Can the people exercise constituent power?

Abstract

Contemporary constitutional theorists sometimes use the phrase “the constituent power of the people” in a way that is, on closer examination, ambiguous. It could mean that the people is the bearer of constituent power, that the people exercises constituent power, or both. This article examines this pivotal, yet rarely explicitly thematized, distinction internal to the concept of constituent power and considers its downstream implications for constitutional theory. The proposition that the people is the bearer of constituent power, I argue, is best read narrowly as a claim about the proper subject of attribution for major constitutional change. The proposition that the people exercises constituent power, however, is best read either as (i) a claim about the capacity of citizens to effect constitutional change through collective deliberation, or (ii) shorthand for the claim that representatives should always engage in processes of constitutional change on behalf of citizens. If these readings are true, the article concludes, then this has important consequences for the theory and practice of constituent power and for its relationship with political representation.

1. Introduction

Constituent power may be defined in general terms as the power to establish, or fundamentally amend, a constitution.¹ On such a broad definition, most modern constitutional settlements would rest on both original and derived exercises of constituent power, regardless of their democratic provenance.² The current resurgence of theoretical interest in constituent power remains largely attributable to normative assumptions about its close relationship with democracy. The concept of constituent power seems to align with a democratic conception of political legitimacy, whether in relation to patently authoritarian regimes or the more subtle forms of elite domination prevalent in liberal democracies. From this perspective, some advocates of constituent power point to its capacity to promote greater participation and openness to change, counteracting the tendency for proponents of liberal constitutionalism to privilege a procedural and representative conception of democracy, and to empower unelected courts to overrule elected legislatures or obstruct the outcomes of popular referendums.³ Yet the idea of constituent power is not only attractive to radical-democratic theorists; it can also provide what appears to be a democratic foundation for liberal-constitutionalist commitments.⁴ Popular sovereignty is a widely accepted source of constitutional legitimation, but lacks a uniform meaning, and is indeed vulnerable on some interpretations to concerns about the manipulation of public opinion by populists and threats to the rights of minorities by “tyrannical” majorities. The proposition that the people is the ultimate source of legitimate public authority because it is the bearer of constituent power is, however, compatible with a broad range of theories, from the view that citizens should retain a permanent capacity to exercise political autonomy as authors of the laws which govern them, to the position that the role of the people is mostly dormant as a “sleeping sovereign.”⁵

The general idea of the constituent power of the people is nevertheless sufficiently broad to encompass both a narrowly attributive and a more robust participatory interpretation. One can see this potential ambiguity, and discern its significance, in the following representative statement:

At least since the French Revolution, the idea of constituent power has been used to indicate the power the people have to create legal-political orders. As such, the history of constituent power is deeply tied to the principle of popular power and, through it, to the history of democracy, to its theory and to its practice [and it also] acts as a reminder that the source of constitutional normativity lies in the will of the people.⁶

The claim that the people have the power to create constitutional orders could mean that the people (i) is the bearer of the constituent power, (ii) exercises constituent power, or (iii) both. At a first approximation, to say that (i) the people is the bearer of constituent power is to identify it as the political or juridical subject of constitutional creation and change. Yet the identification of the people as the bearer of constituent power leaves wide scope for interpretation regarding the level of active participation by citizens in constitution-making and constitution-amending processes and is indeed consistent with a very narrow construal that it involves a retrospective imputation. The claim that (ii) the people exercises constituent power, by contrast, seems to imply that citizens, taken individually or collectively, are the actual agent of constituent power, in the sense (further specified below) that they play an effective role in processes of constitutional change. Crucially, while both (i) and (ii) could be true, the inference from (i) to (ii) is more contentious than the converse. If the people exercises constituent power, then this suggests that it is also the bearer of the power. The path from an identification of the people as the bearer of constituent power to the claim that the people exercises constituent power requires more argument. One could regard the people, for example, as the bearer of constituent power, yet believe that the popular will must always be represented by constituted powers in ways that, given the scale and plurality of modern polities, precludes the collective of citizens from effectively exercising constituent power as agent.

Two prominent historical examples can help to illustrate the distinction and point in the direction of its wider significance. In the monist British constitutional model famously articulated by A.V. Dicey, “the doctrine of parliamentary sovereignty” is the rule that the Crown-in-Parliament “has, under the English Constitution, the right to make or unmake any law whatsoever; and, further, that no person or body is recognised. . . as having the right to override or set aside the legislation of Parliament.”⁷ This model still assumes that constitutional authority ultimately rests on a popular foundation (“political sovereignty”), yet the role of the people as effective constituent agent is muted. In historical terms, one can say that the constituent power of the people was gradually “absorbed into the doctrine of the absolute authority of the Crown-in-Parliament to speak for the British Nation.”⁸ On the monist model there is accordingly no active exercise of constituent power beyond “ordinary” representation, even though one can ascribe “political” or popular sovereignty to the people as bearer. If one turns to the 1787 Philadelphia Constitutional Convention and the subsequent ratification of the US Constitution, the people can also in this case be regarded as the bearer of constituent power, insofar as its authority legitimates the activity of a convention called into operation by the state legislatures for the specific purpose of constitution-making.⁹ Significantly, the most important distinction for both of these examples is not between the active exercise of constituent power by the people and a representative exercise of constituent power. The examples rather illustrate different models of constructing the relationship between the people as the bearer of constituent power and the active exercise of constituent power by representatives. Most obviously, in the US case, in contrast to the British, there is a clear differentiation of the role of a constituent convention from that of “ordinary” representatives.

Contemporary constitutional theorists do not always clearly mark the distinction between the people as the bearer and the people as the agent of constituent power, and its explanatory and normative significance is often understated in the literature. Many theorists have argued that a robust “classical,” or “Schmittian,” model of constituent power, whereby a unified people acts as an “external” dynamic extra-legal force prior to any constituted political structure, is conceptually incoherent, normatively undesirable, and ill-suited to explain contemporary constitution-making processes.¹⁰ A more general and domesticated idea of the constituent power of the people nonetheless remains prevalent in attempts to explain processes of constitutional creation and change. An acknowledgement of these two points motivates greater attentiveness to the institutional mechanisms for processes of constitution-making—which are invariably representative—and their normative constraints. Yet a lack of analytical clarity regarding the distinction between the people as the bearer and as the effective agent of constituent power undermines both tasks. In the first instance, the conflation of questions of attribution and agency can obscure both how constitution-making processes actually work and their conditions of legitimacy. One manifestation of this tendency, I suggest below, is a reluctance directly to confront the implications of the redundancy of “classical” models of constituent power based on a unified people’s will, so that a “romantic” conception of popular constituent agency continues to operate in the background of explanations of constitution-making processes. Relatedly, in a more normative register, the same lack of clarity leads theorists to overstate the requirements on direct citizen participation for the legitimate exercise of constituent power, and conversely to underestimate the significance of representative institutional mechanisms for constitutional change and their associated criteria of legitimacy.

Two recent accounts of constituent power which help to clarify these distinctions are Hans Lindahl’s reflexive theory and Philip Pettit’s neo-republican analysis of the plural and singular people.¹¹ An examination of the bearer/exercise distinction in these and related theories of constituent power, I argue in what follows, raises doubts as to whether any sharp line of demarcation can be drawn between theories which grant the collectivity of citizens an active role in constitution-making processes and those which place a greater weight upon representation. This examination, in fact, strongly supports the thesis that constituent power is best regarded as “a creature of political representation.”¹² Far from it being the case that constituent power is uniquely well-equipped to provide a radical-democratic challenge to premises of liberal constitutionalism, one might add, it can perform an indispensable role of legitimation by “externalizing” the role of elites in constitutional formation towards an imputed subject that is defined by an inability to act.¹³

The article is structured in three sections. Section 2 examines what it means for the people to be the bearer of constituent power and argues for the conceptual separability of this idea from the exercise of constituent power by the people. Section 3 then considers the conditions that would need to be satisfied for the people to exercise constituent power. I argue that—at least outside of highly idealized models—neither a coordinated plurality of citizens nor the people as a unified group agent can exercise constituent power as a political agent. Section 4 then contends that the obvious alternative position, which is that constituent power is invariably exercised by representatives and then attributed to the people as its bearer, more accurately reflects both the way that the concept is applied in constitutional practice and its use in normative argumentation.

2. The people as bearer of constituent power

The proposition that the people is the bearer of constituent power is a predominant assumption in contemporary constitutional theory. Following the decline of monarchical legitimacy, consolidated by the late-eighteenth-century revolutions in America and France, the popular or national will becomes the indisputable source of constitutional authority. With the gradual broadening of the franchise in the nineteenth and twentieth centuries, moreover, the idea of the constituent power of the people can be associated with a broader democratic commitment to the idea that citizens should be ultimate authors of the fundamental laws which govern them. On closer examination, of course, the identity of “the people” and the relationship between the political and juridical dimensions of its role as constituent subject raise difficult questions. This section examines these questions only as they pertain to the status of the people as the bearer of constituent power. I argue that the proposition that the people is the bearer of constituent power is best construed narrowly as a thesis about the proper subject of attribution for constitutional change. Such a conclusion may appear trivial, yet it has significant downstream implications. For the fact that the people is the proper subject of attributions for constitutional creation or change, I demonstrate, is conceptually independent of its capacity to exercise effective constituent agency.

