Notes on the Contributors
Cormac Mac Amhlaigh, University of Edinburgh
Andreas Follesdal, University of Oslo
Martin Krygier, University of New South Wales
Fabienne Peter, University of Warwick
Philip Pettit, Princeton University/Australian National University
Jiří Přibáň, Cardiff University
Joseph Raz, Columbia University/King’s College London
Nicole Roughan, University of Auckland
Wojciech Sadurski, University of Sydney/University of Warsaw
Michael Sevel, University of Sydney
Kevin Walton, University of Sydney
Paul Weithman, University of Notre Dame
Introduction
Wojciech Sadurski, Michael Sevel, and Kevin Walton
Traditionally, discussions of legitimacy have been about states. The present volume engages with this tradition, but goes beyond it to examine, in addition, the legitimacy of suprastate institutions and norms. Yet consideration of legitimacy in these two contexts obviously raises another issue: what is legitimacy? Part One focuses on this general inquiry. But every contributor is concerned, even if only implicitly, with it.
Here is one aspect of this general inquiry: what sort of problem is legitimacy? Some, most notably Max Weber,1 consider it empirical, some think of it as moral, and some believe it to be both. Philip Pettit, the author of the first chapter, is among those for whom it is a moral problem. According to Pettit, a state is legitimate if its exercise of power over its citizens is morally justified. One contributor who takes legitimacy to be a matter of social fact is Jiří Přibáň, whose chapter is a sociological study of the ambiguous role of values in legitimation of the state. Martin Krygier thinks that both moral and empirical tests must be met for an institution to be legitimate. For Krygier, the legitimacy of the state depends on the perceptions of its citizens plus its moral character.
Pettit not only adopts a moral conception of legitimacy; he also separates the issue of legitimacy from those of justice and authority. The distinction between legitimacy and justice is familiar. Wojciech Sadurski also makes it in his chapter. On this common view, a state can be legitimate without being entirely just and can be illegitimate while being broadly just. The question of legitimacy, Pettit explains, is about the vertical relation between the state and its citizens, whereas the question of justice is about the horizontal relations established by the state between its citizens. In making the further distinction between legitimacy and authority, Pettit gives the former priority over the latter. He takes the question of legitimacy to be more fundamental than the matter of authority and, therefore, crucial for any moral obligation owed to the state by citizens.
What is this moral obligation? It is, Pettit supposes, an obligation on citizens who contest the law to do so in a way that does not involve changing the regime. But most
1 See Max Weber, Economy and Society: An Outline of Interpretive Sociology (Guenther Roth and Claus Wittich eds, Ephraim Fischoff and others trs, University of California Press 1968) 212–301.
Legitimacy: The State and Beyond. First Edition. Edited by Wojciech Sadurski, Michael Sevel, and Kevin Walton. © Wojciech Sadurski, Michael Sevel, and Kevin Walton 2019. Published 2019 by Oxford University Press.
theorists take a different view: they assume that the obligation is to obey the law. For Sadurski, the majority view is ill-suited to the supranational realm, where it should be replaced, in his opinion, by a weaker alternative, according to which those subject to legitimate norms have a moral obligation merely to respect them.
Among the many theorists for whom the legitimacy of an authority entails a moral obligation to obey its directives is Joseph Raz, the author of the opening chapter of Part Two. His earlier work features in the contributions of several others, most obviously in Fabienne Peter’s response to the objection that Raz’s view of legitimate political authority fails to distinguish authority adequately from expertise, in Michael Sevel’s extension of Raz’s perfectionist-liberal theory of legitimacy to international law, and in Andreas Follesdal’s reliance on Raz’s conception of authority in his chapter on the legitimacy of international courts.
Another theorist who is mentioned in more than one chapter is John Rawls. Indeed, his treatment of legitimacy in his later work is the exclusive concern of Paul Weithman’s contribution, which offers an interpretation of his view. More critical is Sadurski, who dismisses Rawls’s account of public reason as a conception of legitimacy in the supranational sphere. Sevel considers his work, too, albeit less directly, when defending Raz’s perfectionist brand of liberalism against Martha Nussbaum’s Rawlsian critique.
Besides these general issues and the writings of Raz and Rawls, what else do the contributors discuss? Short summaries of their chapters follow.
From the perspective of republicanism, which he defends elsewhere, Pettit looks at theories of state legitimacy of three types: benefit-, merit-, and will-based. Since benefits can be provided by dictators and the merits of a leader cannot be guaranteed, the third, according to which the will of the state must not be dominating for those who live under it, is best, says Pettit. Yet he does not regard all will-based approaches as equal. Of the three that he examines, only the ‘control’ approach ensures that individuals are not dominated, which means, in his view, that its unwelcome decisions should not be resented, but viewed instead as just ‘tough luck’.
In her chapter, Peter replies to Stephen Darwall, who accuses Raz of conflating expertise and legitimate political authority. According to Darwall, such authority depends on a mode of justification involving mutual accountability, which public reason approaches exemplify. The main point of disagreement between Darwall and Raz, says Peter, is not whether expertise is sufficient for legitimate political authority—Raz admits that it is not—but whether, as Raz thinks but Darwall denies, it is necessary. Having clarified the debate, Peter defends the possibility of grounding legitimate political authority in expertise. She thus dismisses Darwall’s critique. She concludes, however, that the mode of justification on which he insists is nevertheless important in democratic societies, in which, on many issues, expertise is not available.
