A key task of democratic theory in an age of global interdependence is to retrieve and reconstruct core concepts and normative principles from the long history of democratic thought and practice. The need for reconstruction arises from the fact that democratic theory in the modern period developed primarily as an account of the political legitimacy of territorially bounded sovereign states. Many of the defining principles of democratic legitimacy, such as the consent of the governed and equal rights under law, have long carried unquestioned presuppositions about the territorial boundedness of democratic political orders. Although these principles provide potent standards for evaluating the legitimacy of political orders, their built-in presupposition that the principles can only be realized within territorially bounded states makes them less useful for criticizing, resisting, and transforming forms of political and economic power that exceed the boundaries of states.

Recent work in democratic theory has made significant progress in returning to central concepts in the history of democratic thought, reconstructing them so as to free them from their territorial presuppositions, and using them to gain critical purchase on the question of what it would mean to render existing global power relations more legitimate from a democratic point of view. Two principles of democratic legitimacy – the All-Affected Principle and the All-Subjected Principle (hereafter AAP and ASP) – stand out as especially important contributions to democratic theory in a global age. Both principles have deep roots in the history of Western political thought generally, and in the history of democratic thought in particular.

In this chapter, I argue that we can fruitfully historicize recent work on the AAP and the ASP as reconstructions of two distinct traditions in the history of democratic thought. The AAP, I suggest, reconstructs the intuition that legitimate government requires the rational consent of the governed, an idea that lies at the heart of the liberal tradition of democratic thought. The ASP, by contrast, is better understood as a reconstruction of the republican ideal of equal freedom under and through law. And yet useful as these concepts are for criticizing and reforming already constituted structures of power beyond the state, such as supranational institutions and transnational corporations, they fall short as criteria of democratic legitimacy because they lack a normative account of democratic agency. This shortcoming is twofold. First, the two principles become effective as instruments of democratic legitimacy only when they are taken up by democratic collective agents as principled grounds for resisting or overturning unjust or illegitimate sites of decision-making power. Advocacy for affected interests or subjected persons may produce more egalitarian outcomes, but it is not democratic unless advocates are accountable to those they claim to represent.¹ Second, and more fundamentally for my purposes in this chapter, the democratic legitimacy failures of our age go well beyond the undemocratic character of existing institutional orders. The problem is not only that already constituted orders are undemocratic, but that in many key domains we lack any institutionalized capacity to address the urgent collective action problems we face as a consequence of globalization, such as climate change and rampant inequality. Addressing the democratic legitimacy failures of our age requires an account of how new political orders can be constituted in a democratically legitimate way.

Turning back to the history of democratic thought, we find that the doctrine of popular sovereignty was an important device by which to explain how democratically legitimate powers of binding collective decision can come into being. Popular sovereignty, however, is a doctrine tied at a deep conceptual level to the form of the territorially bounded modern state, and hence not useful for making sense of the possibility of democratic collective agency across borders. I turn to a different concept, that of constituent power, as an alternative resource for rethinking democratic agency in a way that liberates it from the territorial presuppositions embedded in the concept of popular sovereignty. Although constituent power and popular sovereignty are often linked tightly to one another in the history of democratic thought, I argue that this link is not unbreakable, drawing evidence from the history of the idea of constituent power from the medieval and early modern periods and from the thought of the French revolutionary thinker, Emmanuel Sieyès. Having reconstructed the idea of constituent power as a “deterritorialized” way of understanding democratic collective agency, in the chapter’s conclusion I reflect briefly on the relationship between the AAP, the ASP, and constituent power as jointly necessary principles of democratic legitimacy.

Deterritorializing Consent: The All-Affected Principle

The AAP expresses the common intuition that “individuals should be able to influence decisions that affect them.”² Stated in such an abstract and general way, the principle is insufficient as a standard for evaluating decisions, since virtually every action has some effect on others, and it would be an unreasonable constraint on individual autonomy if we had to woo the consent of every actually and potentially affected other before acting.³ Democratic theorists have been refining the principle so that it can serve as a defensible standard for establishing and evaluating mechanisms for holding a wide variety of collective agents (state and nonstate, public and private) responsible for the significant and enduring effects of their decisions on others.⁴ Taken together, these arguments constitute an advance in democratic theory in an age of heightened global interdependence because they let go of the assumptions embedded in earlier formulations of the principle: that territorial states were the appropriate context for its application, and citizens of such states were the appropriate bearers of entitlements to have their interests taken into account in decision-making processes. As Ian Shapiro argues, “[t]he causal principle of affected interest suggests that ideally the structure of decision rules should follow the contours of power relationships, not that of memberships or citizenships: if you are affected by the results, you are presumptively entitled to a say.”⁵