Although the view that the people is the bearer of constituent power may be predominant, given prevailing assumptions about the relationship between popular sovereignty and constitutional legitimacy, it is neither historically inevitable nor conceptually necessary. In the first instance, as Joel Colón-Ríos has demonstrated, there is a long tradition—notably represented by Sieyès—which attributes constituent power to the nation as “an abstract entity with a will of its own,” rather than the people.¹⁴ Sieyès’s position, Colón-Ríos notes, is consistent with the view that constitution-making representatives may depart from the actual preferences of citizens. Colón-Ríos contrasts this with a Rousseauvian approach, which, in attributing constituent power to the people, at least endeavors to extend broader citizen involvement beyond the pouvoir commettant. More fundamentally, there is nothing as a matter of principle or conceptual necessity which entails that constituent power could not be held, for instance, by a constitutional monarch (as perhaps in nineteenth-century Imperial Japan), a bureaucratic elite, or even a military junta. A reluctance to acknowledge this reflects confusion between the concept of constituent power and its normative theorization.¹⁵

Perhaps the more obvious difficulty with the identification of the people as the bearer of constituent power is the equivocity of the idea of the people itself. This equivocity has a long historical lineage. The Roman term populus, for example, from which the English term “people” derives, might refer in the constitution of the Roman Republic narrowly to the plebeian class (in distinction to both patricians and slaves), to citizens in general, or to the whole polity as a “collective entity that transcended specific individuals, classes and generations.”¹⁶ There is a similar ambiguity, to mention one more example, in seventeenth-century English Civil War parliamentary propaganda literature.¹⁷ Here the equivocation ranges from the people as enfranchised property owners to all members of the polity or the commonwealth as they are represented by parliament.

For present purposes, it is instructive to distinguish between more robust ideas of the people as either a substantial unity or group agent and thinner “aggregative” or “pluralist” interpretations.

The most robust conception of “the people” postulates its existence as a substantial unity with a shared intention and will. In contemporary constitutional theory, such a view is often associated with Carl Schmitt’s Weimar-period position that the concrete existence of a people as a political unity is a precondition for its constitutional incorporation.¹⁸ While such a conception of the people has continued political and sociological relevance, especially given populist movements, it is belied by the pluralism and diversity of modern polities and is also, for reasons explored more below, unpersuasive from a perspective internal to the theorization of constituent power.

If one assumes that “the people” must refer in a modern polity to the plurality of citizens, this still leaves open multiple interpretative possibilities.¹⁹ First, “the people” could refer to a concrete multitude, so that the singular term picks out and then aggregates every person identified as a member or citizen of a particular territory or political group.²⁰ Second, “the people” could denote the “active” citizens within a polity who play a role—even if only minor—in its established political institutions. Third, “the people” could refer to a unified corporate subject which, in some sense to be further specified, supervenes on citizens as an aggregate or plurality. This third way of regarding “the people” has a long provenance, and can be traced back to medieval theories of the corporation and the people as a universitas in contrast to a mere collection of discrete individuals.²¹

There is little reason to regard these three interpretations as mutually exclusive alternatives. It is possible, that is, to regard the aggregate of citizens, the active plurality of citizens, and the corporate collective of citizens as different but reconcilable aspects of the modern concept of the people, which are granted priority dependent on the direction of practical or theoretical concern.

This brings us back to the meaning of the proposition that the people is the bearer of constituent power. It is possible, under all three interpretations, to attribute constitutional creation and change to the people as subject. The people conceived as an aggregate of citizens, as the plurality of citizens, and as the citizen body as a whole, can each serve as subject of constituent power in this sense, because all that is required is the capacity to support an imputation or attribution. Yet this leaves totally undetermined—and here is the central claim of the current section—the level of effective involvement by the people as agent in constitutional creation or change. The status of the people as bearer of the constituent power is independent of, and does not predetermine, its active exercise of the constituent power, and this is true on all three interpretations of the people.

One way to motivate this claim about the conceptual independence of (i) the attribution of constituent power to the people as subject and (ii) the active exercise of constituent power is to consider permutations in relation to the three aspects of the people. A skeptic about the capacity of the people to exercise constituent power might point to minimal participation by individual citizens—as an aggregate or plurality—in constitutional change, while defending the juridical status of the corporate people as the bearer of constituent power. A proponent of the capacity of the people to exercise constituent power, by contrast, might point to the involvement of citizens, taken aggregatively or as a plurality, in the determination of constitutional essentials (e.g., through ratification of a plebiscite) and construe the corporate people not only as a juridical subject, but also as a group agent capable of acting with a shared intention and will. This second approach is close to the republican construal of constituent power, as will be demonstrated in the next section.

In short, the proposition that the people is the bearer of constituent power is consistent not only with different (but reconcilable) interpretations of the status of the people, but with quite divergent interpretations of the role of the people as effective agent of constitutional creation and change. The separability of (i) the identification of the people as bearer of constituent power and (ii) the claim it has the capacity to exercise constituent political agency has, I argue in the next two sections, important and often overlooked implications. In closing this section, I introduce some of these implications, while also elaborating on the argument above, by reference to Hans Lindahl’s justly influential “reflexive” account of constituent power as the expression of collective selfhood.

Lindahl’s “reflexive” theory of constituent power seeks to deflate the paradox of constituent power by charting a middle way between normativism and decisionism.²² A normativist interpretation of constituent power, associated with Kelsen, interprets the people as an “ethical political postulate” or juridical subject rather than a dynamic political unity.²³ Kelsen’s normativism, Lindahl suggests, collapses constituent and constituted power, agency and representation, and politics and law.²⁴ A decisionist or Schmittian interpretation of constituent power, however, relies on a substantial conception of the people, which (even apart from normative considerations) is implausible given the internal plurality of modern polities. Both accounts are, moreover, susceptible to the paradox of constituent power. Normativist theories, by externalizing constitution-making agency to the domain of politics or formalizing it as a juridical postulate, suggest that constituent power is either juridically irrelevant or always a form of constituted power. Decisionist theories raise the problem how the people can be a substantial unity absent prior constitutional incorporation. Lindahl’s solution is to argue that “collective self-constitution means constitution both by and of a collective.”²⁵ The “self” of self-constitution is a form of reflexive identity, not identity as sameness, insofar as the exercise of constituent power “signals the self-constitution of political community. . . [as] collective self.”²⁶ The people hence arises as a unity of constitutional law through the very exercise of constituent agency which is attributed to it as bearer. As Kelsen suggested, the unity of the people relies on a post-facto attribution following political constituent activity by a plurality. Yet constituent power, understood reflexively, remains a “dynamic” political phenomenon in which the people expresses its intention and will through the constitution-making activity which constitutes itself as a subject of attribution.

At first blush, the “normativist” dimension of Lindahl’s theory appears to predominate. The retroactivity of collective self-attribution means that the unity of the collective “we” which emerges from the action of a plurality is always a function of self-attribution or “represented”; it is never a “present” unity.²⁷ The individuals who become members of the people as citizens can only regard themselves as a collective in an act that was undertaken for their own sake post facto.²⁸ And while Lindahl seeks to retain a temporal distinction between constituent and constituted “moments,” he renounces a radical foundational moment external or prior to constitutional form. As Lindahl puts it, the “paradox of constituent power entails that the foundation of a novel legal order can only come about as its re-foundation, as the continuation, albeit in a legal guise, of an extant collective.”²⁹ What, then, is the status of this extant collective? If one wishes to retain a “dynamic” dimension to constituent power, then it seems necessary to uphold the political status of the people as an effective collective or group agent which is irreducible to its juridical or formal status as a constituted normative unity. This explains why, in his later elaboration of the reflexive theory, Lindahl engages in some detail with theories of group or collective agency.³⁰ As examined in the next section, however, the extent to which theories of group agency can establish that the people, as the bearer of constituent power, is also capable of its effective exercise is questionable.

Lindahl’s reflexive theory is nonetheless instructive in relation to the distinction between the people as the bearer and as the active agent of constituent power. On close examination, Lindahl’s view that constituent power simultaneously constitutes a people is consistent with the argument above that the attribution of constituent power to the people as its bearer is independent of the capacity of the people to exercise constituent power. The two claims are separable in the sense that the propos­ition that the people actively exercise constituent power does not follow necessarily from the identification of the people as the proper subject to which constituent activity should be attributed. If constituent power is understood as reflexive or relational, indeed, this can motivate the view that those who claim to exercise constituent power act as an already constituted power because, for example, “a constituent assembly or convention authorized to draft a constitution is an already constituted governmental institution.”³¹ This implies that the people, as bearer of constituent power, requires representatives to exercise power on its behalf.

Once one postulates the status of the people as the bearer of the constituent power, there would then appear to be two viable options left on the table. The first possibility is that the will of the people must always be represented when constituent power is exercised, so that the proposition that the people exercises constituent power is best regarded as shorthand for the claim that constituent activity by representatives (generally in an authorized constituent assembly or convention) is to be attributed post facto to the people as its subject. Such a theory motivates well-worn legal analogies of principal and agent, but also invites reflection on the descriptive and normative conditions that would need to be satisfied for it to be true to say that constituent representatives genuinely act on behalf of the people as the proper subject of constituent power.³² Alternatively, one could seek to connect the status of the people as bearer of constituent power with its active exercise through a theory of collective or group agency. It is to this second possibility, elaborated in most detail by Pettit, that I now turn in the third section of the article.

3. The exercise of constituent power

The proposition that the people is the bearer of constituent power does not necessarily entail that constituent power is exercised by the people. What, then, does it mean for the people to exercise constituent power? The most plausible argumentative strategy to establish that the people exercises constituent power, the analysis above suggests, is to appeal to the group agency of citizens. The current section examines this strategy by reference to Pettit’s defense of the capacity for the sovereign people to exercise constituent power. A close analysis of Pettit’s defense and his broader views on popular sovereignty, I argue, speaks against the capacity of the people to exercise constituent power in constitutional democracies, except perhaps as a postulate of ideal theory. This conclusion motivates the alternative view that the proposition that the people exercise constituent power is best viewed as shorthand for the claim that constitution-making representatives who effectively exercise constituent power should be accountable to citizens.