The third chapter is, as stated above, an attempt by Weithman to make sense of Rawls’s account of legitimacy in his later work. Weithman thinks that the answers to the questions raised by this account are to be found in Rawls’s claim, familiar from social contract theory, that political power is the power of the public as a corporate body. In so claiming, Weithman suggests, Rawls follows Immanuel Kant, not
John Locke. Understood as Kantian, his conception of legitimacy is less puzzling, says Weithman, even if some questions persist.
Raz’s contribution, which keeps its original character as a talk, is the first of four to focus on the legitimacy of the state. It starts with a question: Do increasing challenges to state sovereignty require legal theory to be re-thought? In response, Raz observes that, despite these challenges, state sovereignty persists, since no alternative is available. He argues that the success of the assortment of international organizations that challenge state sovereignty depends on their ability to attract loyalty, which depends, in turn, on their ability to recognize the truth of pluralism and respond to diverse interests, preferences, and traditions. Perhaps such value-pluralism is best protected by sovereign states, he says.
In the fifth chapter, Nicole Roughan argues that, to be complete, an account of a state’s legitimacy must evaluate not only its powers and its institutions, but also its officials. She gives three reasons for separate consideration of the third: officials are not only institutional, but also moral agents; important aspects of legitimacy will be missed unless the legitimacy of officials is examined; and the identity of those who exercise powers and control institutions is a normative matter. She proceeds to indicate that, to be legitimate, officials, having claimed a state’s authority, must seek to improve the legitimacy of both its powers and its institutions and, in exercising its authority, must try to behave respectfully towards those subject to its powers and institutions. Roughan finishes by emphasizing that an assessment of the ‘microlegitimacy’ of officials is not all that matters in evaluating the legitimacy of a state as one among many claimants of authority.
Krygier begins his chapter on the rule of law and state legitimacy by pointing out that how the rule of law contributes to state legitimacy turns on how the rule of law is understood. Drawing on earlier work, he recommends looking first to its point, which is, in his opinion, to temper the arbitrary exercise of power. Such tempering, he thinks, even if Marxists and others do not, is something without which a state cannot be legitimate. Krygier then queries the assumption of lawyers and legal philosophers that the rule of law depends simply on the formal rules and procedures of states; it also depends, he insists, on the actual and perceived functioning of those rules and procedures, on tempering non-state arbitrariness, and on passing ‘Hobbes’s test’ of effectiveness.
Part Two ends with Přibáň’s inquiry into the legitimation of the nation state in a post-national world. He employs systems theory, according to which law is one of many normative systems and the state is one of many societal organizations, to make sense of the role of values in the state’s legitimation. He argues that processes of legitimation are constituted by systemic communication between law and politics within and beyond the nation state, communication that involves the internalization of values that are external to these systems. Yet, stresses Přibáň, such values are not transcendental foundations; they are instead internally generated expectations that differentiate between legitimacy and illegitimacies in both politics and law.
Legitimacy beyond the state is the subject of Part Three. In his chapter, Sadurski contends that legitimacy in this sphere should be understood, given the democratic deficit of supranational authorities, as a subject-conferred attribute of specific norms
that generates, as previously mentioned, no more than a duty to respect those norms. The conception of this concept of legitimacy that he favours is a public reason one. It is not Rawls’s view of public reason in the supranational sphere, which, points out Sadurski, cannot be understood as a conception of legitimacy. Rather, it is what he calls Supranational Public Reason sensu stricto, which, he says, not only can be so understood, but can also be detected in the legitimating strategies of supranational bodies such as the European Court of Human Rights.
In the ninth chapter, Cormac Mac Amhlaigh replies to critics of attempts to ground the legitimacy of suprastate institutions in constitutionalism. More specifically, he responds to the objection that suprastate constitutionalism is idealistic due to the improbability of a demos beyond the state. He makes two points: first, the objection relies on a normative understanding of the idea of a demos that is itself potentially idealistic and, second, the objection assumes that idealism in normative theory is problematic, which might be doubted.
As already indicated, Sevel examines Raz’s perfectionist liberalism, an alternative to liberal neutrality. Sevel notes that perfectionists, unlike neutralists, have done little to extend their view beyond the state to international law and institutions. He wonders whether perfectionist liberalism can be a theory of legitimacy in this sphere. His discussion focuses on the neutralist worry that the moral pluralism and the conception of autonomy that are aspects of Raz’s view fail to respect moral diversity and the equal standing of citizens across state boundaries. In particular, he looks at Nussbaum’s claim that Raz’s liberalism is less stable than Rawls’s because it is incompatible with the moral views of many people. Sevel argues that this critique is not persuasive in the state context and, even if it were compelling, it would be less so in the suprastate context, due to well-known attributes of international institutions, including their limited jurisdiction and their limited capacity to enforce their decisions.