To my knowledge, no one has yet written a conceptual history of the AAP. Nonetheless, some scholars have marked out some of the starting points and pivotal moments in the history of this idea. As Melissa Lane notes in her chapter for this volume, it makes sense to trace the idea back to a Roman law precept compiled as part of the Justinian civil code: quod omnes similiter tangit, ab omnibus comprobetur, “What touches all similarly must be approved by all.”⁶ Gaines Post has chronicled the evolution of quod omnes tangit from a principle of private law to a procedural principle of public law in the medieval period, showing that it gradually became a procedural principle of consent and a secular source of legitimate political authority.⁷ By the fourteenth century the principle was explicitly linked to the legitimacy of taxation and to the claims for representation in conciliar, administrative, and judicial proceedings.⁸ It became a key tenet of reform in the age of democratic revolutions, as in the American revolutionary slogan, “No taxation without representation.” And from there, it is not difficult to see the connection to interest-group pluralist theories of democracy, in which the equitable representation of potentially conflicting interests becomes a defining criterion of democratic legitimacy.⁹

Read in this way, the AAP captures the normative core of the liberal tradition of democratic thought, in which democratic institutions, together with basic civil and political rights, are seen as instrumentally valuable for the protection of fundamental interests (rather than as intrinsically valuable). Scholars have recently adapted and amended the principle so that it can function as an instrument for evaluating, criticizing, and reforming governance institutions in the global age. As Sofia Näsström has argued, the AAP “has done much work to detach the ideal of democracy from its conceptual reliance on the nation state.”¹⁰

Deterritorializing Equal Freedom under the Law: The All-Subjected Principle

Some democratic theorists have argued that the ASP is a better way to conceptualize democratic legitimacy in the global age than the AAP, and we can trace a similar project of retrieval and reconstruction in their work. Arguing that the AAP is too indeterminate to serve as a stand-alone criterion of democratically legitimate governance,¹¹ these theorists turn to the ASP for greater specificity in identifying the group of persons who should be empowered with civil and/or political rights in relation to an order of rule.

The normative core of the ASP is that a coercive order of rule is legitimate only if it recognizes and secures the status of all subject to it as free and equal persons.¹² As Nancy Fraser states,

Because there are cross-border, suprastate and transnational configurations of political power, the demos, read as the people subject to coercive power, can also be understood to exceed the boundaries of citizenship within a territorial state. On one side, the ASP prescribes that non-citizens subject to the coercive authority of states also be recognized as bearers of basic rights against domination that states have a duty to realize and protect.¹⁴ From another angle, the ASP seeks to recognize non-state forms of institutionalized power as bearing a duty to recognize and protect the rights against domination of those who are subject to them.¹⁵

As with the AAP, different contemporary theorists have offered different interpretations of the ASP. Here, I highlight two, one wide and one narrow, to suggest that while they have in common a reconstruction of the normative core of the republican tradition of democratic thought, each retrieves a different strand of republicanism in order to render it usable for thinking about democracy in an era of globalization. On the wide reading of the ASP, it retrieves the idea of democratic freedom as collective self-legislation. On the narrow reading, the ASP reconstructs the republican ideal of freedom as nondomination, an idea that underwrites but does not entail the wider view. On the wider view, democracy is intrinsically valuable because participation in the exercise of collective autonomy is constitutive of the individual autonomy of those subject to binding law. On the narrower view, democracy is instrumentally valuable because it empowers individuals to contest both public and private forms of domination.¹⁶

My purpose here is not to try to adjudicate between different interpretations of the ASP, but to suggest that it is illuminating to read the contrasting interpretations of the ASP as reworkings of different strands within republican traditions of thought that loosen them from their territorial presuppositions. Following Philip Pettit’s distinction between the “Italian-Atlantic” and “Franco-German” traditions of republicanism,¹⁷ we can read the narrow interpretation of the ASP as the retrieval of an element common to both traditions, whereas the wide interpretation retrieves an idea that is distinctive to the Franco-German tradition of republicanism. The core of the narrow interpretation of the ASP can be summed up as the principle of freedom as nondomination. The second can be summed up as the principle of freedom as self-legislation (auto-nomia) that underwrites the Rousseauean/Kantian tradition of popular sovereignty.