Pettit’s construal of popular sovereignty in terms of citizen group agency is instructive for explicating the bearer/exercise distinction. While the claim that the people is the bearer of constituent power can be read narrowly as a juridical proposition which identifies the subject of attribution for constitutional change, the claim that the people exercises constituent power implies that citizens are the effective political agent of changes to a constitution. Pettit’s account, presented more as a contribution to political philosophy than constitutional theory, addresses the dynamic aspect of constituent power that is suggested by reflexive or relational theories.

Before turning directly to Pettit’s account, it is useful to set out in a preliminary manner some constraints on a satisfactory analysis of the capacity for the people to exercise constituent power. The exercise of a power can be distinguished from the mere opportunity or abstract capacity to exercise a power by the efficacy of the former in the performance of an action (without this entailing, of course, that an exercise of a power must always be successful in achieving a desired outcome; here one can speak of unsuccessful exercise).³³ Speaking in general terms, to “exercise” a power is to be directly responsible for effecting a particular type of action which is directed towards an outcome that is the end or purpose of the exercise of that power. Given this orientation to ends, an adequate explanation of what it means to exercise a power will refer to that power’s success-conditions. The most basic success-condition on an exercise of constituent power is that it directly culminates in constitutional change. For while not all exercises of constituent power need to culminate in constitutional change, it is the establishment of constituted powers or amendment of their terms that is, by hypothesis, the main function served by constituent power.

How “directly involved” do the people need to be for it to be true to say that it is the political agent which exercises constituent power? In the case of a modern written constitution, it would seem too strong a constraint to demand that all citizens be involved in the actual drafting of a constitution. Such work is generally carried out by political representatives and constitutional experts, and it is better understood as the articulation of a prior exercise of constituent power. Yet it would seem equally unsatisfactory to say that constituent power could be exercised through mere participation in the election of representatives to a constituent assembly or convention, or voting in Yes/No referendums or plebiscites proposing constitutional change, let alone the mere opportunity to participate in such processes. This claim is defended in more detail below, but it is prima facie plausible for the following reason. The exercise of a power requires a direct rather than indirect involvement in the effecting of an action oriented towards a particular outcome. In the case of constituent power, the relevant outcome is change to the content of a constitution. For an agent to be directly involved in changes to a constitution implies that they engage effectively with its content. From this perspective, what seems lacking in the case of selecting constituent representatives or voting in plebiscites and referendums is active participation in collective deliberation concerning the content of the constitutional change which is the purpose of the constitution-making activity. In order to exercise constituent power, this is to suggest, the people, as a plurality or unity, would need to at least participate in deliberation about the material content of the constitution, not merely to select the agents who will participate in such a process, or to register simple approval or disapproval of the content of a proposed constitutional change.

A further question is whether, for it to be true to say that the people exercises constituent power, all citizens would need to play an active role or a decisive role, or both, in processes of collective deliberation and decision regarding the content of the constitution. The success-conditions outlined above suggest that both conditions would need to be satisfied for it to be true in the strict sense that the people exercise constituent power: all citizens would need actively to engage with the content of the constitutional change and they would need to play the decisive role in determining its content. While sociologically naïve, it is certainly conceivable as a postulate of ideal theory that all citizens could actively engage in collective deliberation about constitutional change, perhaps through a combination of the above processes, which could be supplemented by structured forums for public debate which inform citizen choices in the decision-making procedures that determine the content of constitutional provisions. Yet even if these constraints can be satisfied in principle, the condition that all citizens play a decisive role in the determination of constitutional content would require more than agreement to participate in, and be bound by the outcome of, a shared decision-procedure. It would also require a properly unified intention, inclusive of all citizens, regarding the determination of the content of the constitution. At this point, in addition to any sociological plausibility concerns, one is confronted with the problem of the doubtful applicability of the group agency model to the determination of constitutional content.³⁴ Whereas the group agency model may seem convincing with respect to the establishment of decision-procedures like voting, it is less well-equipped to explain how the content of a constitution is specified. This is a point best elaborated by reference to Pettit’s theory.

For the people to exercise constituent power, the analysis to this point suggests, either (i) the total plurality of citizens or (ii) citizens as a unified group agent would need to engage directly in processes of deliberation which are decisive in the determination of the content of the constitutional change. Both possibilities are entertained in Pettit’s theory of popular sovereignty.

Pettit’s argument for the agency of citizens as a plurality is expressly presented as a contribution to the theory of constituent power. The initial step in Pettit’s argument is to distinguish between the people as a self-governing plurality of citizens and as a singular group agent or corporation.³⁵ As members of the plural people, citizens can play an active role by electing representatives, by voting in referendums, and by engaging in public debates which may influence lawmaking.³⁶ As members of the singular people, however, citizens can act with a unified intention because they are coordinated under shared decision-making procedures which constitute them as a group agent.³⁷ Pettit’s resolution of the paradox of constituent power is hence to argue that each citizen in a plural people, while insufficiently unified under a shared decision-making procedure to constitute a group agent, can nonetheless share a joint intention to establish a constitutional framework which provides the coordination mechanisms necessary to constitute them as a singular people.³⁸ The status of the plural people as causal agents of constitutional change does not presuppose the higher level of constitutional coordination necessary for the singular people to emerge as a genuine group agent.³⁹ The plural people which exercises constituent power is more than a multitude, because the individuals who comprise it coordinate activity towards the realization of a common goal, but less than the constituted group agent their activity constitutes.

Pettit rejects what he considers the false assumption that causal responsibility for a constitutional order also requires some form of temporal precedence. Both “the constituting and constituted people can be responsible for a constitution that was set up by a particular individual or group, insofar as they have the power to change it and choose not to do so; that is why there can be a democracy without a founding, democratic moment.”⁴⁰ Pettit’s conception of the plural people hence assumes that a constitutional polity is already extant. The constituting plural people is not, on Pettit’s conception, a unified substance that sits outside of a constitutional order as an unmoved mover (as might be suggested by the more extravagant creatio ex nihilo rhetoric of Sieyès or Schmitt), but rather the individual citizens within an existing constitutional order, viewed under the aspect of their plurality rather than as a genuine singular group agent. From an historical perspective, this indeed makes more sense than evocative talk of radical beginnings. If one considers, for example, the two paradigmatic “constituent moments” of the late eighteenth century—post-revolutionary America and France—the constituent activity of members of the constitutional conventions and assemblies rested not insignificantly on the existing political structures of states and Estates in confederation America and ancien régime France, respectively.⁴¹

A further illuminating feature of Pettit’s account is that it clarifies why an extra-constitutional people cannot exercise constituent power. If a shared will is defined as an interconnected network of intentions to engage in a joint activity, accompanied by the awareness of all participants regarding the existence of this network, then it is at least conceivable that citizens incorporated within the constitutional structures of a modern polity could be ascribed a shared will.⁴² Yet the thesis that an extra-constitutional people has the unity of intention, and associated representational and motivational states, let alone the capacity to process those states, which is necessary for a group agent, is very difficult to entertain.⁴³ On standard models of group agency, there is no room for an extra-constitutional constituent agency of the people as effective unity.⁴⁴

It is not immediately apparent, however, exactly how the plural people would be able to exercise constituent power, either. An initial concern is the status of the plural people as an agent of constitutional change. When the people are viewed under the aspect of plural constituent activity, to recollect, one is to suppose that citizens share an intention to establish a new constitution, or reform an existing one, while falling short of group agency. Yet Pettit’s account suggests we are not referring to a temporal founding moment, and must indeed presuppose the unity of group agency which is provided by an existing framework of constituted power. This is to say that a necessary condition for being able to view the people under the aspect of its plurality is that we have always already established the people under the aspect of its unity. It is therefore quite difficult to see how talk of aspects can resolve the paradox here; the shift from the register of temporal preconditions to conceptual preconditions still leaves us with the conclusion that the presupposition—now in the conceptual rather than temporal dimension—for regarding the plural people as a constituent agent is the prior existence of the constituted people as coordinated unity.

This point is supported by the need for a successful exercise of constituent power to rely on procedural decision-making mechanisms. The most common procedural mechanisms are representative conventions and assemblies (inclusive of “round-table” processes which seek to achieve a “consensual” settlement through negotiation of plural viewpoints) and referendums and plebiscites.⁴⁵ Some recent constituent processes have also sought, it is worth noting for completeness, to utilize crowd-sourcing technology.⁴⁶ These various procedural mechanisms for the “channeling” of constituent power can be assessed across descriptive and normative dimensions with respect to their ability to deliver on criteria such as broadness of participation and representation. Notwithstanding differences with respect to these criteria, however, all of the procedural mechanisms must coordinate participants as part of the deliberative process directed towards a new constitutional settlement or reform. It is precisely procedures for coordination of decision-making, however, which are a necessary condition for group agency. Viewed from this perspective, the constituent plural people appears to be a somewhat elusive intermediate category, because it assumes a level of procedural coordination which suggests that we have already identified a singular or unified people brought together under a shared decision-making process.

In any case, all of these considerations suggest that, if the people is to exercise constituent power, it will do so as a group agent under shared procedures for decision-making. Importantly, the people can still be viewed under the aspect of plurality (i.e., as a plurality of citizens) on this interpretation. It is possible to say that the people exercises constituent power as a unity insofar as the plurality of citizens are coordinated so that they have the shared intention of a bona fide group agent. This seems the model implicit, indeed, in Pettit’s more recent defense of the group agency of the people as sovereign.⁴⁷ While not presented as an account of the exercise of constituent power, Pettit’s defense still helps to clarify the conditions that would need to be satisfied for it to be true to say that the people is a unity which can be regarded as effective agent of constitutional change.