Follesdal’s contribution concludes Part Three and the volume as a whole. In it, he tries to bring some order to debates about the legitimacy of international courts. His attempt draws on Raz’s conception of authority and on cosmopolitan theory. Follesdal argues that this approach can reduce apparent confusion about the legitimacy of international courts by explaining the significance of considerations such as states’ consent, states’ compliance, and the legality of courts’ decisions.
These chapters represent a wide diversity of both established and emerging approaches to the problem of legitimacy. They raise the question of what sort of problem it is. They also explore whether an account of the legitimacy of states is adequate for assessing the legitimacy of suprastate institutions. Do the latter have such distinctive features that a different account is needed? As suprastate institutions evolve, as they have rapidly over the last half-century, further reflection will surely be necessary. That reflection will, we hope, be facilitated by these chapters.
I The Control Theory of Legitimacy
Philip Pettit*
This chapter addresses the problem of legitimacy that every state poses and looks at different grounds on which a state might be thought to be legitimate. As I use the term, the problem of legitimacy is that of morally justifying the way a state exercises monopoly power over its adult, able-minded, more or less permanently resident members: in short, its citizens.1 It may be that this exercise of power cannot be justified, of course, in which case we can only view the state in question, despite any good qualities it may possess, as an organization on a moral par with a mafia-like agency. But most of us think that it is at least possible for states to wield political power in a more or less legitimate or justifiable way and the challenge is to identify what would make for their legitimacy.
In focusing on this problem, the chapter assumes that there is no issue about whether or not the system under which the world is divided into states is itself justified. History has left us with that system, which is now close to irreversible. Individual states might choose in principle to renounce their power, but in practice none dares to do so, for fear that others would take over its population and territory.2 They might choose to fuse into larger states, of course—at an unlikely limit, a single world-state—or divide into smaller. But those choices would still preserve the system. That system would cease to obtain only in the vanishingly unlikely event that existing states could agree and entrench an arrangement to let world society operate without anyone exercising political power.
Thus, it makes only speculative sense to ask whether it is justifiable that states should possess the power that they have under the system of states. We have no
* I was greatly helped in revising this chapter by discussion at the conference in Sydney Law School in which the volume originated. I learned in particular from the many astute observations of my commentator, Paul Patton.
1 I put aside here the difficult issues as to how the citizenry ought to be determined, how those who are not adult or able-minded should be identified, and how the state ought to behave towards them.
2 Thus, to anticipate later discussion, the choice of a state to remain in existence is not voluntary— there is no acceptable alternative to its doing so—and so it does not impose its own will on its members and does not dominate them by doing so. See Serena Olsaretti, Liberty, Desert and the Market (Cambridge University Press 2004); Philip Pettit, On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge University Press 2012); Philip Pettit, Just Freedom: A Moral Compass for a Complex World (WW Norton & Co 2014).
Legitimacy: The State and Beyond. First Edition. Edited by Wojciech Sadurski, Michael Sevel, and Kevin Walton. © Philip Pettit 2019. Published 2019 by Oxford University Press.
practical as distinct from theoretical interest in arguing that the system of states ought or ought not to be in place—that the state ought or ought not to exist, as it is often put—since there is no corresponding ‘can’; there is no agent to which we can address a suitable prescription.3 It makes good practical sense, by contrast, to ask whether a particular state is justified in how it exercises the power it has under the system of states. We have a practical interest in knowing the answer, after all, since here there is a ‘can’ that answers to the implied ‘ought’: in saying that it ought to exercise its power in a certain manner, we assume that it can be made to do so.
As the focus on morally justifying how a state exercises its power is distinct from an interest in whether the existence of the state is justified, so it is distinct from a concern with how far the social structure that the state uses its power to establish is morally appealing or just. It is possible for a legitimate state not to establish a fully just order and for an illegitimate state to establish an order that is just in broad outline.4 And so, the two concerns are relatively distinct. Both may be expressed, of course, as concerns with justice. But ours is a concern for the political or vertical justice of the state exercising monopoly power over its citizens, the other a concern with the social or horizontal justice of the relations that the state sets up among its citizenry.5
This chapter is in three progressively longer sections. The first distinguishes the legitimacy of a state from the authority that is sometimes ascribed to the state and explains the sort of obligation—the sort of political obligation—that legitimacy imposes on citizens. The second sketches the case for what I describe as a will-based approach to the problem of legitimacy, indicating its advantages over benefit-based and merit-based alternatives. And the third looks at the different forms that this approach may assume.
This last and longest section argues in favour of a version of the will-based approach that has two characteristics: it takes the state to be a proper agency, not just an apparatus that mediates the agency of individuals; and it takes legitimacy to require not that citizens individually consent to the state or enjoy collective command over the state, but that they share equally in imposing a controlling discipline on what the state does in their name. I have defended this control theory elsewhere on premises drawn from the republican tradition of thought.6 The aim here is not to recapitulate
3 Of course, we might conceivably want to know whether we should welcome the existence of the system of states on the ground that if it is not the best possible system, it is at least better than the state of nature. It is possible to read Kant’s discussion of the state in The Doctrine of Right on these lines. See Arthur Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy (Harvard University Press 2009); Kyla Ebels-Duggan, ‘Kant’s Political Philosophy’ (2012) 7 Philosophy Compass 896.
