DANIEL VIEHOFF
Democratic Equality and Political Authority
Many, philosophers and democratic citizens alike, consider democracy the best form of government, not because it is likely to make the best decisions (it may not), but because it grants citizens an equal say in the making of the laws. Many also hold that democracy has authority other regimes lack, and that the democratic provenance of laws at least sometimes gives us a duty to obey them. Finally, some believe that the two are connected: as citizens of a democratic state, we have a moral duty to obey (at least some) democratically made laws because they are the outcome of an egalitarian procedure.1 On this view, we can at least sometimes
For comments on earlier versions of this article, or valuable conversations on the issues that it treats, I am grateful to Thomas Christiano, David Estlund, Nir Eyal, Evan Fox-Decent, Michael Fuerstein, Amanda Greene, Andrew Hall, Niko Kolodny, Larry Lessig, Jacob Levy, Claudio Lopez-Guerra, Reidar Maliks, Jane Mansbridge, Fred Neuhouser, David Owens, Thomas Pogge, Massimo Renzo, Jon Rick, Thania Sanchez, Tommie Shelby, Bob Stern, and Annie Stilz; to audiences at Columbia University, Harvard University's Safra Center for Ethics, and the University of Sheffield; and to the participants in a workshop on my book manuscript on "The Authority of Democracy" held by the Groupe de recherche interuniversitaire en philosophie politique (GRIPP) at McGill University. For extremely helpful written comments and suggestions, I am particularly indebted to Arash Abizadeh, Eric Beerbohm, Chris Bennett, Pablo Kalmanovitz, Felix Koch, Joseph Raz, Juri Viehoff, Jeremy Waldron, and two anonymous reviewers for this journal.
1. The first is held by, among others, Charles Beitz, Joshua Cohen, Jane Mansbridge, and Jeremy Waldron: Charles R. Beitz, Political Equality: An Essay in Democratic Theory (Princeton, N.J.: Princeton University Press, 1989); Joshua Cohen, "Procedure and Substance in Deliberative Democracy," in Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg (Cambridge, Mass.: MIT Press, 1997); Jane J. Mansbridge, Beyond Adversary Democracy (Chicago: University of Chicago Press, 1980); Jeremy Waldron, Law and Disagreement (Oxford: Oxford University Press, 1999). The second is claimed by John Dunn, David Estlund, Peter Singer, and (perhaps) Jeremy Waldron (but see note 42): John Dunn, Democracy: A History (New York: Atlantic Monthly Press, 2005); David Estlund, Democratic Authority: A Philosophical Framework (Princeton, N.J.: Princeton University Press, 2008); Peter Singer, Democracy and Disobedience (Oxford: Clarendon Press, 1973); Waldron, Law and Disagreement. The third (and so a fortiori the
© 2014 Wiley Periodicals, Inc. Philosophy & Public Affairs 42, no. 4
338
Philosophy & Public Affairs
respond to someone who asks why she ought to obey a law: "Because this law was made democratically, via procedures in which all of us had an equal say; and by disobeying it you fail to respect our equality." The purpose of this article is to justify this egalitarian authority claim.
The significance of the claim is twofold. First, clarifying how, if at all, political equality can contribute to democracy's authority helps us better understand what forms of equality are normatively central to democracy. Second, and more practically, if the egalitarian authority claim is correct, it may provide support for the widely held conviction that democracy has special authority nondemocratic regimes lack.2 Most theorists of political authority hold that, while many states have some authority (based on coordination, consent, or expertise) over some of their purported subjects and with regard to some issues, no state as we know it has general authority, over all its subjects on all issues on which it makes law.3 The state's law thus has "authority gaps": some of the laws the state makes and enforces are not morally binding for their purported subjects.4 If a law's democratic provenance sometimes justifies the subjects' duty to obey, then democratic states may have fewer such
first and second) has recently been endorsed by Allen Buchanan, Thomas Christiano, and Scott Shapiro: Allen Buchanan, "Political Legitimacy and Democracy," Ethics 112 (2002); Thomas Christiano, "The Authority of Democracy," Journal of Political Philosophy 12 (2004); Thomas Christiano, The Constitution of Equality: Democratic Authority and Its Limits (Oxford: Oxford University Press, 2008); Scott Shapiro, "Authority," in The Oxford Handbook of Jurisprudence & Philosophy of Law, ed. Jules L. Coleman and Scott Shapiro (New York: Oxford University Press, 2002). Rousseau too may have been committed to this third claim.
2. It should be clear that democracy could in principle have special authority on other grounds too. (For instance, democracy may have special epistemic virtues; or we may have a special duty to obey democratic decisions because disobedience undermines institutional stability, and living under democratic institutions is preferable to any feasible alternative.) Conversely, democracy could also lack authority that other arrangements have. This article will not discuss these possibilities. Finally, insofar as democracy is not the only way of upholding political equality—coin flipping, lot drawing, and the like being others— the argument for the egalitarian authority claim may bear on the authority of such alternative procedures as well. I briefly discuss this issue in the conclusion.