If the people is to be regarded as sovereign, Pettit claims, then individual members of the polity must be able to participate in its institutional procedures on relatively equal terms. Relatively equal participation, argues Pettit, would justify an identification of the polity as a group or corporate agent and its members. When the polity acts in a sovereign capacity (e.g., by amending constitutional structures and competences or by enacting laws), on this assumption, we can also say that it is the singular people that acts. Pettit attempts to elucidate this argument by analogy with the members of a club or corporation.⁴⁸ For a club to be equated with its members, the expressions “the club” and “the membership” must co-refer.⁴⁹ This rules out the situation where there are two classes of participation in the governing activities of the club so that, for example, some members play a more active role in, or some members have less than equal access to, decision-making procedures.⁵⁰ Moving from the analogy back to the unified sovereign people, we arrive at the following conditions on an identification of the polity with its members.⁵¹ First, the polity will need to have a constitutional framework with “decision-maker rules that are subject to the equally shared power of the people.”⁵² Second, the decision-maker rules constituted by the framework “must constitute a system, to which people have equal access, for disciplining what those in office can do in the name of the state.”⁵³ In short, “ordinary people” must have “power over the framework of government, and, equally, they must have power under that framework.”⁵⁴

Pettit’s argument that the people can be identified with the polity seems to rest, however, on a concept of equal power which vacillates between (i) equal opportunity to exercise a power and (ii) equal exercise of a power. In the case of the club, Pettit says, we need to show that “the members considered severally have more or less equal power—an equal say, or at least an equal chance to have a say, over how the club acts.”⁵⁵ There must be, Pettit says, “sufficient equality,” so that no individuals have “special access” and power must be distributed “more or less equally.”⁵⁶ This would allow us to say that “there is a sense in which [the members] share equally in a system for determining what the club decides and does.”⁵⁷ Yet the distinction between “equal access to power” and “equal share in power” is far from trivial, and Pettit’s argument does not clarify whether we are constrained by the weaker or stronger condition. In the case of a club, the examples are eligibility to stand for election, equal votes in those elections, equal opportunities to challenge committee decisions, the requirement for officials to “act in a way that answers equally to the interests of all,” and equal access “to calling for a review of the constitution under which elections, challenges and constraints are determined.”⁵⁸ Apart from equal voting, these examples appear to entail the weaker condition of equal access rather than equal participation. It hence appears that Pettit’s account allows us to say that the people is sovereign even if most members of the polity only vote intermittently in elections, so long as they have formal equality of access potentially to participate more actively in the determination of fundamental political decisions.⁵⁹

If one abstracts from the sociological reality of unequal access to economic, political, and legal power in constitutional liberal democracies, and restricts Pettit’s claim to formal equality of participation, one could perhaps argue that, in a suitably qualified sense, the equal power condition can be satisfied for a broad understanding of popular sovereignty parsed in terms of democratic equality. Yet the prospects are much weaker once we turn to specific procedural mechanisms for the exercise of constituent power. The problem can be seen by reconsidering constituent power’s success conditions. The concept of constituent power, by hypothesis, would describe the procedurally regulated capacity for creating or fundamentally amending a constitution. An exercise of constituent power by the people as a unified group agent would thus require all members of the polity to be actively involved in deliberation over the creation or amendment of a constitution, not only to have equal formal access to procedural mechanisms for collective deliberation. Returning to Pettit’s “club” analogy, there must be active involvement of all members in deliberative procedures for the establishment or amendment of the club’s fundamental decision-maker rules, not mere formal equality of access to offices or processes for decision-making like majority voting. Consideration of the implications of this difference speaks against the claim that all members of a polity, as a group agent, could exercise constituent power.

As suggested above, constituent power is the power to create or amend a constitution and its exercise would seem to require not merely the potential equal opportunity to participate in decision-making processes, but active involvement in processes of deliberation which play the decisive role in constitutional change. At this point it is salutary to pay heed to actual constituent processes. The most prevalent deliberative mechanisms for the exercise of constituent power, such as the constituent convention or assembly, are invariably representative in character. There is little basis for applying Pettit’s “all-in” model of popular group agency to these institutions insofar as the vast majority of citizens are excluded from their deliberations. This concern about exclusion also pertains to reform proposals which seek to increase participation in constituent processes such as the use of primary assemblies with binding mandates during initial stages of constitutional change.⁶⁰ Primary assemblies, while no doubt significantly widening the net of participation, remain representative bodies which rely on the election of eligible candidates.

It is of course more feasible for all enfranchised citizens to participate in (i) election of representatives to constituent conventions or assemblies; (ii) constitutional referendums or plebiscites; (iii) the ratification of a new constitution; and (iv) informal or formal public debate. Yet, in the case of case of (i), (ii), and (iii), it would seem misleading to say that the people exercises constituent power; here the role of citizens is either to commission representatives to act on their behalf or to endorse or reject a proposal articulated by representatives.⁶¹ In the case of (iv), it is certainly the case that public debates can influence the content of a constitution. Yet it is another proposition altogether to say that each and every citizen is engaged in public deliberation over the content of a constitution and that the unified intention formed on the basis of this deliberation played the decisive role in the ultimate determination of the content of constitutional change.⁶²

These considerations regarding active and decisive participation in the determination of constitutional content are worth examining in detail because they point to the more fundamental issue concerning constituent group agency. Even if one accepts that some citizens outside of a representative constituent convention or assembly can play an active and decisive role in the determination of constitutional content, this falls short of the claim that “the people”—even taken as shorthand for the plurality or unity of citizens—has played an active and decisive role. Most fundamentally, such cases do not entail that the totality of citizens is a group agent, because while they may be united under a shared decision-making procedure, meeting one constraint for group agency, they lack true unity of intention with respect to the determination of constitutional content.⁶³ In the election of constituent representatives, or in a referendum vote, or in public deliberation, there is no people as a unified constituent agent which exercises constitution-making power, even if there is perhaps a constitutional people that is unified under democratic procedures.

Pettit’s defense of the claim that the people can be sovereign and identified with the polity is instructive, I have suggested, because it clarifies the conditions that would need to be satisfied for there to be a singular popular constituent subject, comprised of plural citizens, which exercises constitution-making or constitution-reforming agency.⁶⁴ A close examination of these conditions reveals that, while they may make sense as a postulate of ideal theory, it is vanishingly improbable that they could be satisfied in any actual constituent process. The reason is that Pettit’s identification of the people with the polity depends on the satisfaction of constraints of equality of access or participation. While it may be possible to massage these constraints with respect to a broad idea of popular sovereignty parsed loosely in terms of democratic equality of access, this maneuver does not work with the exercise of constituent power, which would require the effective involvement of all citizens, united by a shared intention, in a decision regarding constitutional content arrived at through processes of collective deliberation. The simpler and more plausible alternative to popular constituent group agency is to acknowledge that when constituent power is exercised by representatives, it can always be attributed post facto to the people as its unified bearer.⁶⁵ The final section considers this alternative and its wider implications.

4. Implications for the theorization of constituent power

The capacity of the people to exercise constituent power does not withstand close scrutiny, yet the status of the people as the bearer of constituent power is axiomatic for contemporary constitutional theory and practice. Any impression of paradox or inconsistency in these claims dissolves, however, in light of a recognition of the conceptual independence of the status of the people as (i) the subject and (ii) the effective agent of constituent power. If constituent power is invariably exercised by representatives on behalf of “the people” (as shorthand for the collective of citizens), then this motivates significant downstream claims. As suggested in the introduction, it undermines the assertion of any natural compatibility between constituent power and radical or romantic theories of direct democracy. From an historical perspective, it is instructive that one finds incipient appeals to constituent power in the writings of sixteenth-century Huguenot theorists and defenders of parliament during the English Civil War, prior to its mature formulation in the work of the liberal-bourgeois theorist Sieyès.⁶⁶ Constituent power has consistently proven amenable to deployment in strategies of constitutional legitimation that are more liberal than radical democratic or populist in orientation. From a perspective internal to liberal constitutionalism, the arguments of the previous two sections also motivate a closer examination of the way that constitution-making representatives can or should be held accountable to normative demands derived from a theory of political representation, such as responsiveness, indicativeness, and inclusivity.⁶⁷ In this final section, I concentrate primarily on the question of representation by reference to the recent constituent processes in Iceland and Chile. My analysis suggests that democratic theorists who wish to persist with constituent power should focus on how representatives can act “on behalf” of citizens, the normative constraints on constituent representation, and the institutional and procedural mechanisms likely to satisfy those constraints.

A clear acknowledgement that constituent power is exercised by representatives and then imputed to the people as its bearer has several explanatory advantages. In the first instance, such an interpretation better reflects the sociological reality of historical constitutional settlements than accounts which make overly strong claims for the effective participatory role of the constituent people. Second, the paradoxes of constituent power are disarmed, because a constitutional people always presupposes a prior constitution-making act, and is not itself regarded as the effective agent of constitutional change. Third, this approach allows for a clearer-sighted perspective on the discrete descriptive and normative dimensions of constituent power.