4 See A John Simmons, ‘Justification and Legitimacy’ (1999) 109 Ethics 739; Philip Pettit, ‘Justice, Social and Political’ in David Sobel, Peter Vallentyne, and Steven Wall (eds), Oxford Studies in Political Philosophy, vol 1 (Oxford University Press 2015); Anna Stilz, ‘The Value of Self-Determination’ in David Sobel, Peter Vallentyne, and Steven Wall (eds), Oxford Studies in Political Philosophy, vol 2 (Oxford University Press 2016).
5 As Martin Krygier points out, the word ‘arbitrary’ is often used to indict a system of law on both counts: on the one side, for being imposed on an unjustified, wilful basis; on the other, for having an unjustified character or content. See Martin Krygier, ‘The Rule of Law: Pasts, Presents, and Two Possible Futures’ (2016) 12 Annual Review of Law and Social Science 199.
6 See Pettit, On the People’s Terms (n 2); Pettit, Just Freedom (n 2); Philip Pettit, ‘The General Will, the Common Good, and a Democracy of Standards’ in Yiftah Elazar and Geneviève Rousselière (eds), Republicanism and the Future of Democracy (Cambridge University Press 2019).
that defence so much as to show that the theory compares favourably with both the alternatives outside and the alternatives inside the will-based family of theories.
A. Legitimacy, Authority, and Political Obligation
1. Legitimacy and authority
The problem of whether a state’s exercise of monopoly power is justified or legitimate is distinct from the question as to whether it has the right, on an independent basis, to exercise such power: that is, the right to coerce its citizens into obeying its laws, punish them for disobedience, tax them for resources, and do all of this to the exclusion of any competitor. If its exercise of power can be morally justified, of course, then it may be right that the state should act as it does. And under one or another morally justified system of rights, it may be said then to have the right or authority to impose on members in that particular way. But that is not to ascribe an independent right or authority to the state, only a right relative to that system of rights and that justification.7
I take the issue of whether the state’s exercise of monopoly power is justified or legitimate, not only to be distinct from the question of whether the state has the independent right to exercise such power, but also to be more basic. On this assumption, the state has a certain authority because it is legitimate, and not the other way around. The assumption marks a difference from any approaches that first try to establish the authority of the state and then aim to derive its legitimacy from its authority.8
Those who want to put authority first often do so out of a wish to argue that on that same independent basis citizens have a certain duty or obligation to their state: that they have a duty akin to the directed obligation debtors have to their creditor, or children to their parents. But even if the state is an agent in its own right, as I argue later, the idea that citizens have this sort of independent obligation or duty towards it is not immediately intuitive. They may have a duty of obligation to the state, as we shall now see, only to the extent that the state is legitimate.
2. Legitimacy and political obligation
Suppose that the state exercises its monopoly power legitimately or justifiably; put aside for the moment the fact that legitimacy is bound to come in degrees. What follows about how citizens morally ought to behave: about how it is right that they should treat the state and the state’s initiatives? What is the political obligation that goes with the legitimacy of a state?
7 See Allen Buchanan, Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law (Oxford University Press 2002).
8 See Fabienne Peter, ‘Political Legitimacy’ in Edward N Zalta (ed), The Stanford Encyclopedia of Philosophy (summer 2017 edn) <https://plato.stanford.edu/archives/sum2017/entries/legitimacy/> accessed 16 March 2018.
One possibility is particularly salient, and commonly invoked in the literature.9 We might say that other things being equal the citizens of a legitimate state ought to obey its laws: that while they may not be obligated towards the state in the directed manner associated with an independent authority, they are obliged in a more general sense to display obedience. The idea is that if the way the state makes and applies law is legitimate, then the compliance of citizens with that law is required pro tanto— required other things being equal. On this account, political obligation is nothing more than legal obligation: the obligation to obey the law.
The identification of political with legal obligation is hardly persuasive, however. For the legal obligation of citizens seems to derive from the character of the laws imposed rather than from the legitimacy of the state that imposes them. Even if the state failed to be legitimate, citizens would be plausibly required pro tanto to obey established, relatively fair laws. The fairness of the laws would argue in itself for conformity; and this, all the more so, when those laws provide a basis for mutual expectation and coordination among the citizenry. Indeed, if the laws established by a legitimate state were unfair—if they imposed unjust or unequal requirements— then it is not clear that citizens would necessarily be required to obey them; if they were required to obey, that could only be because the unfairness was not significant and the laws played an important coordinating role.10
The pairing of legal obligation with the fairness of the laws rather than with the legitimacy of the state is further supported by the fact that non-citizens who reside in the state are obliged equally with citizens to obey the law. There is a sense in which a state does not impose monopoly power on such non-citizens, since they are formally free to leave, assuming they are citizens of another state. And so, we should not expect the legitimacy of the state to be relevant to whether or not they are obliged to obey the law of the country they live in. In their case, then, it is more likely to be the fairness of the established laws, or their role in establishing mutual expectations, that grounds an obligation to obey the law. And if that is the ground of the obligation in their case, it is plausibly the ground in the case of citizens too.11
If the state exercises its monopoly power in a way that is justified, then, is there any more distinctive obligation than the obligation to obey the law supported by this legitimacy? Under almost any conceivable political regime, people are going to differ in how far they are satisfied with the laws imposed, having rival interests in what laws there should be and diverging in their opinions about the justice or fairness of competing candidates.12 The circumstances of politics, as Jeremy Waldron
9 See Christopher Heath Wellman and A John Simmons, Is There a Duty to Obey the Law? For and Against (Cambridge University Press 2005).