3. Leslie Green, The Authority of the State (Oxford: Clarendon Press, 1988); Joseph Raz, "The Obligation to Obey the Law," in The Authority of Law (Oxford: Clarendon Press, 1979); A. John Simmons, "Political Obligation and Authority," in TheBlackwell Guide to Social and Political Philosophy, ed. Robert L. Simon (Oxford: Blackwell, 2002).
4. The subjects may nonetheless have a duty to conform to the law, if the law tracks their law-independent obligations (not to kill, murder, rape, and so on). But while this may be true of some laws, it is unlikely to be true of most.
339
Democratic Equality and Political Authority
authority gaps than their nondemocratic counterparts.5 This would bear on both the duties of citizens and the moral standing of the state.
To establish the egalitarian authority claim, it is not sufficient to show that democratic procedures uphold equality among us and that we have a duty to obey democratic decisions, since the two could both be true yet unconnected. A Hobbesian democrat, for instance, may hold both that only democratic decision making treats each citizen with equal respect (so we have reason to democratize political institutions) and that political authority depends only on the state's capacity to enable order and coordination, a capacity possessed by democratic and nondemocratic institutions alike. A Hobbesian democrat would thus value democracy yet not attribute any special authority to it. Or to put the underlying thought in another way: the Hobbesian democrat accepts that authority need not be inimical to equality, nor equality to authority. But she does not think that either makes a distinctive contribution to the other. This essay, in contrast, argues that realizing a particular form of equality requires authority, and that the value of realizing such equality in turn justifies the conclusion that we have a duty to obey some laws because they are the outcome of an egalitarian procedure.
But before this article develops its own justification of the egalitarian authority claim, it considers two prominent arguments that link democratic authority to a concern for equality. Both are ultimately unsuccessful; but their failures are instructive, and will help us identify the conditions that a plausible defense of the egalitarian authority claim must meet. The first argument (which I discuss in Section II, after explaining, in Section I, what I take authority and democracy to consist in) appeals to the simple idea that fairness disallows granting ourselves special privileges that we deny to others; and it suggests that fairness thus requires obeying democratic decisions rather than acting on our own judgment of the right policy. I show that this argument fails; and its failure indicates that we must invoke a less formal, more substantive understanding of equality to justify democratic authority. The second argument, recently suggested by Thomas Christiano, does just that. Democratic authority is, on this account, not justified by the formal demand for equal treatment but by the distinctive value of showing
5. I leave it to the reader to fill in the steps required to bridge the gap between antecedent and consequent.
340
Philosophy & Public Affairs
public equal respect to our fellow citizens. Such public equal respect requires treating as authoritative the democratic decisions in which citizens had an equal say. This argument does not, however, succeed in its goal either. While it plausibly establishes the value of democratic institutions, it does not provide grounds for a duty to obey democratic decisions (Section III). This article thus develops an alternative argument, according to which egalitarian procedures have authority because, by obeying them, we can avoid acting on certain considerations that must be excluded from our intrinsically valuable egalitarian relationships. The authority of democratic decisions rests on the egalitarian fact that none of us has more of a say than any other, not on the further fact (crucial to the previous argument) that each of us has a positive say (Sections IV and V). The argument in turn helps us understand the limits of democratic authority: it is constrained not only by a demand for equality, and by considerations of justice, but also by a requirement of mutual concern without which our egalitarian relationships lack their distinctive value (Section VI).
I. AUTHORITY AND DEMOCRACY
Authority is centrally the power to morally bind others by giving them directives that they must follow. So if a state has genuine authority on the matter on which it legislates, its subjects are duty-bound to obey its laws. Consider a typical instance of the exercise of political authority. A state lays down a law prohibiting unlicensed gun ownership, or requiring citizens to pay a certain amount in income tax. That a law with this particular content—"do x"—has been enacted gives the subjects a reason to act as the law requires—to do x. The authoritative directive or law provides, however, not just any reason but is meant to largely determine the subject's action, or be binding: That a law requires citizens to do x (to pay a certain amount in taxes) is not just a reason for doing x that is to be added to all the other reasons they have for and against doing x. Rather, when a law on a particular matter has been enacted, citizens are to set aside (their own judgment of) the reasons that bear on that matter and to follow the authority's determination instead.6 As Locke writes in
6. A more detailed discussion would highlight that an authority's purview, and with it the preemptive force of its directives, may be limited: it might, for instance, only cover economic reasons, but not others, which may thus still be considered by the subject. But
341
Democratic Equality and Political Authority
describing the establishment of political authority, "all private judgment of every particular member being excluded, the community comes to be umpire."7 Crucially, the authority's decision is binding for us even if the authority's judgment on the matter at issue is substantively mistaken— if, for instance, there is no good reason to require gun owners to have a license, or for citizens to pay as much (or as little) income tax as the law demands. If democratic decisions have authority for me, then I must abide by them even if I believe they are mistaken, and even if my belief is true.