The US Constitution can again be taken as a paradigmatic case to motivate the first historical claim. The delegates who met in Philadelphia in 1787 were selected by the state legislatures to amend the Articles of Confederation. When confronted with the criticism that they were acting extra-legally or unconstitutionally in proposing a new federal constitution, the delegates appealed to a conception of popular sovereignty which had already been articulated in the independence movement.⁶⁸ This appeal was certainly legitimated by the establishment of ratifying conventions to justify the status of the Constitution as a creation of a sovereign “We the People.” Yet it was not the people that exercised constituent power in this process, but delegates of the Philadelphia Convention who engaged in deliberation regarding the content of the constitutional settlement. The ratifying conventions are best understood as a process which allowed prospective citizens of the new federal union to confirm that the Philadelphia delegates had acted as their true representatives. As suggested in Section 2, the retrospective imputation of constituent power can also be articulated (either sincerely or strategically) in political vocabularies and narratives which ascribe ultimate responsibility for constitutional settlements to the collective citizenry or people.

From a juridical perspective, the exercise of constituent power by representatives can, no doubt, be imputed to the people as a personification of the polity or constitutional order.⁶⁹ The claim that constituent power is generally exercised by representatives of the people should be distinguished, however, from the (closely related) thesis that all constitutions are imposed. As Yaniv Roznai points out, “there are many ways and degrees of imposition and all constitutions can be considered as imposed to some extent.”⁷⁰ While this “continuum of imposition” is significant, however, it seems better to reserve the term “imposition” for more clear-cut cases. Obvious examples include constitutions imposed by military juntas or post-conflict occupying authorities. In the case of the Chilean Constitution of 1980, for example, constituent power remained in the hands of the Junta, guided by a Constituent Commission (under the influence of Pinochet regime ideologist Jaime Guzmán), which submitted the constitutional text approved by the Junta to a (far from open and transparent) plebiscite as a fait accompli.⁷¹ Well-known examples of (subsequently quite resilient) constitutional texts more or less imposed by constitutional committees of victorious occupying authorities on defeated populations are the still extant constitutions of (West) Germany and Japan. Revealingly, both of these constitutions include robust rhetorical assertions of popular sovereignty, despite their far-from-democratic provenance.⁷² In clear cases of imposition, constituent representatives may indeed still intend to act on behalf of the constitutional people, even if “bottom-up” popular involvement remains comparatively minimal.

These examples of imposed constitutions can assist the search for normative criteria of legitimate constituent power or constituent authority. Armed with a minimalist definition of constituent power, which does not conflate purportedly democratic or more inclusive and participatory exercises of the capacity with the capacity simpliciter, one can proceed to stipulate constraints on legitimate exercise beyond the mere successful creation of a constitution. What is needed is a principled justification for asserting that, for example, Pinochet and the military junta in Chile illegitimately exercised, or usurped, the constituent power. This approach distinguishes constituent power and constituent authority by stipulating that the latter designates an exercise of constituent power that satisfies criteria of legitimacy set out in the more normatively demanding theory.⁷³ One could then say that the military junta in Chile exercised constituent power, but not constituent authority, because its constitution-making activity did not respect the separation of powers, democratic legitimacy, the terms of its popular commission, or other normative criteria. Constituent authority, or legitimate constituent power, should be grounded, such a view suggests, in the normative requirement for those exercising constituent power to do so on behalf of all citizens.

It might be countered that an appeal to constituent representatives acting on behalf of the people is vague and open to abuse. A claim to act on behalf of the people is indeed, without further specification of criteria for effective and legitimate political representation, completely general and open to abuse. My argument is that it is precisely a specification of the content of the representative claim to act on behalf of the people which needs to be elaborated in a theory of constituent authority or legitimacy. This view is more reconcilable with leading theories of legitimate constituent power than it may first seem. Consider theories of constituent legitimacy grounded in appeals to (i) democratic inclusiveness or (ii) a direct commission or mandate from the people. An example of the former approach is found in Arato’s proposal for constituent episodes to include plural views in round-table discussions, with open communicative channels to and from civil society, as a way of mitigating dominance by establishment figures and political elites.⁷⁴ Arato points to the 1996 South African constitution-making process as an exemplar on the basis of its inclusion of diverse stakeholders in decisions over constitutional structures and competences.⁷⁵ The more radical-democratic approach, represented by Colón-Ríos, argues that an exercise of constituent power is legitimate when it occurs in line with the mandate or binding instructions of the people as a multiplicity of individuals (rather than abstract nation) and is subject to law in the sense that the mechanism of its exercise—an elected constituent convention or assembly—respects the content and limits of its mandate.⁷⁶ In both cases, the various constraints are specifications of the higher order normative requirement that constituent representatives consider a wide range of viewpoints when acting on behalf of the people. And in neither case is it necessary that the people as a unified group agent effectively exercises constituent power. A normative account of constituent authority should explain what it is for citizens to be involved in, and to influence, the constitution-making deliberative process, not require that they are the effective constituent agent. If one accepts the normative premises of modern constitutionalism, the real desideratum, in short, is that constituent representatives should act on behalf of “the people,” not elite or “vested” interests.⁷⁷

The capacity of a constituent process to satisfy the constraints of a normative theory of political representation can be measured by reference to demands such as responsiveness, indicativeness and inclusiveness. On a “classical” responsive model, representatives are regarded as “deputies” authorized to speak on behalf of the interests or views of those whom they represent.⁷⁸ One finds an appeal to “responsive” representation in theories of constituent power which advocate for the revival of “imperative mandates” or binding instructions in constituent conventions.⁷⁹ An “indicative” model of representation regards representatives as proxies who “stand for” the whole and “ought to exemplify and indicate the presumptively significant aspects of the people they represent, reproducing salient variations amongst its members.”⁸⁰ A robust version of this model is found in seventeenth-century English writers, such as Henry Parker, who argued that the English parliament “offered a recognisable image or likeness of the populace as a whole.”⁸¹ While it is rare to encounter Parker’s “symbolic” model of representation in the contemporary political theory literature, the “indicative” model remains identifiable in constituent processes which draw on random sampling and use proportional voting systems.⁸² Although elements of both responsive and indicative models reflect an aspiration to inclusiveness, recent normative theories of inclusive representation strongly emphasize the proposition that neither the people as a whole, nor social groups in a majority or minority, can claim fully to express or embody the beliefs and interests of all citizens.⁸³ As a matter of principle, inclusive representation seeks to ensure that more citizens are involved in political decision-making, such as constituent processes, than they would be prior to, or without, any representation.⁸⁴ Inclusive representation in this sense thus seeks to integrate not only perspectives of minority groups, but also viewpoint diversity across socio-economic levels. As a consequence, truly inclusive models of representation will allow for a significant degree of “partisanship” and “politicization” in constitution-making processes, so that previously or potentially excluded groups find a collective voice, whether inside the formal processes of constitutional change or in less structured political action which informs the constituent process.⁸⁵

The retrospective imputation of constituent power exercised by representatives to the people is, this brief discussion suggests, an appropriate subject for normative assessments of constitutional legitimacy: one can say schematically that an exercise of constituent power by representatives is more likely to be legitimate if it is exercised on behalf of the people in reflecting a concern for the beliefs or interests of citizens in a way that is responsive, indicative, and inclusive. A normative model of constituent representation which privileges inclusion will prioritize mechanisms and procedures which “multiply and challenge” claims by constituent representatives to represent the people, and is in this sense inherently in tension with a more unitary or populist construal of the role of the people.⁸⁶ The plausibility of this schematic normative account of constituent representation can be seen by turning to two recent examples—Iceland and Chile—of unsuccessful constitution-making processes which endeavored to implement novel forms of representation.

The unsuccessful Icelandic constitution-making experiment between 2010 and 2013 was notable for its attempt to enhance citizen participation in the constituent process through opportunities for multilevel input and utilization of technology platforms. One of the central Icelandic innovations reflecting a concern for indicativeness and inclusiveness was the “mini-public,” understood as a representative body which would allow for increased consideration of a multiplicity of viewpoints on the nation’s proposed new fundamental law. Responsive, indicative, and inclusive modes of representation can also be identified in other aspects of the constituent process, such as a national forum of 950 “randomly selected” citizens, an assembly of non-professional politicians, and use of crowdsourcing techniques.⁸⁷ In a multi-staged process, the Icelandic draft constitution was considered by the 950-member national forum, refined by a Constitutional Committee of seven experts, and only then entrusted to a Constitutional Council which was comprised of twenty-five elected citizens.⁸⁸ Following the work of the Constitutional Council, the draft constitution was presented to the Icelandic public, which was then also offered the opportunity to give feedback to the Council. After twelve iterations of the Council draft informed by this multi-staged process, the Constitution was submitted to a referendum proposal, which was endorsed by approximately two-thirds of citizens.

The use of both random selection to the National Forum and crowdsourcing techniques at the crucial stage of finalization by the Constitutional Committee did result in a relatively open constituent process in Iceland. Of course, the fact that the citizens of Iceland engaged in a relatively open constituent process in comparison to traditional models does not entail that the people—all members of the polity—were the effective constituent subject. It was the Constitutional Council, on the basis of reports from the Constitutional Committee, which played the decisive role in deliberations over the content of the constitutional text. Even if one takes into account the National Forum and crowdsourcing, only a relatively small proportion of Icelandic citizens actively participated in collective deliberations during the constituent process. Although the endeavor to implement a more open constituent process may have been better aligned with the schematic normative model of representation outlined above than a classical convention or assembly, moreover, it did not translate into the successful creation of a new constitution. On the one hand, those coordinating the constituent process were never able to overcome perceptions of elite capture.⁸⁹ On the other hand, the Icelandic process did not culminate in a new constitutional settlement in large part because of the resistance of established political elites.⁹⁰ The Icelandic constitution-making episode hence reflects some of the challenges confronting attempts to implement novel forms of representation, including potential tensions between inclusiveness and effective constitution-making, and the irreducibility of inclusiveness (at least in the more “agonistic” sense above) to multiplication of channels allowing for citizen participation.