10 There may be a range of reasons apart from the fairness of the laws and their role in grounding mutual expectations for why citizens and other residents should obey the law. See Buchanan, Justice, Legitimacy, and Self-Determination (n 7).
11 Admittedly, there is another possible ground of obligation in the case of non-citizen residents. It may be said that those who choose to reside in a state where they are not citizens, with the permission of the authorities, commit implicitly to conforming with the local laws and regulations. In that case, their commitment to conform would help to explain why they are obliged to obey the laws, and perhaps even to obey unfair laws.
12 See Thomas Nagel, ‘Moral Conflict and Political Legitimacy’ (1987) 16 Philosophy and Public Affairs 215.
says, are circumstances of disagreement.13 That raises a question about the obligations that fall on citizens who disagree with some laws and are disposed to contest them. And it is plausible that the legitimacy of the state has implications for the obligation of citizens in this domain.
If the state were illegitimate, so that its exercise of monopoly power were not justified, then we might say that other things being equal—collateral costs being intuitively acceptable, for example—it would be permissible for citizens to look for radical regime change. If the state was legitimate, however, then changing the regime would seem to be ruled out: it would amount to overthrowing a body that by hypothesis was justified in how it acted. And so, a plausible proposal is that the legitimacy of a state imposes on citizens the conditional obligation, if they contest the law, to contest it within the system: that is, to contest it in a manner that does not require regime change.14
We have been putting aside the fact that legitimacy is bound to come in degrees. Taking that fact into account, the proposal supported here is that to the extent to which the state is legitimate, there is an obligation on citizens to contest any public laws or decisions they regard as unjust—or indeed illegitimate in their particular genesis—only within the system. I shall assume in what follows that this conditional, contestatory obligation is the correlate for citizens of any legitimacy enjoyed by their state. That this is the correlated obligation does not presuppose that people ought to contest the laws or decisions with which they disagree, but merely that such contestation is unobjectionable or permissible. And, as noted, it does not presuppose that people ought to seek the overthrow of any illegitimate regime; collateral costs might argue for trying to work for change within the system rather than outside it.
In presupposing the permissibility of contesting a legitimate state only within the system, the proposal naturally suggests that the more or less legitimate state ought to establish ways in which citizens may contest the laws that it enacts or proposes. While any plausible regime will offer citizens many different ways in which they may challenge its laws, there is always one mode of contestation—civil disobedience— that any system is required by its very design to leave in place. This might consist in publicly breaking the law to which you object, and not protesting about your conviction and punishment; such behaviour would communicate, on the basis of shared assumptions, that you oppose the law in question and hold that it should be changed. Or, in a slight variant, it might consist in publicly breaking some other law, not protesting your conviction or punishment, and communicating explicitly that
13 Jeremy Waldron, Law and Disagreement (Oxford University Press 1999).
14 See Pettit, On the People’s Terms (n 2) ch 2. Overthrowing the state and achieving radical regime change need not involve armed rebellion. It might take the form, for example, that it took when the American founders proposed the new constitution of 1787 and established it with popular, electoral support; no such change of constitution was possible under the regime as it stood: that is, under the Articles of Confederation that had united the states. But it is possible to take a narrower or wider view of the processes for amending a constitution that are allowed within it. Thus, while a narrower view would suggest that the US constitution can only be amended under Article 5, Akil Amar argues that it would not amount to a rejection of that constitution to amend it under a countrywide majority vote. See Akil Amar, ‘Philadelphia Revisited: Amending the Constitution outside Article V’ (1988) 55 University of Chicago Law Review 1043.
you are doing this as a means of challenging the law to which you object. Such civil disobedience constitutes a form of contestation within the system, since it does not challenge the power exercised by the state in convicting or punishing you; it does not suggest that the state is illegitimate.
To sum up these observations, then, the legitimacy of a state is fixed by how far its exercise of power over citizens is morally justified. Such legitimacy is distinct from, and more basic than, any independent authority that a state might claim to exercise this power and any independent duty that citizens might be said to have towards the state; and the duty with which it correlates is not an obligation to obey the law, but an obligation on the part of citizens to contest the law, if they contest it, within the system.
B. Benefit-Based, Merit-Based, and Will-Based Theories
1. Three candidate legitimators
Any legitimator of a state, any justifier of how a state exercises its power, should explain why the citizens ought to contest its laws only within the system: why they should seek amendments only in a way that does not require starting again from scratch. Thus, it should explain why citizens ought to seek a change in the constitutional order only under the accepted ways of amending the constitution. Assuming an electoral democracy, it should explain why they ought to seek other changes only at the ballot box or in the courts, in the media, in public demonstrations, or in acts of civil disobedience.
