The idea of democratic legitimacy

6    The Idea of Democratic Legitimacy

Constituent power, it has been suggested in previous chapters, has an important connection with the democratic ideal. In its modern and contemporary formulations, constituent power is attributed to the nation, the people or the community; in short, to all those who will become subject to the future constitutional regime. Constituent power points toward a democratic constitution-making power, and its popular and collective character makes it incompatible with given or imposed constitutions. This is why, paraphrasing Antonio Negri, to talk about democracy at the level of the fundamental laws – about popular sovereignty in the context of constitutional change – is to talk about constituent power.¹

Not surprisingly, constituent power has traditionally been associated by constitutionalists with instability and the risk of political revolution. A multitude always getting what it wants, continually making and unmaking laws, represents the antithesis of good government; the rule of people’s ever-changing wishes against the empire of law and reason. As we saw in the previous chapter, since constituent power is about creating new constitutions without being subject to any form of positive law, constitutionalists’ fears are exponentially heightened. Nevertheless, as a result of its connections to the basic principles of democracy, constituent power can provide us with a way of assessing the democratic legitimacy of a constitutional regime.

The chapter is organised in the following way. It begins by exploring the idea of legitimacy and distinguishing it from the related concepts of authority and justification. The conception of legitimacy that will emerge from that analysis will be highly procedural, but will not be able to assist us in identifying the specific procedures that must be present for the relevant institution (in our case a constitutional regime) to be considered legitimate. The chapter then proposes to fill that normative vacuum with the theory of constituent power. A conception of legitimacy centred on the theory of constituent power, however, might be seen as inevitably accompanied by an uncontrollable constitution-maker that may be tempted to abolish democracy or to perpetuate itself. In response to these fears, this chapter will argue that in virtue of its connections to the basic principles of democracy and its finality of establishing a new constitution, the exercise of constituent power should not be seen as a disaster waiting to happen.

It is rarely clear what is meant when someone says that the exercise of political power by some human beings over others is legitimate. In fact, as David Beetham has argued, the meaning of the term legitimacy usually depends on the training of the academic who is speaking or writing: whether she is a lawyer, a social scientist or a philosopher will have an important impact on her approach to the question of what constitutes legitimate political power.2 For instance, lawyers are traditionally concerned with how legal rules are posited, revised and enforced. For them, a legitimate political power is one acquired and exercised according to established law; in their view, says Beetham, legitimacy is equivalent to legal validity.³ This kind of approach is reminiscent of (though not identical to) Hans Kelsen’s pure theory of law. According to the pure theory, the question of legitimacy is only relevant if understood in terms of legal validity.⁴ A legitimate constitution is thus one that has been adopted in the manner prescribed by a superior norm (e.g., the amendment rule of the previously valid constitution). Questions such as: ‘What is the content of the laws?’ ‘Who posits them and how?’, which are fundamental to other conceptions of legitimacy, simply do not figure under the pure theory.⁵

Social scientists have a very different approach to legitimacy. The social scientist is not normally interested in questions of legal validity, but in looking at the extent to which those who exercise political power can count on the obedience of those subordinate to them.⁶ Social scientists are not concerned with the idea of legitimacy in universal or normative terms. Their objective is to show how legitimacy affects power relations in particular societies. By trying to stand back from their own values and beliefs (e.g., from their idea of what would amount to a ‘truly’ legitimate political power according to some set of normative criteria), they aim to discover what is actually believed in the society they are studying. Beetham sees this approach exemplified in Max Weber’s conception of legitimacy as the belief in legitimacy of the relevant social agents.7 In this vision, power relations are legitimate when “those involved in them, subordinate as well as dominant, believe them to be so”.⁸ Such an approach would consider a constitution legitimate if both officials and citizens believe it to be legitimate (regardless of the causes of that belief).⁹ For the social scientist, to say that a determinate political power is legitimate is to make a report, to describe the beliefs of a particular group of human beings. The question of legitimacy does not involve a juridical assessment of how a regime came into being, but an empirical judgment.¹⁰

Beetham distinguishes the approaches of the lawyer and of the social scientist from that of the philosopher.¹¹ Moral and political philosophers are usually not interested in legal validity or in the actual beliefs of those involved in a relation of power. They are concerned with the question of how political power ought to be arranged.¹² For them, political power is legitimate when its rules can be justified according to normative principles with which any rational and unbiased person would agree to. What is legitimate to the philosopher “is what is morally justifiable or rightful; legitimacy entails the moral justifiability of power relations”.¹³ This approach is exemplified in the political philosophy of John Rawls, who attempts to show how the idea of justice can guide the establishment of a just constitution.¹⁴ Consistent with Beetham’s depiction of the philosophical approach, Rawls expresses his conception of legitimacy in terms of what is justifiable to citizens: “As we have said, on matters of constitutional essentials and basic justice, the basic structure and its public policies are to be justifiable to all citizens, as the principle of political legitimacy requires.”¹⁵ In this sense, the project of those engaged in the philosophical approach is that of elucidating the general principles according to which political power may be justified.¹⁶