The recent process of constitutional reform in Chile offers similar lessons. The Chilean constituent process was triggered by the government (after the initial imposition of a state of emergency and other attempts to contain popular protests) in response to the large protests which erupted in Santiago and other Chilean cities in October 2019.⁹¹ On April 26, 2020 approximately 78% of voters (51% turnout) voted “yes” in response to the proposal for the creation of a new constitution and 79% of voters chose the option for a fully elected Constitutional Convention over a Mixed Constitutional Convention (which would have been comprised of 505 popularly elected members and 50% parliamentarians).⁹² On May 16–17, 2021 an election was then held to select the 155 members of the Constitutional Convention who would be responsible for drafting the new constitution. In a national plebiscite held on September 4, 2022, 62% of the citizens of Chile rejected the proposed new constitution drafted by the Constitutional Convention. Initial commentary on the reasons for this resounding rejection of the proposed constitution, despite widespread support for constitutional reform, has pointed to the increased citizen participation in the second plebiscite relative to the first, and the fact that the content of the draft constitution was not perceived adequately to represent the views of many in the Chilean population.⁹³ What one arguably finds in the Chilean case, then, is a combination of attempts to ensure a more responsive and inclusive process (reflected in frequent appeals to the electorate to ratify proposals or vote for constituent representatives, and the diverse character of those elected to the Constitutional Convention) and contested claims regarding the inclusion of voices across the political spectrum.⁹⁴

The participatory and popular elements of the Chilean process should also not distract from the fact that it was the elected—and representative—155-member Constitutional Convention which played the key role in constituent deliberations. This would be true, indeed, even if voter turnout for the election of the representatives on the Convention had exceeded the disappointing figure of 43.3%. As might be expected with a representative body commissioned for the extraordinary power of constitutional change, disputes arose between members of the Convention regarding the precise scope of the powers that it is to exercise.⁹⁵ The principal focus of the dispute was Article 135 of the existing Constitution, which stipulates, in a subsection setting parameters to the constituent processes, that the Convention cannot exercise powers that have not been assigned to it, especially the sovereign powers which reside with the Nation. A significant minority (thirty-four) of members of the Convention nevertheless asserted that the Convention, as the bearer of the original constituent power, is not bound by the “country’s existing institutional framework.”⁹⁶ This dispute is quite revealing, because it confirms the status of the Convention both as a representative body and as the actual agent of (a perhaps limited) constituent power. The thirty-four members of the Convention made a disputable claim to be able to represent the people in the full extent of its putative sovereign power, rather than only speak on its behalf on the more narrowly constrained question of the best constitutional framework within the existing juridical structure.

All of these points motivate the conclusion that constituent representation is, like all forms of political representation, inevitably a site of political struggle. Theorists of constituent power who seek to enhance levels of citizen involvement in processes of constitution-making should, the arguments above suggest, jettison any lingering or residual romanticism regarding the direct action of the collective people, and focus attention on the development of processes which, while always politically contested, might also support more responsive, indicative, and inclusive forms of representation.⁹⁷ The unsuccessful outcomes of the constituent processes in Iceland and Chile may demonstrate the difficulties which confront such a project, but they also underline its importance.

5. Conclusion

A non-reductive interpretation of the claim that the people exercise constituent power is confronted, I have argued, with demanding requirements upon equal participation by citizens. All citizens would need to be involved in processes of deliberation which are decisive in determining the content of constitutional change—and not merely have the opportunity to be involved in such processes—for that settlement to be the effective product of the people as a unified group agent. While not impossible in principle, such participation is unrealistic and impractical, which explains the ubiquity of representative processes and mechanisms like the constituent convention or assembly. While no amount of case studies could prove as a matter of strict necessity that constituent power must be exercised by a representative body, the processes in Iceland and Chile demonstrate the difficulty of establishing the contrary. While it may be desirable to involve as many diverse citizens as possible, constituent power remains in the hands of representatives, and usually representative elites, in the most decisive junctures of the process. Yet an acceptance of the view that constituent power is exercised by representatives does not preclude a normative democratic theory of constitution-making. What should matter, for advocates of such a theory, is that representatives are constrained to exercise constituent power on behalf of citizens in a responsive, indicative, and inclusive manner, not that the people is an effective constituent agent.⁹⁸

What are the motivations for a more literal construal of the people’s capacity to exercise constituent power, given that it is historically and sociologically counter-intuitive and does not sustain close conceptual investigation? An adequate answer would require a shift from the normative register of legitimacy to the sociological register of legitimation. It of course remains highly significant for exercises of constituent power that they be recognized, or be capable of being recognized, as legitimate. What this suggests—although it is not possible to do justice to the issue here—is that a romanticism of the people’s effective constituent power can actually obscure the reality of constitutional settlements in a way which supports established power constellations.

Footnotes

*

Associate Professor, School of Humanities and Social Sciences, Deakin University, Melbourne, Victoria, Australia. Email: [email protected]. The support of Australian Research Council grant DP220100967 is also gratefully acknowledged.

1

Joel Colón-Ríos, Constituent Power and the Law 223, 259 (2020).

2

As Colón-Ríos notes, the distinction between original and derived exercises of constituent power has been employed in diverse ways by theorists of constituent power. Id. at 8–17. Yaniv Roznai distinguishes between the “primary” constituent power of “the people” and the “secondary” or delegated constituent power vested in a constitutional organ. Yaniv Roznai, Unconstitutional Constitutional Amendments: The Limits of Amendment Powers 113–23 (2017). Markus Patberg restricts secondary constituent power to circumstances where constituted powers are excluded from the determination of constitutional structures and competences. Markus Patberg, Constituent Power and The European Union 35–7 (2020). My argument is intended to encompass constituent activity ranging from the creation of a completely new constitution to secondary or derived constituent acts in Patberg’s sense. This article abstracts from complexities which arise for federal states with plural constituent subjects. SeeStephen Tierney, The Federal Contract: A Constitutional Theory of Federalism 76–81, 115–19 (2022); Nicholas Aroney, Constituent Power and Constituent States: Towards a Theory of the Amendment of Federal Constitutions, 17 Jus Politicum, 5, (2017).

3

SeeJoel Colón-Ríos, Weak Constitutionalism (2012) (offering the most articulate defense of this view). See alsoMartin Loughlin, Against Constitutionalism 77–86 (2022) (discussing the relationship between constituent power and the “normativist” model of constitutionalism).

4

SeeZoran Oklopcic, Beyond The People: Social Imaginary and Constituent Imagination (2018) (examining the “vocabularies” of popular sovereignty and constituent power). For further historical context—including the provenance of contemporary usage of the concept in the work of Emmanuel Joseph Sieyès and Carl Schmitt—see Lucia Rubinelli, Constituent Power: A History (2020); Martin Loughlin, Foundations of Public Law 221–8 (2010). For dissenting views with respect to both the explanatory usefulness and the normative desirability of constituent power, see the critiques in David Dyzenhaus, The Politics of the Question of Constituent Power, inThe Paradox of Constitutionalism: Constituent Power and Constitutional Form 129 (Martin Loughlin & Neil Walker eds., 2007); Lars Vinx, The Incoherence of Strong Popular Sovereignty, 11 Int’l J. Const. L. 101 (2013); Sergio Verdugo, Is it Time to Abandon the Theory of Constituent Power?, 21 Int’l J. Const. L. 14 (2023).Cf. Yaniv Roznai, The Boundaries of Constituent Authority, 52 Conn. L. Rev. 1381, 1406–8 (2021).

5

For the association of constituent power with political autonomy, see Patberg, supra note 2, at 5–6. On the “sleeping sovereign,” see Richard Tuck, The Sleeping Sovereign: The Invention of Modern Democracy (2015). There is an important question, addressed briefly below, about the implications of attributing constituent power to the nation rather than the people. As Colón-Ríos’s analysis demonstrates, such theories (represented notably by Sieyès) tend to restrict the role of the people to the commissioning of the extraordinary constituent representatives who determine a constitutional settlement. SeeColón-Ríos, supra note 1, at 56–76, 101–26. For the reasons discussed below, a clear-cut distinction between the attribution of constituent power to “the people” and to the nation is often difficult to maintain.

6

Joel Colón-Ríos, Eva Marlene Hausteiner, Hjalte Lokdam, Pasquale Pasquino, Lucia Rubinelli, & William Selinger, Constituent Power and Its Institutions, 20 Contemp. Pol. Theory 926, 926 (2021), https://doi.org/10.1057/s41296-021-00467-z (emphasis added).

7

A. V. Dicey, The Law of the Constitution, in 1 The Oxford Edition of Dicey 27, 27 (J. W. F. Allison ed., Oxford Univ. Press 2013). See further Martin Loughlin & Stephen Tierney, The Shibboleth of Sovereignty, 81 Mod. L. Rev. 989 (2018).

8

Martin Loughlin, Constituent Power Subverted: From English Constitutional Argument to British Constitutional Practice, inThe Paradox of Constitutionalism: Constituent Power and Constitutional Form, supra note 4, at 27, 28.

9

William Partlett, The American Tradition of Constituent Power, 15 Int’l J. Const. L. 955 (2018); David Ciepley, Is the US Government a Corporation? The Corporate Origins of Modern Constitutionalism, 111 Am. Pol. Sci. Rev. 418 (2017).

10

See, e.g., Christopher Thornhill, The Global Legal System and the Procedural Construction of Constituent Power, 5 Global Const. 405 (2016); Andrew Arato, The Adventures of The Constituent Power: Beyond Revolutions? chs. 1–2 (2017); Vinx, supra note 4.