Very broadly, there are three kinds of candidates that might be proposed for the role of state legitimator: the role of providing on its own a sufficient ground for determining legitimacy. One is that the state delivers some significant benefit to citizens, whatever that may be taken to be. A second is that the state is guided by personnel who are suitably meritorious: they enjoy appropriate expertise in the issues they confront and, being equally responsive to every member of the citizenry, they display appropriate virtue. And a third is that the state does not impose an independent will on its citizens: it does not hold them in its power.
In each case, the legitimator may come in degrees and confer legitimacy only in a corresponding measure. But again, for convenience, I shall generally speak as if legitimacy is an on- off property: a feature that the state has or fails to have, period.
2. The republican tradition
There is a long republican tradition of thought under which the first two approaches, respectively benefit-based and merit-based, are not plausible candidates for legitimating the state. I shall draw here on this tradition in order to identify a common failure in the approaches and to provide ground for exploring the will-based approach instead.
The republican tradition begins with figures like Polybius, Cicero, and Livy, who defended the Roman Republic. And it is associated with those in later ages— for example, in the Italian Renaissance, the English revolution, the American war of independence—who took that republic to hold out an ideal for how the state should be organized in their own era. The paramount ideal in these writers was that in a republic all citizens should count as equally free persons. And their guiding institutional assumption was that citizens could only achieve that status if they were protected in the same range of personal choices—the basic liberties, as these were often known—under a mixed constitution: that is, before a law over which they maintain a multi-centred, multi-level, form of more or less equally shared control.15
In the Renaissance period, these writers included Machiavelli of the Discourses on Livy’s History; in the period of the English revolutions, authors such as James Harrington, John Milton, and Algernon Sidney; and in the eighteenth century, radicals like the authors of Cato’s Letters, American rebels like John Adams and James Madison, and English supporters of the American cause like Joseph Priestley and Richard Price. They uniformly celebrate the idea that being a free citizen means enjoying independence from others in the exercise of the basic liberties that any law should recognize. And they equally support the multi-modular system of government associated with the mixed constitution. Variously interpreted in the tradition, this is an arrangement that distributes power among different authorities, puts those authorities in a position to check and balance one another, and exposes them to the electoral and contestatory power of the people as a whole.
15 There is now an extended literature on the history and nature of republican thought and on the break that classical liberal modes of thought initiated; for an overview, see Frank Lovett and Philip Pettit, ‘Neo-Republicanism: A Normative and Institutional Research Program’ (2009) 12 Annual Review of Political Science 18. Contemporary republicanism has its origins in Zera S Fink, The Classical Republicans: An Essay in the Recovery of a Pattern of Thought in Seventeenth Century England (Northwestern University Press 1962); Caroline Robbins, The Eighteenth Century Commonwealthman (Harvard University Press 1959); and especially JGA Pocock, The Machiavellian Moment: Florentine Political Theory and the Atlantic Republican Tradition (Princeton University Press 1975), which first revived interest in the classical republican writers and charted the historical continuity of their political ideas. Quentin Skinner argued in a number of essays, later collected (and somewhat revised) in Quentin Skinner, Visions of Politics, vol 2: Renaissance Virtues (Cambridge University Press 2002), that these works had failed to recognize that classical republicanism did not endorse an allegedly ‘positive’ view of freedom as equivalent to a right of participation in government. And building on this insight, Pettit—and Skinner himself—cast the republican conception of freedom as one according to which it is the absence of domination or dependence on the arbitrary will of another, and not the absence of mere interference, that matters. See Philip Pettit, ‘Freedom and Antipower’ (1996) 106 Ethics 576; Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford University Press 1997); Quentin Skinner, Liberty before Liberalism (Cambridge University Press 1998). This idea of freedom as non-domination has become the crucial unifying theme for those who work within the neo-republican framework, although of course within that frame there are also some differences of emphasis and detail. See Philip Pettit, ‘Keeping Republican Freedom Simple: On a Difference with Quentin Skinner’ (2002) 30 Political Theory 339. For a recent, alternative history of thinking about freedom, see David Schmidtz and Jason Brennan, A Brief History of Liberty (Wiley Blackwell 2010).
3. Against the benefit-based approach
The benefit-based approach to legitimating the state is rejected in this long tradition on the basis of a spectre that became particularly salient in the eighteenth century: that of the benevolent despot. The idea was that if the benefit provided by a state could legitimate it as a regime, then the dictatorship of a well-disposed monarch—say, a dictatorship that delivered a just and efficient system of law— would be legitimate. And that was taken in this tradition to be utterly implausible, even absurd.