These three approaches to the question of legitimacy are different from the one that I will defend in this chapter.¹⁷ This chapter is about democratic legitimacy, not about what we may call ‘legitimacy as such’. These accounts are useful, however, because they throw light on the ways in which the ‘legitimacy’ aspect of ‘democratic legitimacy’ has been treated from different academic perspectives. But in order to fully explain the way in which the concept of legitimacy is used in this book, some further clarifications are needed. Accordingly, in what follows I distinguish the idea of legitimacy from two other concepts that, although related to legitimacy in important ways, are sometimes used interchangeably. The first of these concepts is that of justification. There is, of course, no single way of distinguishing between justification and legitimacy, and it is not necessary to undertake a comparative review of different understandings of these concepts.¹⁸ Instead, I propose to examine one influential account of this distinction, and build from there. The view that I have in mind is that of A.J. Simmons, and I will briefly outline its main points as a way of introducing the discussion.19

Simmons is particularly interested in the legitimacy and justification of the state. Under his view, legitimacy and justification provide different dimensions of institutional evaluation and involve different kinds of arguments: to legitimise a state, one must show that it has a special relationship with its citizens that gives it a right to rule over them; to justify the state, one must show that some realisable type of state is preferable to any (feasible) non-state alternative.²⁰ If the distinction between legitimacy and justification collapses, Simmons argues, political philosophers would be robbed of one important mode of institutional evaluation. Simmons finds the model for the kind of distinction that he is trying to advance in the work of John Locke.²¹

In Locke’s political philosophy, political power is legitimate only if subjects have freely consented to it and if it is continuously exercised within the terms of those subjects’ consent. The legitimacy of particular states lies in the actual history of that state’s relationship with its subjects: if the subjects freely consented to the exercise of state power, their state is a legitimate one; if they did not, then they live in an illegitimate state.²² Nevertheless, when Locke advances his argument in favour of the preferability of the limited state (the state ruled by limited government) over life in the state on nature, he does not rely on consent theory (but on the idea that leaving the state of nature and creating a state would allow for a superior protection of individuals’ lives, liberties and estates).²³ In other words, the fact that it is good to live in a state might be enough reason to support it, but does not create a duty to obey its dicta.

What Simmons wants to take from Locke is that the considerations that serve to justify the state (which make it preferable to life in the state of nature) cannot by themselves legitimise it.²⁴ In this way, Locke captures the distinction between legitimation and justification: “[T]he Lockean, I take it, wants to say the following: the general quality or virtues of a state (i.e., those features of it appealed to in its justification) are one thing; the nature of its rights over any particular subject (i.e., that in which its legitimacy with respect to that subject consists) are quite another thing.”²⁵ That the limited state is justified does not say anything about its legitimacy. Justification is about demonstrating that the state is, on balance, a good thing; legitimation about showing that the state has the kind of relationship with its citizens that gives it the right to require their obedience. Simmons contrasts this approach with one that finds in the justification of the state the very conditions of its legitimacy. He identifies this view in Immanuel Kant and his followers. Kantians use the same kinds of arguments in attempting to justify the state as in demonstrating its legitimacy. Accordingly, they think of institutional evaluation in terms of what ought to be chosen by people and not in terms of their actual choices.²⁶ Under their approach, the actual historical relation between the state and its citizens is irrelevant, and a state that can be rationally justified enjoys political legitimacy. ²⁷

In distinguishing between legitimacy and justification, Simmons associates the idea of legitimacy with the ways in which the state arises. In his view, citizens have to do something before the juridical apparatus that governs them can be legitimated. Because he is committed to political voluntarism, for Simmons this ‘something’ must refer to the actual consent of individuals (a condition that Simmons, the philosophical anarchist, maintains is not met by any modern state).28 Although for reasons that will soon become clear, I do not entirely subscribe to Simmons’ conception of legitimacy, I believe that it has certain advantages over the Kantian approach. That approach (as the one identified earlier as the ‘philosophical approach’) treats legitimacy and justification as being the same things. Nevertheless, while Simmons’ conception gives us valuable insights into the distinction between justification and legitimacy, it does so at the price of identifying legitimacy with the idea of authority.²⁹ For Simmons, a state’s legitimacy gives it the exclusive right to impose duties on subjects through legally binding directives and to coerce those who refuse to comply with them.³⁰ This right to rule and the correlative duty to obey is what political philosophers have traditionally referred to by the term ‘authority’.³¹ Authority and legitimacy, in my view, should be understood as different (although closely related) concepts.³²