11

Hans Lindahl, Constituent Power and Reflexive Identity: Towards an Ontology of Collective Selfhood, inThe Paradox of Constitutionalism: Constituent Power and Constitutional Form, supra note 4, at 9; Hans Lindahl, Constituent Power and the Constitution, inPhilosophical Foundations of Constitutional Law 141 (David Dyzenhaus & Malcolm Thorburn eds., 2016); Philip Pettit, On the People’s Terms: A Republican Theory and Model of Democracy (2012).

12

Joel Colón-Ríos, Constituent Power and Referendums, in Constituent Power and Its Institutions, supra note 6, at 935, 936.

13

See Niklas Luhmann, Verfassung als evolutionäre Errungenschaft, 9 Rechtshistorisches Journal 176 (1990) (discussing the distinct but related idea of an externalization of legal paradoxes of self-reference towards “the people”).

14

Colón-Ríos, supra note 1, at 262–3.

15

George Duke, Inherent Constraints on Constituent Power, 40 Oxford J. Legal Stud. 795 (2020).

16

Margaret Canovan, The People 12 (2005). See alsoAndrew Lintott, The Constitution of the Roman Republic 72 (1999); Benjamin Straumann, Crisis and Constitutionalism: Roman Political Thought from the Fall of the Republic to the Age of Revolution 119–29 (2017). Cf. Kinch Hoekstra, Athenian Democracy and Popular Tyranny, inPopular Sovereignty in Historical Perspective 15, 50–1 (Richard Bourke & Quentin Skinner eds., 2016) (discussing the demos). The continued relevance of these distinctions is seen in contemporary populist movements, which commonly rely on a pars pro toto claim about the “authentic” people in relation to the polity as a whole. See Camila Vergara, Populism as Plebeian Politics: Inequality, Domination and Popular Empowerment, 28 J. Pol. Phil. 222 (2020).

17

Alan Cromartie, Parliamentary Sovereignty, Popular Sovereignty, and Henry Parker’s Adjudicative Viewpoint, inPopular Sovereignty in Historical Perspective, supra note 16, at 142; Lorenzo Sabbadini, Popular Sovereignty and Representation in the English Civil War, inPopular Sovereignty in Historical Perspective, supra note 16, at 164.

18

Carl Schmitt, Constitutional Theory 75–88 (Jeffrey Seitzer trans., Duke Univ. Press, 2008); Carl Schmitt, The Concept of the Political (George Schwab trans., Chi. Univ. Press, 2007). See also Vinx, supra note 4 (offering an astute critical analysis of Schmitt’s “strong” idea of popular sovereignty).

19

Cf. Philip Pettit, Popular Sovereignty and Constitutional Democracy, 72 U. Toronto L.J. 251, 276–8 (2022) (developing a similar, but not identical, tripartite distinction). As elaborated below, the corporate people need not necessarily be regarded as a group agent as per Pettit’s analysis, i.e., it also admits interpretation as a juridical subject.

20

This seems, for example, to be close to Antonio Negri’s position, although the reference to citizenship in this definition might also be thought incommensurate with the materialist and resolutely anti-constitutionalist tenor of Gramsci’s account. Antonio Negri, Insurgencies: Constituent Power and the Modern State 1–34 (1999).

21

Daniel Lee, Popular Sovereignty in Early Modern Constitutional Thought 51–78 (2016).

22

Cf. Loughlin, supra note 4, at 226. On the paradox of constituent power—here construed broadly as the question of how constituent activity is possible or legitimate in the absence of prior constitutional (i.e., constituted) forms—see The Paradox of Constitutionalism: Constituent Power and Constitutional Form, supra note 4; Christopher F. Zurn, The Logic of Legitimacy: Bootstrapping Paradoxes of Constitutional Democracy, 16 Legal Theory 197 (2010).

23

Lindahl, Constituent Power and Reflexive Identity, supra note 11, at 9–11. Cf. Hans Kelsen,On the Essence and Value of Democracy, in Weimar: A Jurisprudence in Crisis 84, 89–90 (Arthur Jacobson & Bernhard Schlink eds., 2000).

24

Lindahl, Constituent Power and Reflexive Identity, supra note 11, at 9–11. The contention that Kelsen’s views on constituent power entail a collapse of politics and law could be challenged. See, e.g., Hans Kelsen, General Theory of Law and State 259 (Anders Wedberg trans., Lawbook Exchange 1945). Here Kelsen associates constituent power with institutional and procedural mechanisms, such as the constituent convention, in modern national constitutional state orders. Kelsen also insists, consistent with the arguments below, that “the original constitution of a State is the work of the founders of the State.” Id. I return briefly to Kelsen’s views on imputation in Section 4.

25

Lindahl, Constituent Power and Reflexive Identity, supra note 11, at 10.

26

Id. at 9, 18. Cf. Hans Lindahl, The Paradox of Constituent Power: The Ambiguous Self-Constitution of the European Union, 20 Ratio Juris 499 (2007). One variant of this reading is that the people should be regarded in the first instance as the “object,” not the effective subject, of constituent power. See further Eoin Daly, Alchemising Peoplehood: Rousseau’s Lawgiver as a Model of Constituent Power, 47 Hist. Eur. Ideas 1278 (2021).

27

Lindahl, Constituent Power and Reflexive Identity, supra note 11, at 19; Lindahl, Constituent Power and the Constitution, supra note 11, 151.

28

Lindahl, Constituent Power and Reflexive Identity, supra note 11, at 16.

29

Lindahl, Constituent Power and the Constitution, supra note 11, 150.

30

Id. at 141–59. As an anonymous reviewer suggests, an alternative reading of Lindahl’s position to the one I offer here is that it seeks to collapse the distinction between the bearer and the exercise of constituent power.

31

Loughlin, supra note 4, at 227 (discussing Lindahl).

32

Significantly for the argument that follows, the attribution of constituent power to the people in this sense need not signify a “mere” juridical construct; it might also, for example, enter the public sphere and indeed become part of a widely held political narrative as in the constitutional discourse of the United States. On the relationship between concepts of public law theory and public narratives, see Patberg, supra note 2, at 6–10. The juridical dimension of attributions of constituent power remains pivotal. From a historical perspective, a conception of the people as a juridical corporation played a decisive role in the theory of popular sovereignty developed in post-independence America. Yet the imputation was not only juridical, but political, because widely promulgated and legitimated by statements of political representatives, public intellectuals, journalists etc. See Ciepley, supra note 9, at 432; Christopher Thornhill, A Sociology of Constitutions: Constitutions and State Legitimacy in Historical-Sociological Perspective 158–251 (2011). I return to this point below.

33

Cf. Sandra Field, Potentia: Hobbes and Spinoza on Power and Popular Politics 97 (2020).

34

A possible objection is that a judge delivering a minority judgment might be said to exercise judicial power, even if they do not play the truly decisive role in bringing about the outcome. This case is, however, disanalogous insofar as an effective exercise of judicial power would not involve the collective intentionality of a large group of citizens. On the success conditions for the exercise of constituent power above, for the people to exercise constituent power requires that they jointly and severally play a decisive role in the effective determination of the content of a constitution through deliberation, which presupposes the shared intention of a large collective agent. The capacity for citizens to exercise constituent power through voting is considered below. I am grateful to an anonymous reviewer for pressing these points.

35

Pettit, supra note 11, at 286.

36

Id.

37

Id. Pettit’s strategy has acknowledged affinities with the arguments of the Italian post-glossators, particularly Baldus’s theory of the hominum collectio in unum corpus mysticum. Significantly, for the argument which follows, Baldus considers the populus or the people-as-collective-unity under the aspects of both a “plurality of humans” and as a persona ficta or legal fiction. Cf. Lee, supra note 21, at 74–5; Brian Tierney, Religion, Law and the Growth of Early Modern Constitutional Thought 1150–1650, at 26 (1982).

38

Pettit, supra note 11, at 286.

39

Id. at 292.

40

Id.

41

See K. M. Baker, Inventing the French Revolution 224–51 (1990); William Doyle, The Oxford History of the French Revolution 86–111 (2018); Gordon S. Wood, The Creation of the American Republic 1776–1787 (1998); Bernard Bailyn, The Ideological Origins of the American Revolution 160–229 (2017); Gordon S. Wood, Power and Liberty: Constitutionalism in the American Revolution 32–99 (2021).

42

Christian List & Philip Pettit, Group Agency: The Possibility, Design and Status of Corporate Agents 32–40 (2011).

43

Id.

44

Id. at 194.

45

On the procedural pre-conditions for the exercise of constituent power, see Colón-Ríos, supra note 1, at 8–13 and passim. On “roundtable” constituent processes, see the illuminating discussion in Arato, supra note 10, at 365–418. On constituent power in relation to referendum processes, see Colón-Ríos, supra note 1, at 274–81; Richard Albert, Constitutional Amendments: Making, Changing, Breaking and Changing Constitutions 72–3 (2019). I return to the question of the people’s role in constituent referendums and plebiscites at the end of the current section.

46

Hélène Landemore, Open Democracy: Reinventing Popular Rule for the Twenty-First Century 152–78 (2020) (offering a useful overview of the use of crowd-sourcing and other innovative deliberative and consultative techniques in the attempted constituent process in Iceland—discussed in detail below in Section 4).

47

Pettit, supra note 19.

48

One may certainly raise doubts about the potentially misleading character of analogies between modern polities and clubs or corporations, but I leave this mostly to one side here. An obvious concern is that whereas membership of clubs tends to be voluntary, one is usually “born into” membership of a polity. A further concern is that clubs tend to have more limited and easily identifiable purposes. The difference of scale is also significant, as I briefly discuss below.

49

Pettit, supra note 19, at 281.

50

Id.

51

Id.

52

Id. at 283.

53

Id.

54

Id.

55

Id.