Freedom in the sense in which it requires independence from others in the exercise of the basic liberties implies that no slave can be free, not even one whose master is wholly benevolent; if the slave may choose to act as he or she wishes, this is only possible so long as that is the wish of the master. Any master or dominus, no matter how indulgent, has the power to interfere at will with his or her slaves in the exercise of their basic liberties. And this dominatio, as the Romans called it,16 is enough to put the will of the master in control. Freedom requires the absence of such domination, then, not just the absence of the interference it might make possible. It requires that so far as possible no one has the power of interfering at will in a person’s exercise of their basic liberties: that is, in the exercise of suitably established, suitably extensive choices that each can exercise without detriment or disadvantage to others.17 Thus, in the words of Algernon Sidney, writing in the seventeenth century, ‘he is a slave who serves the best and gentlest man in the world, as well as he who serves the worst’.18
This republican understanding of freedom implies that to live under a dictator, no matter how benevolent, is to be effectively a slave of that authority. Even if you can exercise a suitable range of basic liberties under such a regime, your access to those liberties—your access to a law that defines and defends them—is dependent on the willingness of the ruler to establish them. And, of course, your unhindered exercise of the liberties that are actually established is dependent equally on the ruler’s willingness to allow you in particular to benefit from them in the same manner as others. This means that in relation to the ruler you are in the position of a slave: a lucky slave, to be sure, but still a slave.
The supporters of the American cause in the eighteenth century took up this theme when they challenged the legitimacy of British Colonial Rule, as it had come to be conceptualized and practised in the previous decade. The Westminster Parliament may have proved itself to be gentle and benevolent when, in response to American complaints, it withdrew the Stamp Act in 1766. But at the same time, it claimed in the Declaratory Act, as it was called, that it enjoyed the status of a master; it asserted as ‘of right’ that it had ‘full power and authority to make laws and statutes’ to bind the Americans. This is one of the things that particularly enraged the colonists, conscious as they would have been that as a people they were then no
16 See Frank Lovett, A General Theory of Domination and Justice (Oxford University Press 2010) app.
17 See Pettit, On the People's Terms (n 2).
18 Algernon Sidney, Discourses Concerning Government (Liberty Classics 1990) 441.
better than slaves in relation to the British. Richard Price expressed their grievance particularly well. Commenting that individuals under the power of masters ‘cannot be denominated free, however equitably and kindly they may be treated’, he argued that the same lesson applied to societies like Great Britain and its colonies: ‘This is strictly true of communities’, he said, ‘as well as of individuals’.19
Assuming a suitable conception of the basic liberties, it is surely plausible that freedom in the republican sense, freedom as non-domination in the exercise of these liberties, is an ideal we ought all to cherish.20 And it is equally plausible that the benevolent dictatorship, whether it be that of an individual, an elite, or a colonial power, is a form of subjection or slavery that compromises people’s freedom. This being so, it is hard to see how the mere fact that a state produces various benefits can justify the way in which it exercises power and thereby establish its legitimacy.
No matter what the benefits it produces, then, a state may resemble a benevolent dictatorship in the way in which it exercises power and may count as a dominating presence in its people’s lives. And when a state has this profile, it is hard to see why the subjects should be morally obliged to operate within the system, even when there are no collateral reasons to refrain from seeking regime change. Perhaps Kant is the only otherwise liberal philosopher who ever held that, regardless of domination by the state, its people were always obliged to operate within the system. He argued that the subjects of any government, even a government that behaves ‘tyrannically’, should only enjoy the freedom of the pen: the freedom to try to persuade the ruler to act differently, but not to call ‘by word or deed’ for a change of regime.21
The complaint about the benevolent dictator supports a critique of many different sorts of arrangements, each identifying a different benefit that a state might achieve for its citizenry. The fact that a particular state promotes the aims of citizens more effectively than those citizens could themselves promote them;22 the fact that the state is epistemically reliable in the judgements of fact and value that it makes;23 the fact that by selecting rulers by lottery it allows power to rotate on an impartial basis;24 or the fact that it delivers a form of governance that could be justified to reasonable citizens:25 these will all fail, at least on their own, to justify the state’s exercise of power.26 For, regardless of the advantages for citizens, the state’s exercise
19 Richard Price, Political Writings (Cambridge University Press 1991) 77–78.
20 Whether or not non-domination is used to define freedom, of course, it may still be regarded as an ideal worth promoting in social life. The argument in the chapter is strictly independent, then, of the republican linkage between non-domination and freedom.
21 Immanuel Kant, Practical Philosophy (Mary J Gregor tr and ed, Cambridge University Press 1996) 298–392. For an argument in support of a more congenial reading of Kant, see Ripstein, Force and Freedom (n 3).
22 See Joseph Raz, The Morality of Freedom (Oxford University Press 1986).
23 See David Estlund, Democratic Authority: A Philosophical Framework (Princeton University Press 2007).
24 See Alexander A Guerrero, ‘Against Elections: The Lottocratic Alternative’ (2014) 42 Philosophy and Public Affairs 135.
25 See John Rawls, Political Liberalism (Columbia University Press 1993).
26 The different proposals gestured at in this paragraph are loosely characterized as legitimating the state by reference to a benefit it provides, and the different authors cited are only loosely associated with those proposals. The list is meant to be indicative only of the different directions that a benefit-based approach might take.
of power may still be objectionable on the most basic of grounds, namely, that it puts those citizens severally as well as collectively under the rule of an independent will; it makes them into slaves, albeit slaves of a relatively benign master.27
While these and other benefits might not be sufficient for legitimating a state, however, it might be asked if they are necessary for legitimacy. Without considering the question properly, it is worth noting that on the approach defended later in the chapter, they are not necessary in one sense, but they are in another. They are not necessary in the sense that they are not part of what determines or grounds legitimacy: the fact, according to the line taken, that the state is under the equally shared control of the people. But, still, some of them may be necessary in the sense that it is barely conceivable that the state might be controlled in that way without delivering such benefits. They may not themselves help to ground legitimacy, but still be entailed by the realization of its proper ground.