The features of authority just described can be usefully captured by thinking of authoritative directives as reasons of a special sort: they are content-independent and preemptive.8 A reason for doing x is content-independent if it is not based on the nature or merit of x, but on facts extraneous to x. Thus, a directive requiring doing x is intended to be a content-independent reason for doing x because it is the fact that such a directive has been given, rather than the nature of action x itself, that is meant to provide a reason for the agent to do x. And a reason for doing x is preemptive if it is both a reason for doing x and a (second-order) reason for not acting on certain otherwise relevant considerations against doing x, or for "excluding" them from among the reasons that figure in our deliberation. This captures the idea that we must obey, or defer to, the authority: If an authority requires that I do x (which falls within the authority's purview), I am not only meant to add the fact that it requires doing x to the reasons I consider relevant for and against doing x. I am also meant to set aside (at least some of) these other reasons, which are thus replaced or preempted by the directive.9
For the purposes of the following discussion, I take democracy to be a particular procedure for making binding legislative decisions for a group
the law's purview is not normally limited in this way: it purports to take into account all the relevant reasons and to give all-things-considered directives. It is on such cases that I focus here.
7. John Locke, "The Second Treatise," in Two Treatises of Government and A Letter Concerning Toleration, ed. Ian Shapiro (New Haven, Conn.: Yale University Press, 2003), p. 137 (§87).
8. Joseph Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986).
9. That obligations in general have an exclusionary structure has also been suggested by T. M. Scanlon and defended by David Owens. See T. M. Scanlon, What We Owe to Each Other (Cambridge, Mass.: Harvard University Press, 1998), pp. 156-57; and David Owens, "Rationalism about Obligation," European Journal of Philosophy 16 (2008).
342
Philosophy & Public Affairs
of people. What sets democracy apart from other such procedures, like aristocracy or monarchy, is that all or almost all of those to whom the decision is meant to apply are also granted an equal say in its making. The focus on political procedures is not meant to deny that democracy may also be associated with a distinctive social ethos, or that democracy may also require laws with particular content, for example, protecting free speech. I start from democratic procedures because my interest is in understanding how such procedures can imbue even substantively mistaken laws with authority these laws would otherwise lack. (The connection between democratic procedures and a distinctive social ethos will be central to the argument in Sections IV and V. How the procedures' authority may be restricted by substantive commitments will briefly be discussed in Section VI.)
There are many different institutional arrangements that have been taken to implement the ideal of an equal say in the making of the law. For the purposes of this article, I will simply assume that citizens have an equal right to vote on laws or (as is usually the case in modern democracies) for representatives who in turn vote on laws, as well as a right to participate as equals in prevoting deliberation, to run for office, and so on. Democratic citizens are, furthermore, equal before the law: the law applies to all, and to all equally. Both the institutional complexities of modern states and a more developed grasp of the values central to democracy might require refining this very general characterization. Nonetheless, I rely on it as a helpful starting point for understanding core features of democracy. Notice that many (perhaps all) existing democratic states fall short of even these quite minimal requirements in one way or another, and thus share only to a limited degree in the value—and perhaps the authority—possessed by democracy so understood.
II. EQUALITY AS FAIRNESS: TREATING YOU AND ME ALIKE
Formal equality—or equality as fairness—is often deemed the basis of justice. It has also at times been taken to provide an argument for the authority of democratic decisions: fairness requires that we not grant ourselves special privileges that we deny to other, similarly situated citizens. Democratic decisions are, at least ideally, made by all citizens equally: everybody has an opportunity to cast a vote based on her own
343
Democratic Equality and Political Authority
judgment of the proposal under consideration,10 and everybody's vote counts equally. So, one may think, we have a straightforward duty of fairness to obey democratic decisions: If I do not comply with the democratic decision, but instead act on my judgment of the matter at issue, I grant myself special status that I deny to others. By contrast, if I submit to the democratic decision (and set aside my own view of the matter under consideration), I treat equally, or show equal regard for, others and myself.11
Yet despite its intuitive appeal, this fairness argument fails; and its failure will prove instructive for thinking about what more must be added to our understanding of political equality if it is to be a basis for authority. The force of the fairness argument rests on the idea that, when I obey democratic decisions, I treat myself and others alike; by contrast, when I deny the authority of these decisions and act directly on what I judge to be the correct assessment of the matter, I grant myself special privileges and thus fail to live up to the demands of fairness. But once we consider the details of the argument more carefully, we recognize that no such privilege need be asserted when I disobey a democratic decision.12
Since proponents of the fairness argument spell out in different ways the somewhat inchoate intuition behind their position, I will consider two different versions of the argument to make this point.13 First, the fairness objection to disobeying democracy is often put by saying that, when I disobey, I act on my own judgment and thereby deny equal standing to the judgments of my fellow citizens. But this explanation is more confusing than illuminating. There is a straightforward sense in
10. I assume throughout that the democratic decisions under discussion aggregate voter judgments about justice or the common good (broadly conceived), and concern what Dworkin calls "choice-insensitive" issues: Ronald Dworkin, Sovereign Virtue: The Theory and Practice of Equality (Cambridge, Mass.: Harvard University Press, 2000), p. 204.