56

Id. (emphasis added).

57

Id.

58

Id. at 281.

59

One wonders, for example, what Pettit’s account would make of the not unrealistic scenario of a constitution which contains a provision allowing for a citizen-led legislative change; but this is rarely, or never, enlivened due to the high threshold requirements for participation and relatively low citizen engagement in the relevant political process.

60

Colón-Ríos, supra note 1, at 300–2.

61

As Colón-Ríos notes, this view about the effective exercise of constituent power is implied in Sieyès, for whom the citizenry (in contrast to the Nation) exercises only a commissioning power or pouvoir commettant. Id. at 104, 275. The now-“dominant” view of constitutional amendment, which derives largely from Schmitt, suggests that it is only the people as constituent power that is able to make fundamental amendments to a constitution. Albert, supra note 45, at 72. See also Pasquino, supra note 12 (on the role of the citizens in the authorization of modifications to fundamental laws in referendums). The prevalence of pluralism and disagreement, the limits of deliberative rationality, and the unlikelihood of unanimity nonetheless support the view that it is implausible to regard participants in a referendum as a group agent.

62

It could be argued that a combination of (i), (ii), (iii), and (iv) might satisfy the active participation constraint, even if it is implausible that the constraint could be satisfied by one of the avenues independently. The idea that all citizens would participate (or even wish to participate) in public deliberation about the detailed content of a constitution is of course hard to warrant but could, as noted above, nevertheless be entertained as a postulate of ideal theory. In any case, in order for this deliberation, in combination with voting, to be considered as an exercise of constituent power more than universal participation would be required; it would also require a group intention (for the reasons explained below).

63

One might perhaps argue that there is a shared intention to participate in the election or referendum process, but this surely falls well short of the requirements for the effective exercise of constituent power. For different but related concerns about constituent power and constitutional referendums, see Jeffrey A. Lenowitz, The People Cannot Choose a Constitution: Constituent Power’s Inability to Justify Ratification Referendums, 83 J. Pol. 617 (2021).

64

Pettit, supra note 19.

65

It could, of course, be countered that the statement that constituent power is always exercised by representatives is a contingent rather than a necessary truth. It is true that it is not inconceivable that all citizens of a polity could deliberate about constitutional change; but this is to deal in thought experiments which do not reflect the reality of any actual historical constituent process or any currently imaginable. A Rousseuavian model of constitution-making could only operate in a modern state through significant restrictions on the franchise. It is surely significant, moreover, that attempts to “implement” the idea of constituent power have historically, since its early modern antecedents in the resistance theory of the Monarchomachs and explicit formulation in Sieyès, invariably appealed to institutional procedures.

66

On incipient theories of constituent power in Monarchomach theorists and seventeenth-century England, see Lee, supra note 21, at 121–58, 273–315. On Sieyès, see Pasquale Pasquino, Sieyès et l’invention de la constitution en France (1998); William H. Sewell, A Rhetoric of Bourgeois Revolution: The Abbe Sieyes and What Is the Third Estate (1994).

67

On representative claims about who “we” are and the identification of the collective will, see Michael Saward, The Representative Claim (2010). There are, of course, a variety of interpretations of political representation, only some of which are considered here. See generallyPolitical Representation (Ian Shapiro, Susan C. Stokes, Elisabeth Jean Wood, & Alexander S. Kirshner eds., 2009).

68

Wood, The Creation of the American Republic 1776–1787, supra note 41, at 533–5.

69

Kelsen, supra note 24, at 99.

70

Yaniv Roznai, Internally Imposed Constitutions, inThe Law and Legitimacy of Imposed Constitutions 58, 60 (Richard Albert, Xenophon Contiades, & Alkmene Fotiadou eds., 2019). See alsoDieter Grimm, Constitutionalism: Past, Present and Future 114 (2016).

71

Octavio Ansaldi & Maria Pardo-Vegara, What Constitution? On Chile’s Constitutional Awakening, 31 Law & Critique 7, 17–18 (2020).

72

SeeArato, supra note 10, at 22. Commonwealth constitutions are arguably a sui generis category, because the formal constituent legislative activity of the British Imperial Parliament was often exercised in conjunction with colonial constitutional conventions and ratificatory referendums which included relatively high levels of popular involvement for the time.

73

Andrew Arato, Post-Sovereign Constitution-Making: Learning and Legitimacy 3–30 (2016).

74

Id. at 9.

75

Arato also points to constituent process in Nicaragua (1986), Brazil (1988), Uganda (1995), and Kenya (2001). Id.

76

Colón-Ríos, supra note 1, at 289. Importantly, respect for the limits of the people’s mandate entails “double differentiation,” or “a negation of the power of the ordinary legislature’s power of constitutional change and a negation of the power of constituent assemblies to produce ordinary governmental acts.” Id. at 161.

77

Arato’s theory quite self-consciously disposes of the idea of a “unitary popular sovereign” in favor of a Arendtian–Habermasian model in which the measure of the democratic legitimacy of a constituent process is its openness to flows of communication from civil society and the channeling of these flows up to constituent bodies through the deliberative involvement of multiple viewpoints. While Colón-Ríos’s account seems more populist in its implications, on closer examination it also requires representation at the crucial stage of constitutional deliberation. The people should be able to initiate a constituent process (through a citizen-led initiative or popular protest for example), feed into that process through primary assemblies, and have the final say in ratifying a constitutional text. Notwithstanding all these points, it is ultimately still a constitutional convention or assembly which exercises constituent power on behalf of the people.

78

Philip Pettit, Varieties of Public Representation, inPolitical Representation, supra note 67, at 61, 74.

79

See, e.g., Colón-Ríos, supra note 1, at 289–92.

80

Pettit, supra note 78, at 67.

81

See Quentin Skinner, Hobbes on Representation, 13 Eur. J. Phil. 163 (2005).

82

Pettit, supra note 78, at 66.

83

Arato, supra note 73, at 109, 118. The concern for an “inclusive” model of representation is in part a response to the fact that representation must always be “constructive” or “interpretative.” An emphasis upon inclusive representation, that is to say, counterbalances or mitigates the tendency for representation to motivate the postulation of a “coherent mind” or a substantive unity of intention. This dimension of inclusive models corresponds with the arguments of Section 3. For discussion, see Pettit, supra note 78, at 76.

84

Samuel Hayat, Varieties of Inclusive Representation, inCreating Political Presence: The New Politics of Democratic Representation 141, 142 (Dario Castiglione & Johannes Pollak eds., 2018).

85

Id. at 143–50.

86

See Bryan Garsten, Representative Government and Popular Sovereignty, inPolitical Representation, supra note 67, at 91; Nadia Urbinati, Representative Government: Principles and Genealogy (2006).

87

SeeLandemore, supra note 46, at 152–78 (offering a detailed treatment of the Icelandic experiment). See also Bjarki Valtysson, Democracy in Disguise: The Use of Social Media in Reviewing the Icelandic Constitution, 36 Media, Culture & Soc’y 52 (2014); Thorvaldur Gylfason, The Anatomy of Constitution-Making: From Denmark in 1849 to Iceland in 2017, inRedrafting Constitutions in Democratic Regimes: Theoretical and Comparative Perspectives 217 (Gabriel L. Negretto ed., 2020); Thorvaldur Gylfason, Iceland’s Citizen Constitution: The Window Remains Wide Open, VerfassungsBlog (Apr. 4, 2016), https://verfassungsblog.de/icelands-citizen-constitution-the-window-remains-wide-open/.

88

Landemore, supra note 46, at 160 (observing that only two members of the Council of twenty-five were former politicians).

89

Critics of the process noted the prevalence of “highly educated, urban, wealthy individuals on the Council.” Landemore, supra note 46, at 161.

90

The Icelandic Parliament, when asked to pronounce on the validity of the draft constitutional text, did not actually vote on the constitutional proposal. This followed a referendum in which two-thirds of Icelandic citizens endorsed the draft constitution presented by the Constitutional Council. SeeLandemore, supra note 46, at 156. One also needs to take into account the relatively small (329,000 citizens) and homogenous character of the Icelandic population.

91

See Ansaldi & Pardo-Vegara, supra note 71, at 11 and passim (offering a detailed account of this process).

92

Id. at 29.

93

See, e.g., Maria Carrasco, Why We Failed to Approve the New Chilean Constitution: The Need for a Cultural Transformation, London School Econ. Latin Am. & Caribbean Ctr. Analysis Blog (Sept. 15, 2022), https://blogs.lse.ac.uk/latamcaribbean/2022/09/15/why-we-failed-to-approve-the-new-chilean-constitution/.

94

For a persuasive argument that the Chilean constituent process did in fact exemplify central features of the “post-sovereign” paradigm of constitution-making, see Sergio Verdugo & Marcela Prieto, The Dual Aversion of Chile’s Constitution-Making Process, 19 Int’l J. Const. L. 149 (2021). See also Jorge Contesse, Chile’s Kaleidoscopic Constituent Assembly, Verfassungsblog (June 22 2021), https://verfassungsblog.de/chiles-const/.

95

See William Partlett, The Dangers of Constitution-Making, 38 Brook. J. Int’l L. 193 (2012).

96

See Rodrigo Kaufmann, Constituent Process and Constituent Power, Verfassungsblog (June 25, 2021), https://verfassungsblog.de/constituent-process-and-constituent-power.

97

As an anonymous reviewer noted, the relation between constituent power and the representative role of political parties deserves closer study. It is certainly arguable that political parties—and the crisis of the party system more generally—played an important role in the failure of the projects for constitutional change in Iceland and Chile.

98

See Ansaldi & Pardo-Vegara, supra note 71, at 36 (in relation to Chile).