That is to say, from the idea that a juridical order is legitimate, it does not necessarily follow that citizens have an obligation to obey the law. It is true that there might be conceptions of authority that say: every legitimate state has a right to rule and to be obeyed.³³ And it is also true that Simmons’ looks like one of those conceptions. But those are conceptions of authority that make a state’s authority dependent on its legitimacy, not conceptions of legitimacy. It is not surprising that authority is usually conceived of as involving legitimacy: the argument that an illegitimate power could have the right to be obeyed is unpopular for good reasons. Nevertheless, from this does not follow that those subject to a legitimate power have a moral duty to obey it; that is a separate idea.³⁴ In that respect, it may be said that authority entails legitimacy, but not the other way around.³⁵ Legitimacy, as Simmons maintains, is about whether the way in which the state arose is considered the correct one according to some external criteria. However, whether or not those subjects are obliged to obey the laws that emanate from that political power is a different matter; it is a matter of the lawgiver’s authority (which might in turn depend on other considerations).

In this section I have distinguished between the concepts of justification, authority, and legitimacy, and maintained that these three concepts, when applied to the exercise of political power, should be conceived in the following way: (a) Justification: to say that a state is justified is to say something positive about it, to suggest that it is better to have that type of state than to live in a non-state situation; (b) Authority: to say that a state enjoys authority is to say that it possesses the right to rule and to be obeyed by those within the scope of its power; (c) Legitimacy: to say that a state is legitimate is to say something about the ways it arose, about its pedigree. My principal objective has been to make clear that when I talk about legitimacy, I am neither making any claims about the advantages of having a state nor attempting to demonstrate that people have a moral obligation to obey the state’s directives. Legitimacy does not point toward the moral qualities of the institutions that are being assessed (in my analysis, constitutional regimes) or to their right to rule: it points to the way in which those institutions were created, and (as I will argue shortly), when combined with democracy, to the ways in which they can be altered.

In the context of democratic legitimacy, which is the modality of legitimacy that interests me here, the distinction between legitimacy and authority is even more important.36 Notice that in Simmons’ approach, the link between the concepts of legitimacy and authority is a result of his strategy of connecting legitimacy with the idea of consent. Consent theory is about trying to explain the authority of the state and its basic idea is that those who wield political power over other human beings have the right to be obeyed only if the latter have freely consented to their authority.³⁷ Democratic theory, in contrast, does not provide (and does not attempt to provide) an explanation of the state’s authority. In fact, democracy does not even involve the idea that citizens have a moral obligation to obey the law. What democracy requires is the equal participation of citizens in the positing of the laws and institutions that govern them (not merely their consent to the establishment of a political community).³⁸ By virtue of that participation, citizens would have good reasons to take compliance with laws seriously as they resulted from procedures that expressed a commitment to their political equality.³⁹ But if these laws (even if they were the result of a democratic process) involve serious violations of democratic principles (e.g., disenfranchising a segment of the population or establishing a dictatorship through popular vote), those that disagree with them would have a good reason to consider disobedience.⁴⁰

But the idea of legitimacy that I subscribe to is similar to that advanced by Simmons in the sense that it partly looks for the legitimacy of a constitutional regime in the way that it arose, or, as he puts it, in the kind of historical relationship that it has with its citizens. It is also ‘procedural’ in a similar way to Jeremy Waldron’s approach to the legitimacy of political decisions: to ask whether a decision is legitimate is to ask whether it was taken according to the right procedures.⁴¹ As I will argue later, this does not mean that substance is irrelevant for the question of legitimacy: if the conditions that allow for the continuing legitimacy of a constitutional regime are abolished, its claims to legitimacy are immediately put into question, regardless of the ways in which such abolition took place. Nevertheless, my conception of legitimacy differs from each of the three approaches considered in the previous section: it does not provide enough tools to adequately differentiate the legitimate from the illegitimate.

Each of the approaches considered in the first section provides (at least in theory) a straightforward test for assessing the legitimacy of political power. According to the legal approach, political power is legitimate if it is acquired and exercised according to established law (no matter the content of that law). Here, the test of legitimacy requires an exercise in juristic interpretation: if power was acquired and exercised legally, it is legitimate (valid). The social science approach demands that those involved in power relations believe them to be legitimate. Here, the test of legitimacy requires us to engage in an exercise in sociology: if the relevant group of human beings believes in the legitimacy of their political system, then the system is legitimate (believed to be legitimate). Lastly, the philosophical approach maintains that political power is legitimate where its basic structure would be favoured by every rational and unbiased person. Here, the test of legitimacy requires an exercise in reason: if a political structure can be rationally justified, then it is legitimate (justifiable). As noted above, the conception of legitimacy presented here associates the legitimacy of laws and institutions with the ways through which they come into existence (and, as we will see later, with the ways they can be changed) but, unlike the previous approaches it does not tell us which ‘ways’ are the right ones.