As Pettit has argued throughout the development of his neo-republican project, the origins of the idea of freedom as nondomination lie in Roman republican thought, and in particular the distinction between the free person (liber) and the slave (servus). To be a slave is to live according to the arbitrary will of a master, that is, under another’s rule or dominium; to be a free person is to be publicly recognized as an equal under law, that is, a citizen, and not subject to another’s will (civis).¹⁸ The purpose of republican legal order is to secure each citizen’s nondominated status. Law is not a will that stands over and against individuals’ wills as a master, but secures the equal freedom of all citizens through a mixed constitution and by protecting the contestatory powers of citizens so that they can defend themselves against both private and public domination.¹⁹ The narrow reading of the ASP, in which subjection to coercion generates a right to contest the order of law to which one is subject, loosens this tradition of freedom as nondomination from its longstanding presupposition that legal citizenship within a polity is what gives a person standing to contest the coercive power of its laws. Instead, buttressed by the idea of human rights, which recognizes all human beings as persons with standing to claim equality under the laws to which they are subject, the ASP extends the principle of freedom as nondomination to anyone who is subject to any coercive order. This expansion does not entail that all who are subjected to a legal order in any respect have a right to participate in the lawmaking process, but that they have a claim to contest any coercive law that unjustifiably restricts their freedom.

The wide interpretation of the ASP, in which being subject to coercive law generates a right to participate in the making of law as its equal co-author, ties back into a different tradition of republican thought in which Rousseau and Kant are key figures. In that tradition, freedom as nondomination is reinterpreted not only as independence from the arbitrary will of another, but as autonomy, that is, being the self-originating source of the law by which one is bound. Its locus classicus is Rousseau’s idea of moral freedom as stated in the Social Contract: “obedience to a law one prescribes to oneself.”²⁰ This reflexive understanding of freedom-as-independence is read into the civic realm as the idea of collective autonomy, in which individuals’ obligation to obey law can be reconciled with individual autonomy just insofar as they participate as co-equal members of the self-legislating community. Law is not the imposition of an alien will, but a self-imposed restriction on individuals’ own arbitrary wills, aimed at securing the possibility of realizing collective goods that would be impossible to realize if individuals pursued only their separate interests. In Abizadeh’s reconstruction of the ASP as a principle of collective self-legislation, he strips away the deeply rooted presupposition that the self-governing demos must be bounded by common culture, nationhood, or historically settled territorial borders, arguing that since borders themselves are coercively enforced, those excluded by them are subject to the coercive power of the states concerned and of the state system as a whole. This idea transforms the ASP into a regulative ideal that can never be fully met in practice, but which is nonetheless analytically clear as a standard for evaluating any actually existing coercive order, including territorial borders.²¹

Deterritorializing Democratic Agency: Constituent Power

My purpose thus far has been to make explicit what contemporary theorists are doing when they debate and refine theoretical articulations of the AAP and the ASP as resources for rethinking the possibilities and criteria for democratic legitimacy under conditions of globalization. I have argued that we can understand their inquiries as efforts to rework familiar ideas from the long history of democratic thought in order to make them usable for analyzing and criticizing existing structures of political power, and for generating normative insights into how these structures might be rendered more legitimate from a democratic point of view. As the contributors to this volume demonstrate, the AAP and the ASP provide rich argumentative resources for identifying the democratic deficits of a wide range of contemporary global structures: immigration regimes,²² global governance institutions,²³ international trade and its impact on labor,²⁴ international philanthropy,²⁵ climate change,²⁶ and so on.

