 127–133.

¹⁷John S. Dryzek (1999) Transnational democracy, Journal of Political Philosophy, 7 (1), 30–51.

¹⁸Molly Cochran (2002) A democratic critique of cosmopolitan democracy: pragmatism from the bottom-up, European Journal of International Relations, 8 (4), 517–548.

¹⁹James Bohman (2004) Republican cosmopolitanism, Journal of Political Philosophy, 12 (3), 336–352; James Bohman (2005) From demos to demoi: democracy across borders, Ratio Juris, 18 (3), 293–314.

²⁰Seyla Benhabib (2007) Another cosmopolitanism. Oxford, Oxford University Press, 70.

²¹Bonnie Honig (2001) Democracy and the foreigner. Princeton, Princeton University Press, 39.

²²Bonnie Honig (2007) Between decision and deliberation: political paradox in democratic theory, American Political Science Review, 101, 1–17.

²³See for example Cornelius Castoriadis (1983) The greek polis and the creation of democracy, Graduate Faculty Philosophy Journal, 9, 79–115; Paul Veyne (1983) Les grecs ont-ils connu la démocratie? Diogène, 124, 3–33; Christian Meier (1990) The greek discovery of politics. Cambridge, MA, Harvard University Press.

²⁴See Richard Bourke (2008) Enlightenment, revolution and democracy, Constellations, 15 (1), 10–32.

²⁵Denis Diderot (1992) Political writings. Cambridge, Cambridge University Press, 21.

²⁶Robert Wokler (1975) The influence of Diderot on the political theory of Rousseau, Studies on Voltaire and the Eighteenth Century, 82, 55–111.

²⁷Jean-Jacques Rousseau (1990) The social contract and discourses. London, Everyman's, 216.

²⁸Christopher Kelly (1987) To persuade without convincing: the language of Rousseau′s legislator. American Journal of Political Science, 31 (2), 321–335.

²⁹Emmanuel Joseph Sieyès (1963) What is the third estate? London, Pall Mall Press, 129.

³⁰Sieyès, 124.

³¹Näsström, What globalization overshadows.

³²Robert Wokler (1998) The enlightenment and the french revolutionary birth pangs of modernity, in: Johan Heilbron, Lars Magnusson, & Björn Wittrock (Eds), The rise of the social sciences and the formation of modernity. Dordrecht, Kluwer, 22–40.

³³For two excellent accounts, see David A. Bell (2003) The cult of the nation in France: inventing nationalism. Cambridge, MA, Harvard University Press, 1680–1800; Sudhir Hazareesingh (2004) The Saint-Napoleon: celebrations of sovereignty in nineteenth-century France. Cambridge, MA, Harvard University Press.

³⁴See Lucien Jaume (2003) Citizens and state during the French revolution, in: Quentin Skinner & Bo Stråth (Eds), States and citizens. Cambridge, Cambridge University Press, 131–144.

³⁵Istvan Hont (2005) Jealousy of trade: international competition and the nation-state in historical perspective. Cambridge, MA, Harvard University Press, 447–528.

³⁶See Erik Ringmar (1998) Nationalism: the idiocy of intimacy, British Journal of Sociology, 49 (4), 534–549; Andreas Benhke (1997) Citizenship, nationhood and the production of political space, Citizenship Studies, 1 (2), 243–265.

³⁷See Bernard Yack (1998) The myth of the civic nation, in: Ronald Beiner (Ed.), Theorizing nationalism. Albany, SUNY, 103–118; Bernard Yack (2001) Popular sovereignty and nationalism, Political Theory, 29 (4), 517–536.

³⁸Rousseau, The social contract and discourses, 203.

³⁹Rousseau, 204.

⁴⁰Michel Foucault (1988) The political technology of individuals, in: Luther H. Martin, Huck Gutman, & Patrick H. Hutton (Eds), Technologies of the self. London, Tavistock, 145–162.

⁴¹Marquis de Argenson (1787), Considérations sur le Gouvernement Ancien et Présent de la France, Comparé avec Celui des Autres États. Liege, C. Plompteux, 2.

⁴²de Argenson, 22–25.

⁴³Compare Jacques Derrida (1992) Force of law: the mystical foundation of authority, in: Drucilla Cornell, Michel Rosenfeld, & David Gray Carlson (Eds), Deconstruction and the possibility of justice. London, Routledge, 3–67.

⁴⁴Thomas Nagel (2005) The problem of global justice, Philosophy and Public Affairs, 33 (2), 113–147, 146–147.

⁴⁵See for example John M. Headley (2002) The universalizing principle and process: on the west's intrinsic commitment to a global context, Journal of World History, 13 (2), 291–321; Sankar Muthu (2003), Enlightenment against empire. Cambridge, MA, Harvard University Press.