That is to say, it does not tell us what processes, what type of ‘historical relations’, are the preferred ones. It provides us neither with a ‘test’ of legitimacy nor with a conception of legitimacy that, like Simmons’, comes accompanied by a theory about what must happen for the relevant institution to be legitimate (in Simmons’ approach that role is played by consent theory). In that respect, it is a conception of legitimacy that needs to be supplemented with some external criteria: it has a void that must be filled with a theory that can tell us how to differentiate between the right and the wrong ways of constitution-making. That void can initially be filled with democracy; that is to say, to talk about ‘democratic legitimacy’ rather than about ‘legitimacy’ as such.42 The idea of democratic legitimacy can take us a long way to determining what kind of procedures are the right ones. At the very least, it suggests that for a constitutional regime to be considered legitimate from a democratic perspective, its constitution must be created through democratic procedures: procedures that are consistent with the principles of popular participation and democratic openness (discussed in Chapter 4).

Under this view, a constitutional regime which has been created through a process in which ordinary citizens are free to propose, deliberate and decide about the content of the constitution is certainly more likely to be considered democratically legitimate than a constitutional regime that has been implemented by a foreign power or a military elite. However, the manner in which a constitutional regime arises – its democratic pedigree – cannot be enough to satisfy the demands imposed by the democratic ideal. Democratic legitimacy, unlike a conception of legitimacy based on consent theory, cannot merely look to the past of the constitutional regime (i.e., to the historical relation it has with the citizens who live under it), but it must also look toward its potential future. A participatory and open constitution-making episode is, in that sense, only part of a democratically legitimate constitutional regime: the constitution must also remain permanently open to fundamental constitutional change; that is, to the future exercise of constituent power.

In fact, I believe that in virtue of its connections to the democratic ideal (and, as a result, to the principles of democratic openness and popular participation), constituent power can help us judge the democratic legitimacy of a constitutional regime. First, as argued in Chapter 5, constituent power is not about a one-time constitution-making event: the people, as constituent subject, may engage in important constitutional transformations whenever they consider it necessary. Second, constituent power requires that those transformations can be understood as having been made by the people, and that means that they must take place through highly participatory procedures. When assessed from the perspective of constituent power, the democratic legitimacy of a constitutional regime would thus depend on: (a) whether the constitution has a democratic pedigree (i.e., whether it was created through an open and participatory process); and (b) whether it is susceptible to re-constitution (or, what is the same thing, to future exercises of constituent power). Only a constitutional regime that does not see the people’s constituent power as a threat can meet these requirements.

Granted, a conception of democratic legitimacy that rests on the theory of constituent power, that seeks to leave the door open for constituent power to manifest from time to time, might be seen to be accompanied by necessary and serious risks. As noted earlier, the exercise of constituent power has been traditionally seen as a threat to both democracy and constitutions. There is something to the fears associated with constituent power, for constituent power (as well as democracy) has frequently been invoked by dictators and despots. The most famous example are probably the words attributed to Napoleon Bonaparte, “je suis le pouvoir constituent”,⁴³ but there are also more recent situations in which a dictatorship has declared itself the bearer of the constituent power (as in Chile under Pinochet, and Spain under Franco).⁴⁴ Moreover, France, where the theory of constituent power was first developed, experienced a wave of constitution-making during the eighteenth century that culminated in dictatorship, and in Latin America, where constituent power has long been part of the constitutional tradition, one can identify numerous constitutional breaks followed by constitution-making episodes (and, literally, hundreds of constitutions). Before considering in more detail the two criteria mentioned before, it is thus necessary to defend the theory of constituent power from these possible critiques.

Democratic Legitimacy and The Risks of Constituent Power

It should come as no surprise that some authors have depicted the exercise of constituent power as lawless and arbitrary, as incapable of creating a constitution or unable to resist the temptation of perpetuating itself.⁴⁵ For instance, in the sequel to We the People, Bruce Ackerman identifies constituent power as an arbitrary will that manifests itself in acts of upheaval in which “law ends, and pure politics (or war) begins”.46 A group of human beings engaged in the exercise of constituent power shows no respect for the established constitutional forms; such a group is simply putting into practice the (frequently violent) ‘right of revolution’, to use Carl Friedrich’s formulation.⁴⁷

Interestingly, Ackerman’s recommended constitutional politics do not involve the “sheer acts of will”⁴⁸ that allegedly characterise constituent power: even though Ackerman’s revolutionaries (the Founding Federalists, the Reconstruction Republicans and the New Deal Democrats) failed to follow the established rules for constitutional change and in that sense engaged in constituent activity, they “experienced powerful institutional constraints on their revisionary authority”, as they created new, higher laws without entirely repudiating the previous constitutional order and tradition.⁴⁹