Taken together, these arguments make a strong case that the AAP and the ASP are valuable (and perhaps necessary) criteria of democratic legitimacy. It also makes sense, as several theorists have argued, to view them as complementary to one another rather than as rivals.²⁷ But the question remains whether they are sufficient to ground a full or adequate account of what democratic legitimacy entails. I believe they are not. The AAP and the ASP provide valuable insights into the normative constraints that must be placed on the exercise of power in order for it to count as legitimate. But democracy requires more than constraints on the exercise of political power; it also requires the capacity to generate political power, that is, the capacity to produce binding (i.e. coercive) collective decisions aimed at advancing common interests. As Jane Mansbridge has argued, democratic theory and activism have long been preoccupied with the important task of resisting illegitimate forms of coercive power. Invaluable as this “resistance tradition” has been for making political power more legitimate by rendering it more democratically accountable, an almost exclusive emphasis on resistance has led to the underdevelopment of theories of democratically legitimate collective action.²⁸ This neglect takes on a particular urgency under circumstances of globalization, when domestic and international or transnational capacities for collective action are less and less of a match for the complex problems that arise from increasing global interdependence and mutual affectedness.

Viewed from this angle, the AAP and the ASP are valuable reworkings of elements in the venerable resistance tradition of democratic theory, but offer limited resources for understanding the character and origins of democratically legitimate collective action. If the democratic deficits of a globalized world arise not only from the illegitimate use of coercive power, but also from the lack of collective agency, then the AAP and the ASP are not adequate for a thorough-going diagnosis of the demands of democratic legitimacy under circumstances of globalization.

In the history of democratic thought, the concept of popular sovereignty has performed important work as an account of democratically legitimate collective agency. It is perhaps for this reason that several contemporary theorists have characterized popular sovereignty as a principle of democratic legitimacy that stands in a complementary relationship to the AAP and/or the ASP.²⁹ Yet in an age of globalization, where the cross-border effects of state decisions are increasingly visible, it is clear that the principle of popular sovereignty as a form of democratic agency at the scale of the state is not adequate to secure the democratic legitimacy of the state system as a whole. The AAP and the ASP offer criteria by which to clarify the external legitimacy constraints on the democratic agency that is realized through states whose internal legitimacy is grounded in the idea of popular sovereignty.

The idea of popular sovereignty, however, like the idea of democratic self-determination to which it bears close kinship,³⁰ may be impossible to extricate from the presupposition that democratic collective agency can only be realized within territorially bounded political communities – if not states, then state-like jurisdictions that enjoy considerable autonomy in relation to other polities.³¹ It is for this reason that some theorists have argued against the idea that democracy is possible, or even conceivable, beyond the scale of the territorial state.³² Others have argued that if democracy is realizable beyond the scale of the state, it will be through a divided or multilevel form of popular sovereignty institutionalized through nested territorial jurisdictions, as may eventually be possible in the European Union.³³ These reconstructions of popular sovereignty as a way of reconceiving democratic collective agency beyond the state remain tied to the presupposition that democratic agency can only be exercised by territorially bounded demoi. This supposition is based, in turn, on the idea that only a collectivity whose members see themselves as participants in a common, stable, durable, and substantial system of cooperation is capable of sustaining a project of collective self-legislation, and hence of democracy properly so-called.³⁴

The idea that collective political agency in the form of self-legislation is possible only among people who see themselves as bound to one another in relations of interdependence or mutual affectedness into the foreseeable future carries a great deal of intuitive good sense. The question I wish to explore in the remainder of this chapter is whether it is possible to extricate this idea from the presupposition that shared occupancy of a determinate territory is a necessary presupposition of democratic collective agency. In other words, is the concept of popular sovereignty, read as the power of collective self-legislation of a territorially bounded demos, the only imaginable form of meaningfully democratic agency? Or is there a normative core to the idea of democratic agency expressed by the concept of popular sovereignty that it is possible to extricate and reconstruct in a manner that liberates it from its territorial presuppositions, much as the AAP reconstructs the normative core of quod omnes tangit and the ASP reconstructs the normative core of equal freedom under law?

I believe that there is in fact a normative core to the idea of popular sovereignty that can be retrieved and reconstructed in this way: the idea of constituent power, that is, the rightful authority that people have to freely associate with one another in jointly establishing the law-governed order that will regulate their relations with one another and through which they will generate binding collective decisions aimed at serving their common interests. Through much of the history of modern Western thought, the concept of popular sovereignty and that of constituent power have been read as nearly synonymous terms. Constituent power – the power to constitute an order of government, to create a constitution – is often read as the power that properly belongs to the popular sovereign as the ultimate authority on which legitimate order is based.