11. Versions of this thought appear to underpin the arguments advanced in Buchanan, "Political Legitimacy and Democracy"; John Rawls, "Legal Obligation and the Duty of Fair Play," in Collected Papers (Cambridge, Mass.: Harvard University Press, 2001); and perhaps Shapiro, "Authority." (Some passages in Shapiro's essay do, however, suggest that he may ultimately hold something closer to the view that I develop and defend in Section V.)
12. I might still be doing something wrong by disobeying. But to say that is simply to affirm the egalitarian authority claim. Yet what is at issue here is what the wrong consists in, and specifically whether it consists in granting myself special privileges.
13. At times the inchoate intuition comes to little more than an endorsement of the egalitarian authority claim. In those cases, my own positive argument in defense of the claim may evidently provide yet another way of spelling out the intuition.
344
Philosophy & Public Affairs
which I always act on my own judgment, whether or not I obey another's directives. In either case, I judge what reasons apply to me—including, if another has authority over me, the reasons provided by her directives— and act on that basis. Yet if I act on my own judgment whether or not I obey an authoritative directive, then it cannot be argued that I must obey another's directive in order to avoid acting on my own judgment.
This objection points toward a more general observation and a more careful restatement of the fairness argument: Useful though this way of speaking sometimes is, statements regarding authority are not statements about whose judgment the subject acts upon. Instead, they are statements about reasons the subject has for action—namely, that someone's directives are binding reasons for the subject. So the problem with which the fairness argument is concerned is not on whose judgment the subject acts, but rather to whose judgment the subject grants reason-giving status. In accepting a democratic decision as authoritative, I count all people's views equally in determining what I ought to do and thus treat them as reasons for action. By contrast, if I deny the authority of a democratic decision and act on my own judgment of the underlying reasons instead, then, according to the fairness argument, I unfairly grant my own view reason-giving status that I deny to the views of others. So if I am to treat myself and others equally, I must obey democratic decisions rather than act on my own judgment of the underlying reasons.
But this argument rests on a mistake. It is indeed true that, in denying the authority of the democratic decision, I deny that the views of others, as expressed in their votes, are a source of authoritative reasons for me. (I deny, that is, that the sheer fact that they judge—rightly or wrongly— that I ought to do x is a reason to do x.) But it is false that I take my own view or judgment as such a reason instead. I need not, and normally will not, do so. That I judge that justice requires x is not a reason for me to do x. My reason for doing x is that justice requires it. My judgment, rather than being a reason itself for doing x, seeks to properly capture the reasons there are for doing x and to guide my conformity with them. When my judgment is wrong, I have no reason to do x,14 even though I judge that I do. Since an authoritative judgment is a reason even if it is
14. To be precise: I have no reason over and above the reasons that I had anyway apart from my judgment, and that my judgment was meant to capture.