4. Against the virtue-based approach
As resources provided by the republican tradition enable us to challenge any benefitbased candidate for legitimating a state, so they also make it possible to challenge any merit-based candidate for that role. An example of such a candidate is provided recently by Daniel Bell in his defence of what he describes as the China model.28 This is a model of how state authorities might be selected that idealizes current practices in the People’s Republic of China. The idea is that a state’s exercise of power would be justified if it was guided by those who proved meritorious or worthy by the standard of such practices: they proved not only to have the expertise required for governing well, but also the virtue to govern impartially.
Assume for the moment that we could rely on a ruler or set of rulers to be suitably expert and, in particular, to be virtuous or impartial. Would that really support the claim that the state is legitimate? The suggestion is that even if the benevolence of a dictatorial ruler or set of rulers—presumptively, expert or knowledgeable rulers— would not be enough to make the state legitimate, their virtue would.
There is indeed a difference worth registering between the benevolent and the virtuous ruler. For a ruler might be benevolent as a result of certain contingencies: for example, the happy accident of wanting to be remembered kindly in history. But a ruler would not be virtuous if the benevolence they displayed were dependent on that sort of contingency. In order to count as virtuous, we may suppose, they would have to be disposed to be benevolent, not just as things actually stand—not just so long as they want to enjoy historical esteem—but robustly over variations in the personal attraction or ease of displaying such goodwill. In short, they would have
27 A critique that converges with this complaint is that in benignly imposing on its citizens the state usurps their judgements about what is in their interest and, assuming a paternalistic role, does not recognize or respect them as autonomous subjects. See Seana Shiffrin, ‘Paternalism, Unconscionability Doctrine, and Accommodation’ (2000) 29 Philosophy and Public Affairs 205.
28 Daniel A Bell, The China Model: Political Meritocracy and the Limits of Democracy (Princeton University Press 2015).
to be robustly benevolent.29 In a phrase of Niko Kolodny’s, they would have to be ‘resolutely disposed’,30 not just disposed in a more or less adventitious manner, to deal benevolently with subjects.
Would this virtue in rulers do more than mere benevolence to secure the freedom of citizens who live under their power, gaining access to a protected range of basic liberties by grace of their goodwill? The problem with the benevolent dictator is that citizens depend on the ruling will remaining unchanged in order to enjoy their basic liberties, so that in that sense they are subject to that will; let it change and their access to the basic liberties may cease. Is there a similar problem with the virtuous dictator? Or does the dispositional robustness of virtue make for a crucial difference?
I do not think that it does. For even if the rulers were virtuously or robustly disposed to deal well and fairly with the citizens, they would still be liable to lose that disposition or liable not to manifest it on this or that occasion. Consistently with deserving the name of virtue, it could not be hard-wired into their make-up and could not deprive them of choice, making them into involuntary do-gooders. Thus, like the rest of us, they would be exposed to the possibility of corruption, on the one side, and more or less spasmodic episodes of akrasia weakness of will—on the other. And that means that the citizenry would still depend on the will of the rulers for being able to access and exercise the basic liberties established in law. If that will failed to be virtuous, whether permanently or occasionally, they would cease to enjoy that benefit. The virtuous dictator may behave well more reliably than the merely benevolent counterpart, but citizens would still depend on the survival of the dictator’s goodwill and still be subject to domination.
This critique of the virtuous dictator extends to any regime where the exercise of power is supposed to be legitimated by the virtue of those in authority. As mentioned, it would lose force if the authorities were supposed to be incapable of losing their virtue, or acting against it, and in that respect not to be as other human beings. But such a supposition would be fantastical. And in any case, it is a long republican theme that far from the power of office tending to support or even elicit virtue in rulers, it usually has the effect of corrupting them. Providing new opportunities for abuse, and often providing a better prospect for hiding any abuse practised, power tends to destroy virtue rather than maintain it. Lord Acton gave expression to a lesson long rehearsed in the tradition when he said: ‘Power tends to corrupt, and absolute power corrupts absolutely.’31
This lesson about the effects of power has strong empirical support in the social science of regulation.32 And equally it is borne out in the long history of politics, where patrimonial interests have always threatened to corrupt politics.33 While most
29 See Philip Pettit, The Robust Demands of the Good: Ethics with Attachment, Virtue, and Respect (Oxford University Press 2015).
30 Niko Kolodny, ‘Rule over None II: Social Equality and the Justification of Democracy’ (2014) 42 Philosophy and Public Affairs 287, 295.
31 Letter from Lord Acton to Archbishop Creighton (5 April 1887) accessed 10 October 2018.
32 See Pettit, Republicanism (n 15) ch 7.
33 See Francis Fukuyama, The Origins of Political Order: From Prehuman Times to the French Revolution (Farrar, Straus and Girous 2011).