It would take a far more detailed historical treatment than is possible in a brief essay to trace the concepts of constituent power and popular sovereignty through the history of Western thought, let alone the complex relationship between them. Other scholars have done this work far more thoroughly than I can hope to do here.³⁵ Instead, I will briefly sketch two moments in the history of the concept of constituent power to show that it is unquestionably possible to read it as independent of the concept of sovereignty, and hence of the concept of popular sovereignty. First, the concept of constituent power long antedated the concept of popular sovereignty, and indeed the concept of the sovereign state. Second, the modern thinker most strongly associated with the concept of constituent power, Emmanuel Sieyès, embraced it at the same time that he rejected the concept of sovereignty as having a proper place in his theory of political legitimacy.

Although the concept of constituent power is usually associated with the age of democratic revolutions, Daniel Lee has recently shown that the idea has much older roots in the history of European thought.³⁶ Lee traces the evolution of the concept of constituent power from sixteenth-century figures such as Donellus and Brutus (who used the concept to ground a claim of the people’s authority over the king) back to the much older doctrine of lex regia, which dates back to late Roman thinkers. The fiction at the core of the lex regia was that the people of the Roman republic had transferred their legislative authority to the emperor;³⁷ in the medieval period, the doctrine was mobilized again to make sense of the (de facto) authority of free cities to establish their own orders, free from interference from the Holy Roman Empire, kings, or lords. Here, the doctrine combined with the concept of a “free people” (populus liber) within Roman ius gentium, in which allies of the Roman empire were treated as having the right to govern themselves according to their own laws. Medieval jurists used these concepts to extend the logic of lex regia to all free peoples.³⁸

For the purpose of retrieving and reconstructing the idea of constituent power as a way of conceptualizing democratic agency, it is worth highlighting two points from this quick sketch of a long history. First, the idea of constituent power as the foundation of democratic authority is not inextricably tied to the idea of the territorially bounded sovereign state. This is clear from the simple fact that the normative core of constituent power was expressed well before the emergence of the idea of the sovereign state, in the form of arguments on behalf of the right of free cities to govern themselves. Second, the history shows that the idea of “the people” that possesses the right of self-legislation is indeterminate; the concept of a populus liber was variously used to denote the people of the Roman republic, the free peoples in the Roman law of ius gentium, the citizens of medieval city-states, and, eventually, the people of the territorially bounded modern state.³⁹ The character and composition of “the people” on whose behalf this potent idea has been mobilized across the centuries is not singularly identifiable with the people of the modern territorial state, and there is no reason in principle why the idea cannot be mobilized for new constructions of “peoplehood” in the twenty-first century. Although the concept of popular sovereignty is inextricably linked to the concept of the sovereign territorial state, the concept of constituent power is not.

The separability of the concept of constituent power from that of popular sovereignty is also evident from the thought of Sieyès, the person most often credited with distinguishing the constituent power of the people from the constituted authority of the state. According to Sieyès, the legitimacy of ordinary positive law rests on its having been promulgated in accordance with constitutional laws fixing the organization and powers of legislative and executive governance bodies. The legitimate authority of the constitution, in turn, issues from its consonance with the will of the people (or, in Sieyès’ term, “the nation”). Constitutional laws, he wrote, “are said to be fundamental, not in the sense that they can be independent of the national will, but because bodies that can exist and can act only by way of these laws cannot touch them. In each of its parts a constitution is not the work of a constituted power but a constituent power.”⁴⁰ Sieyès’ theory of constituent power has a number of features that are problematic from the standpoint of a democratic theory committed to egalitarian inclusion.⁴¹ Nonetheless, two features of his argument stand out as useful resources for a contemporary reconstruction of constituent power as democratic agency. The first is that Sieyès explicitly distances the concept of constituent power from the concept of sovereignty, including popular sovereignty. The second is that although at points he does describe “the nation” as a quasi-natural and pre-political subject, a more nuanced reading of his text shows that what makes the demos as a potential collective agent is not a common history, culture, or language but a dense network of social and material interdependence that is the objective condition of possibility for the formation of a subjective sense of collective purpose and agency. For Sieyès, what transforms a latent or potential “nation” or “people” into a political agent is an act of representation through which a plurality of individuals can be reimagined as a unified collectivity capable of acting jointly toward their common good.