345
Democratic Equality and Political Authority
substantively mistaken, my judgment is not authoritative for me (nor need I take it to be so). In denying authoritative status to democratic decisions, I do not deny reason-giving status to the judgments of others that I grant my own judgment. It is then not the case that I treat myself with special regard when I deny the authority of a democratic decision (and instead act directly on what I judge to be the relevant reasons apart from the alleged authoritative directive) but show no such special regard for myself when I accept its authority.15
Notice that this response to the fairness argument is fully compatible with recognizing our own fallibility. If I have good reason to distrust my own assessment of certain reasons, then I ought to hedge my reliance on that assessment accordingly. Nor does it require treating myself as inherently more reliable than others. If I find myself disagreeing with you, and I have no reason to think you are any more likely to be wrong than I am, I ought to take this into account as well.16 Next, by denying the authority of democratic decisions as such, we do not claim that our own view is authoritative for others. (If I judge that justice requires doing x, and others ask me what they ought to do, I tell them to do x. But they ought to do x because that is what justice requires, not because I judge that it does.) Nor, finally, need we deny that we should obey if the authority is reliable or well placed to help us coordinate with others in the pursuit of a valuable goal. All that we have denied here is that the fairness argument's appeal to formal equality—treating ourselves and others alike—
15. There is another point worth emphasizing: It is often assumed that it would be unfair to act against democratic judgments because we cannot universalize the precept that I may act on what I judge to be just or unjust. This thought is, for instance, central to Rawls's early discussion of fair play and its applicability to democratic legislative institutions in Rawls, "Legal Obligation and the Duty of Fair Play." It is also suggested by Christiano, The Constitution of Equality, p. 253, who discusses whether citizens have a right to disobey "when they think the results are unjust" (emphasis added), and concludes that this is not a right we can grant all citizens without undermining collective decision making and action. But that I cannot generalize the precepts Rawls and Christiano are considering does not show that I must obey democratic decisions. It merely shows that the fact that I take a decision to be unjust is not enough to show that I have a right to disobey it. If, by contrast, the decision is in fact unjust, I might very well have such aright because one could universalize a precept of the sort "obey democratic decisions unless they are unjust; if they are unjust, conform to the demands of justice instead."
16. Much will depend on what counts as a reason for thinking you are more likely to be wrong. For useful discussion, see Adam Elga, "Reflection and Disagreement," Nous 41 (2007); and David Enoch, "Not Just a Truthometer: Taking Oneself Seriously (but Not Too Seriously) in Cases of Peer Disagreement," Mind 119 (2011).
346
Philosophy & Public Affairs
explains why the democratic pedigree of the law gives us reason to obey it. More specifically, we have shown that disobeying democratic laws as such need not involve granting ourselves special privileges or violating the demands of equality as fairness. To justify the egalitarian authority claim, we must thus look for another argument that grounds the authority of our laws in the egalitarian features of our legislative procedures.
III. EQUAL PUBLIC RESPECT FOR JUDGMENT
The fairness argument rests on a conceptual mistake about the nature of authority. But its failure points to a significant issue in accounting for authority on the basis of equality: since both obeying and disobeying democratic decisions are compatible with treating ourselves and others in the same way, we must find independent grounds for granting reason-giving status to democratic decisions in the first place. We need an account of the egalitarian relationship established or sustained by democracy that explains why that relationship requires treating democratic decisions as authoritative.
A view that promises to do just that has recently been suggested by Thomas Christiano.17 We have a weighty interest in relationships to others in which we are not simply treated alike, but are being publicly respected as equals; and such public equal respect is best (or perhaps only) shown by treating as authoritative the decisions of a procedure in which each citizen has an equal positive say. The authority of democratic procedures is based on the value of relating to others as equal rulers: our interest in seeing others obey decisions in the making of which we participate as equals—an interest derived from our concern for being publicly respected—justifies each citizen's duty to obey democratic decisions.
17. Christiano, "The Authority of Democracy"; Christiano, The Constitution of Equality: Democratic Authority and Its Limits. I say that the strategy I discuss is "suggested" by Christiano because his complex argument for democratic authority also appeals to epistemic considerations and the importance of "feeling at home" in society. I set aside these additional strands of Christiano's theory because I believe that, even if they succeed, they provide conceptually quite separate foundations for our duty to obey that are largely independent of the egalitarian character of democratic procedures. I am also rather doubtful that they do succeed, for the reasons discussed in Steven Wall, "Debate: Democracy, Authority and Publicity," Journal of Political Philosophy 14 (2006).
347
Democratic Equality and Political Authority
Let me fill in this sketch of an egalitarian argument for democratic authority in some more detail before I consider it critically. We have, Christiano proposes, a "pre-eminent" interest in "being publicly treated as an equal member of a society,"18 or in publicly "being treated as a person with equal moral standing among [our] fellow citizens."19 One may think that this interest would be best met by making laws that are substantively just, which for Christiano is a matter of advancing people's interests equally.20 But, Christiano argues, this is not so. We are fallible in our assessment of our own interests, the interests of others, and how they stack up; and there is widespread disagreement about how significant different interests are. Fallibility and disagreement make whatever respect may be involved in equally advancing people's interests problematically nonpublic: my good-faith (indeed, my correct) judgment of what equal advancement of your interests requires may look like blatant disrespect to you. So another way of affirming equality is needed to satisfy the preeminent interest in public equal respect.