On the first point, Sieyès explicitly criticizes the concept of sovereignty because he regards it as tied to absolutism. The decision to form a constitutional order was a choice to live under a system of law that would secure the freedom of each, which on his view (as in the Roman republican tradition) depended upon a mixed constitution in which no element reigns supreme. “It is a mistake,” he argued, “to talk of the sovereignty of the people as if it had no bounds.”⁴² In a later writing, pace Schmitt’s interpretation, he rejects the political theology with which the concept of sovereignty is bound.⁴³ As Lucia Rubinelli has recently argued, Sieyès “never relied on the notion of sovereignty to describe the principle of the people’s power,” and instead substituted the concept of constituent power for that of sovereignty.⁴⁴

The significance of Sieyès’ rejection of sovereignty for this chapter’s project of retrieval and reconstruction is straightforward. These passages demonstrate that one can embrace the concept of constituent power while rejecting the absolutism and decisionism that is constitutive of many theories of sovereignty, including the most common readings of Hobbes and Rousseau and, certainly, Schmitt’s politico-theological account of popular sovereignty. These passages show clearly that for Sieyès the constituent power of “the people” did not entail that the will of the people was self-validating. Rather, a legitimate political order is always constrained by principles respecting the equal freedom of its individual members.

The second key point I wish to retrieve from Sieyès’ theory of constituent power is that his conception of the people as a collective agent is not best understood as pre-political. In the opening chapter of “What is the Third Estate?”⁴⁵ Sieyès defines a nation not as a people defined by shared history, language, or culture but as “a body of associates living under a common law, represented by the same legislature, etc.” In other words, the people as a collective agent is itself constituted by its members’ joint decision to live together under a common order of law which they share a role in making. Sieyès’ description of the activities that make a society are explicitly materialist and based on his analysis of the social division of labor. The heart of Sieyès’ critique of the existing order is that the very people whose productive activities and relationships jointly create the conditions for a flourishing social life are excluded from a role in the political order under which they live.

Yet the objective reality of the interdependence and cooperative activities of individual members of society is not in itself sufficient to make them a political agent. Sieyès invites his readers to imagine a three-stage process by which the multitudinous participants in these common social and economic activities become a political unity capable of collective action. The first stage is simply the emergence of a will among “a more or less substantial number of isolated individuals seeking to unite.” “This fact alone,” he writes, “makes them a nation.” At this stage, it is the convergent wills of individuals who see the advantage of acting together that constitutes a collectivity who make their association itself the object of their work.⁴⁶

The second stage is one of deliberation through which associates’ several individual wills are forged into a “common will.”⁴⁷ Here Sieyès explicitly introduces the concept of power as something that is generated only through a process of common will formation:

The third stage in Sieyès’ account of the formation of collective agency is representation. For Sieyès, the formation of political agency should be understood as a part of the larger social division of labor. The inconvenience, inefficiency and, in a society with large numbers of people, the impracticability of assembling to form a common will through deliberation generates the need to “entrust” the common will “to the exercise of some of their number.”⁴⁹ This trust, he emphasizes, does not mean that the community “divest[s] itself of the right to will,” and the appointed delegate has no authority to “alter the limits of the power with which it has been entrusted.” Yet the character of the common will is transformed by the shift from deliberation to representation: “it is no longer a real common will that acts, but a representative common will,” which can only ever be an incomplete and limited expression of the common will.⁵⁰ In other words, the authority of the representative to act on behalf of the community is always qualified by the fact that there is a gap between what the members of the association truly want and the representation of their common will by their delegates. Because of this gap, the possibility remains open that the community, whose constituent power has set up the system of representation, may reclaim its authority if its current delegates have misinterpreted its will.