One such way is suggested by the observation that human life and well-being depend to a large extent on engaging with values. Such engagement is a matter of judging what values there are and what reasons they provide for us, and then acting on these reasons. As we cannot show public respect by what may seem the most direct route— treating people justly by advancing their interests equally—we must take a more indirect route that better satisfies the demand for publicity: we must act in ways that publicly show that we value another's judgment. If being publicly shown equal respect is a matter of being publicly respected for one's judgment, and if public respect for another's judgment is best shown by treating her vote (her publicly expressed judgment of how we ought to live together as a community) as authoritative, then a duty to advance everyone's interest in public equal respect may ground a duty to obey democratic decisions, decisions in the making of which everybody's judgment is given equal weight.21 The argument is, we may say, "quasi-transcendental": it appeals to the conditions of the possibility of relationships of public equal respect under circumstances of
18. Christiano, The Constitution of Equality, p. 251.
19. Ibid., p. 63.
20. Ibid., chap. 2.
21. Ibid., chap. 3.
348
Philosophy & Public Affairs
disagreement and fallibility. And because only obedience, and not the direct pursuit of justice, meets Christiano's publicity condition, his account, unlike the view considered in the previous section, explains why the requisite equality is incompatible with disobeying democratic decisions and pursuing justice directly.
Though ingenious, this argument is ultimately unsuccessful. While a number of questions could be raised about the significance of citizens' interest in public respect and whether individual citizens indeed have a duty to advance that interest, here I focus on just one claim central to Christiano's argument: that, if a citizen has reason to advance others' interest in public equal respect, she also has reason to obey democratic decisions. To show respect for others as equals (and, in particular, as equal moral agents capable of judgment), we must, Christiano proposes, treat the fact that they support, and cast a vote for, doing x as a content-independent reason for doing x. But this is not correct. In fact, public equal respect cannot normally be advanced by granting the status of content-independent reason to people's votes, because granting them this status is not an appropriate way of responding to the value of others' judgments. Yet responding appropriately to an object's value is what respect for that object consists in.
Being capable of engaging with reasons—having the capacity for judgment—may itself be valuable because it opens up to us a realm of value and action that would otherwise be closed. It may be so even where the actual exercises of that capacity misfire: the sheer possibility of proper engagement may be sufficient to impose on us a duty to respect others as possessors of the capacity for judgment. Yet it is difficult to see why respect of this sort would involve treating another's actual exercises of that capacity (and thus her votes) as content-independent reasons for action. The appropriate response to the value of judgment just described would rather be to protect another's capacity for judgment from being stifled, and to enable its free exercise and development. It may require upholding an extensive sphere of freedom of speech, so that agents can hone and improve their judgments through exchanges with others, and an extensive sphere of personal freedom, so that they can form and act on their own judgments in their personal affairs. We may even have reason to democratize existing political institutions and to grant everybody a right to an equal say, so that people can develop and act on their capacity for judgment on matters of public as well as private concern.
349
Democratic Equality and Political Authority
But protecting people's capacity, and enabling them to develop and exercise it, does not require—indeed, it does not normally provide us with reason for—treating their judgment as a content-independent reason for action.22 Yet it is this response to others' judgments that Christiano's argument demands.
It may be objected that the argument just considered adopts too impoverished a view of the value of judgment. Agents exercise their judgment in response to reasons; and they aspire to judge correctly what actions these reasons require or make appropriate. From the point of view of the agent, the value for which she wishes her judgments to be respected is thus, at least in part, that of correctness. If we democratize our institutions, and let people form and express judgments on what justice and other political values require, then respecting them as they wish to be respected is a matter of responding to them as persons whose judgment is correct. And if another's judgment is indeed correct, then following or deferring to it may be an appropriate response.23 So respect for others' judgment as correct may indeed justify treating their votes as content-independent reasons for action.
But it is difficult to see how the public respect we owe to our fellow citizens could be respect for their judgments as correct. Democracy's claim to authority comes into its own when we obey laws we deem substantively mistaken, against which we canvassed and cast our ballot. But then we cannot genuinely value the judgments of the majority supporting these laws as correct. Nor can the majority genuinely take our obedience to be a public expression of respect for the correctness of their judgments if they know that we believe the decision to be wrong. (Notice too that, if public respect required treating the decision as correct, we would have reason not only to treat the decision as binding but also to not try to change it in future elections, not campaign against it, and so on at least as long as there is no relevant change in circumstances or in available information. Yet this would seem incompatible with central tenets of democracy as we know it.)
22. Except insofar as the institution does have authority on independent grounds, which the democratic voters help administer. But in such cases, equal respect does not provide the justification of authority we are concerned with.
23. The link between correctness and deference is in fact more complicated. I leave these complications aside.
350
Philosophy & Public Affairs
Someone may complain, finally, that neither an argument from the sheer capacity for engaging with reasons nor an appeal to correctness does justice to the role judgment plays in our life. Personal autonomy requires that agents give shape to their own life—exercise authorship over it—by forming judgments and acting on them. The value of autonomy thus explains why judgment is valuable, and worthy of respect, even if it is not correct. In a familiar liberal inflection, this thought justifies an extensive sphere of private freedom. But perhaps a democratic version of this argument is also available: Respect for another as an autonomous agent is expressed by letting her judgment play a role in determining not just how she lives, but also how we all live together. And this requires treating her judgment (as well as the judgments of thousands or millions of others to whom equal respect is also owed) as a source of content-independent reasons for action for all the other members of the community.
The force of this final proposal depends crucially on connecting judgment to autonomy. Yet once we spell out the connection in detail, it becomes clear that the purported link between respect for judgment and obeying democratic decisions breaks down. Let me grant that we each have an autonomy-based interest in determining, or playing a role in determining, how we live together in our community. But we must not overlook that each of us also has an autonomy-based interest in determining for herself how to live her own life. And even if the former interest can be advanced by requiring others to obey democratic judgments that they would otherwise lack reason to obey, such a requirement will simultaneously set back the latter interest. After all, once a democratic decision has authority over me, my ability to give shape to my life in light of my own judgment is limited.24 Furthermore, this loss in autonomy almost certainly far outweighs the purported gain. When the authority of democratic decisions is extended, I gain a very small share of control over the lives of very many other people, and specifically over that aspect of their lives that our collective decision now regulates. Yet I also lose a large share of control over the corresponding aspect of my own life.
24. This is not to deny what I affirmed in Section II, that even when we obey an authoritative directive we act on our own judgment. Even a moral marionette, someone whose every action is determined by another's directive, is acting on his own judgment. Still, we would normally say that he lacks authorship over his own life, because his life is not shaped by his views of what projects and purposes matter. ("Normally" only because one could imagine someone who autonomously endorses a monkish life of self-abnegation, of which obedience is an essential part.)
351
Democratic Equality and Political Authority
Autonomy is a matter of self-rule, of giving shape to one's own life in accordance with one's own judgment. Since the control I gain is mostly control over others rather than myself, it barely advances my autonomy. By contrast, since the control I lose is largely control over myself, my autonomy is actively set back. But then the appropriate response to the value that judgment has for autonomous agents can hardly be to treat people's expressions of these judgments as sources of authoritative directives for others, who are thereby prevented from giving shape to their own life in accordance with their own judgments.
There is, then, no room for publicly showing equal respect for another's judgment by treating it as an equal source of content-independent reasons for action. Treating it in this way is not an adequate response to any value a citizen could reasonably think others, who disagree with her, take her judgment to have. Consequently, we cannot justify treating democratic decisions as authoritative by appeal to the public respect we owe to our fellow citizens' judgments and votes.
IV. RELATIONAL EQUALITY
In light of these difficulties, I propose an alternative route from egalitarian relationships to democratic authority. Rather than focus on what considerations we must treat as reasons if we are to uphold a valuable egalitarian relationship, I suggest we begin by asking what considerations we must not treat as reasons. If the idea that a valuable relationship might require not treating certain considerations as reasons seems surprising, it is worth reminding oneself that this is in fact a relatively common phenomenon. Being a friend, spouse, or fellow citizen involves (in the words of T. M. Scanlon) "seeing reason to exclude some considerations from the realm of relevant reasons (under certain conditions) just as it involves reasons for including others."25 For instance, if my spouse has an accident and is hospitalized, I not only have reason to go see her at once. I must also go see her without first considering whether I should not run some useful work-related errands instead. Similarly, I argue, there are certain considerations that we must exclude from among the reasons on which we act if we are to relate to one another as equals. This
25. Scanlon, What We Owe to Each Other, p. 157 (emphasis added). Scanlon talks here of what it means to be moral. I assume that the same thought applies even more strongly to valuable relationships of the sort I discuss here.
352
Philosophy & Public Affairs
leads to a second thought: to justify democratic authority, we must ask not what considerations we treat as reasons if we obey democratic decisions, but what considerations we avoid acting on if we do.
The argument for the egalitarian authority claim I advance proceeds in two parts. First, it is an important feature of certain valuable relationships—relationships committed to an ideal I call "relational equality"—that they require us to set aside, and not act on, unequal power advantages in shaping our interactions and the norms and expectations governing them. (To establish this is the purpose of this section.) Second, a concern with excluding such considerations from our interaction provides the basis for justifying the practical authority of certain persons, bodies, or procedures: unless we accept as binding the directives of an authority whose decisions have not been shaped by our unequal power advantages, it is often impossible for us to achieve certain valuable ends (in particular, ends that require coordination among people who disagree about which coordination solution is best) without violating the demands the egalitarian ideal makes on us. (I argue for this in Section V.) To borrow an image from Rousseau: on the view defended here, egalitarian political procedures like democracy have special authority because they answer our need to replace "whatever physical inequality nature may have placed between men" with "a moral and legitimate equality," so that men, "while they may be unequal in force or in genius, ... all become equal by convention and by right."26
A. Relating to Others as Equals
It is a widely accepted ideal in our culture that many of our intrinsically valuable relationships—such as those between friends, siblings, spouses, neighbors, and even fellow citizens—should be relationships among equals.27 My purpose here is to sketch certain conditions that a
26. Jean-Jacques Rousseau, "The Social Contract," in The Social Contract, and Other Later Political Writings, ed. Victor Gourevitch (Cambridge: Cambridge University Press, 1997). (p- 56 (bk. 1, chap. 9, §8).