We find, then, in the locus classicus of the concept of constituent power, the basic elements of a methodologically and normatively individualist theory of group agency⁵¹ for a particular kind of group – one that seeks to bring into being, where it did not exist before, a law governed order that treats all who live under it as free and equal persons, over the longue durée.⁵² Sieyès’ theory provides the basic conceptual resources for separating democratic collective agency both from the concept of sovereignty and from the supposition that territory provides the relevant material underpinnings for the formation of a democratic collective agent.⁵³ Rather, it is the coming-into-consciousness of ongoing relations of material interdependence that generates the first movement toward the formation of joint intentions to identify common interests and, through deliberation, forge a common will. This common will is not a general will in Rousseau’s sense, as it acknowledges the ongoing plurality of its constituent members. Sieyès’ theory also recognizes that in the moment of transition between the constituent power and the constituted power – the moment of representation – there is always some violence done to this plurality. The representation of the collective agent as a unity always leaves a remainder, which is why the constituent power is not extinguished by the creation of the constituted power, and why claims to represent “the people” must always remain open to contestation.⁵⁴

Conclusion

I have argued that recent works aimed at specifying the AAP and the ASP as criteria of democratic legitimacy for a global age can be understood as endeavors to retrieve and reconstruct much older ideas in the history of democratic thought, liberating these ideas from democratic theory’s long-standing entanglement with the usually unexamined presupposition that democracy is possible only within territorially bounded forms of political community. The AAP and the ASP are potent rearticulations of normative principles that all forms of political power, and not only those centered in territorially bounded states, must meet if they are to claim even a modicum of democratic legitimacy. Any structure of power that does not satisfy these criteria fails to treat as equals those human beings who are affected by its decisions or subject to its coercive power. Since the equal moral worth of all persons as such is the sine qua non of any conception of democracy, no political order that fails these tests can validly claim to be legitimate from a democratic point of view.

Yet both the AAP and the ASP fall short of a thoroughgoing account of democratic legitimacy because both are focused on appropriate normative constraints on political power, and not on the democratically legitimate conditions under which power, understood as a capacity for binding collective decision, can emerge. I have suggested that both are reconstructions for a global age of elements in the long-standing resistance tradition of democratic theory, where the principal concern is to restrain the illegitimate exercise of political power. In the era of globalization, there has been a proliferation of forms of political and economic power that is not constrained to track all affected interests or the equal rights under law of all subjected. The AAP and the ASP provide useful analytical toolkits by which to diagnose these normative deficiencies of the current global order. They track the objective facts of the matter about whose interests are adversely affected and whose equal standing as a subject of lawful or unlawful subjection is being violated by contemporary arrangements of political power. But these principles operate, more or less, from a juridical point of view. They are addressed to the normative strengths or weaknesses of already constituted powers, but offer little insight into the injustices that result from the absence of collective decision-making capacity in those domains where a power vacuum not only reproduces an unjust status quo but, as Mansbridge emphasizes, amplifies the drift of the complex global system toward outcomes that are unquestionably disadvantageous for the vast majority of human (and non-human) beings.⁵⁵ In order to understand the potentials for democratic legitimacy under conditions of globalization, we need, as well, a better understanding of the possibility of democratic agency which, like the AAP and the ASP, is unmoored from the presupposition that democracy is possible only within territorially bounded political communities. We need to understand how democratically legitimate forms of binding collective decision-making capacity can be generated in domains where it does not already exist.

Through this chapter’s provisional reconstruction of the idea of constituent power as democratic agency, I have sought to show that we need not hold onto the supposition that the powers of collective self-legislation are available only to territorially bounded political peoples. The democratic peoplehood of territorial states will undoubtedly be an important resource for democratic agency for some time to come, as the institutionalization of collective agency through elections, projects of constitution making, and the like is still concentrated at the level of the territorial state. Moreover, the citizen empowerments that are crucial to democratic mobilization and will formation – rights of expression, association, and participation – are now institutionalized only at the scale of the state, and these are crucial instruments for leveraging political influence at other scales of politics.

The link between democratic collective agency and the powers of territorial states is a historically contingent phenomenon, not written into the concept of democracy as the self-rule of the people. In principle, there are only two constraints on the formation of democratic collective agency. The first is that there are enduring objective conditions of social and material interdependence among the people. In the absence of lasting social and material conditions of interdependence, which Sieyès characterized in terms of a social division of labor, there is no clear reason why individuals should strive to forge an association united around common interests over a sufficiently long term to warrant the establishment of a durable order of democratic self-rule. In addition, there must be a subjective consciousness of the fact of interdependence potent enough to motivate diverse and dispersed individuals to form an association around their common interests and make these common interests a site of collective self-legislation. They must discursively represent, first to themselves as consociates in a shared process of material and social production, and later to others who are also implicated in this process, their character as participants in a common social project. This moment of discursive representation, the articulation of an imagined political relationship between a multitude of individual actors as parts of a larger social whole that can and should be made democratically legitimate, is a condition of possibility for the political representation of common interests and the constitution of new powers of binding collective decision.

The boundaries of “the people” as the collective subject of constituent power, then, are constrained but not determined by either territorial boundaries or the facticity of material relations of social interdependence. Objective conditions of interdependence form one limit of the possibility of constituent power as democratic agency. The history of the post-Westphalian system of territorially bounded states carries the consequence that material relations of interdependence are especially strong at the scale of the state. Legal regimes of property, labor, taxation, and redistribution remain concentrated at the scale of the state, and state policies around economic development continue to exert enormous influence over the future prospects of individuals within state jurisdictions. But the fact of globalization has generated material and social relations of interdependence that cross state boundaries, much as the colonial policies of European states historically crossed the boundaries of ethno-cultural peoplehood to generate global divisions of labor in which some classes but not others were represented in the decisions by which the benefits of economic codependence were distributed.

Historically, the idea of constituent power has commonly been associated with revolutionary moments and moments of constitutional founding. This is a mistake; such moments are important instances of constituent power, but they do not exhaust the category. Democratic agency as constituent power is much more common than revolutionary moments. It exists wherever individuals freely associate with the purpose of instituting an institutional order that is capable of generating binding collective decisions aimed at advancing common interests. Whether or not it succeeds in instituting a new order or reconstituting an existing one, the essence of constituent power is contained in the joint intention to form a democratically legitimate order.

In concluding, let me briefly return to the proposition, noted earlier, that we should understand basic principles of democratic legitimacy as complementary to one another rather than as rivals. Whereas other theorists have suggested that the AAP and/or the ASP should be understood as complementary to the principle of popular sovereignty, I want to suggest that we should read them as complementary to the principle of constituent power. Indeed, the AAP, ASP, and constituent power can be understood as jointly necessary and mutually constraining principles of democratic legitimacy. An element of constituent power is internal to the democratic bona fides of the AAP and the ASP. A decision-making process cannot take affected interests into account or preserve the equal freedom of those subject to its decisions unless those interests and persons are represented as having a valid claim within the process. The representation of interests and persons, in turn, is not democratic unless it reflects the self-understanding of the represented as to the interests they have at stake in a given process. If the interests of the affected or subjected are represented from a juridical point of view, based on an analysis of the objective facts in a given context of decision making, the resulting decision may be just but it cannot properly be called democratic. Conversely, an exercise of constituent power, aimed at creating institutions capable of generating binding collective decisions, is democratic to the extent that it treats as equals all persons who are included in the collectivity. However, democratic constituent power is not legitimate if it does not take into account the interests that are significantly affected by its exercise, or the claim to equal freedom of those who fall subject to the institutions it establishes, even if those interests or persons fall outside the boundaries of the collective agent. In other words, the AAP and the ASP can be read as side constraints on the legitimate exercise of constituent power.

The argument advanced in this chapter proceeds at a regrettable level of abstraction. Ideally, I would turn to some illustrative cases to show how this retrieval and reconstruction of the idea of constituent power enables us to see, as forms of democratic agency, political formations that have arisen across borders in the global era. The climate change movement is one instructive example: the cross-border mobilization of diverse people who understand themselves as bound to one another by the shared human condition of vulnerability to climate change, and act jointly toward the goal of instituting a global order of binding rules that would limit climate change and address its effects.⁵⁶ Other examples include the transnational movement of Indigenous peoples and the United Nations Declaration on the Rights of Indigenous Peoples that their mobilization generated, and the transnational peasants’ movement, La Vía Campesina, which is organized democratically at local, state, regional, and global levels and has made significant progress toward the goal of ratifying a UN Declaration on the Rights of Peasants and Other People Working in Rural Areas. For the moment, my hope is to have contributed to the larger project of reworking democratic theory for a global era through the retrieval and reconstruction of old ideas in the history of democratic thought.