27. Cf. David Miller's discussion of what he calls "social equality" and describes as "the ideal of a society in which people regard and treat one another as equals, in other words a society that is not marked by status divisions such that one can place different people in hierarchically ranked categories, in different classes for instance." David Miller, "Equality and Justice," Ratio 10 (1997): 224. For discussions of a related ideal that focus primarily on questions of economic equality, see Elizabeth S. Anderson, "What Is the Point of Equality?" Ethics 109 (1999); and Samuel Scheffler, "What Is Egalitarianism?" Philosophy & Public
353
Democratic Equality and Political Authority
relationship must meet if it is to live up to this egalitarian ideal. The sketch is necessarily selective. It starts from certain paradigmatic instances of egalitarian relationships (centrally, friendship and marriage); and it only briefly suggests why the egalitarian ideal should also have a grip on us in our relationship as fellow citizens. It does not argue for the ideal of relational equality, but instead assumes that the reader recognizes its appeal, and seeks to articulate more clearly than we commonly do what the ideal requires. And in doing so, it focuses on those features of egalitarian relationships that will ultimately help us see the connection between such relationships and a duty to obey egalitarian procedures. Much, then, will be left out. But I hope enough will be said to make intelligible the central idea: that relating to others as equals will often require excluding from our relationship certain considerations—in particular, unequal power advantages—that would threaten our equal control over our common life.
A starting thought is that relational equality requires the equal advancement of one another's well-being or interests. To relate to one another in accordance with the ideal of equality is to see one another's interests as being equally significant when it comes to determining how to act within the relationship. Call this the requirement of equal concern: if I am to relate to you as an equal, then I must take your interests to be of as much significance as mine when it comes to those matters bearing on our relationship.
But despite the significance of equal concern, it does not suffice to capture the ideal of relational equality. To see this, we need only notice that relationships between parents and children might be such that each has equal concern for her own and the other's well-being, and yet the relationship is not one among equals.28 The parent makes decisions that the child must follow; the parent has great power over the child while the child has little power over the parent; and so on. So even if (as I accept)
Affairs 31 (2003). Not all intrinsically valuable relationships need be among equals. Relationships among parents and their dependent children, for instance, are not normally held to this ideal. I will say more about such cases later on in this section.
28. Relationships between masters and servants, at least as they existed (and were understood) historically, may provide another counterexample: The master and the servant may have equal concern for each other. Their relationship may even be seen by the parties as having noninstrumental value precisely because they have such equal concern. And yet they need not think of it as a relationship among equals.
354
Philosophy & Public Affairs
equal concern is necessary for the existence of intrinsically valuable egalitarian relationships, it is not sufficient.
What more is required? It might be thought that the relationship must also be structured so as to give equal rights and duties to each party. Among friends, for instance, we would expect that my duty to help you when you are in need is mirrored by a similar duty on your part when I need a hand. And if I have a prerogative to pursue my child's interests even at my friend's expense, then she must in turn be permitted to pursue her child's interests at mine. By contrast, in the relationship between parent and dependent child, no such expectation of equal rights and duties applies. Call this then the requirement of equal rights: if we are to relate to each other as equals, we must also have, and see each other as having, equal rights in our relationship.
But even a relationship that satisfies the requirements of equal concern and equal rights need not be one among equals. For imagine a relationship in which each person has the same rights: a right not to be physically injured, a right that promises be kept, and perhaps even positive rights to the provision of certain basic benefits. But within the constraints imposed by the rights just mentioned, each is at liberty to use whatever power she has at her disposal to advance both her own and her partner's well-being. This is a relationship of equal rights. It may also be one of equal concern. Yet it is not one between equals if it is also the case that one of the parties has significantly greater power than the other and can thus shape the relationship as she sees fit. Just consider a marriage in which this holds true: Within the general restrictions that apply to all human interactions, A and her spouse B consider themselves at liberty to threaten, bribe, or cheat as long as they do so for the sake of advancing their common good (suitably specified). Yet A has much greater resources than B to make offers or threats. Hence A will end up making most of the relevant decisions about their common life. Such a relationship, in which one party has the power to make most decisions, is, it seems to me, not one among equals, yet the requirements of equal concern and equal rights may both be met.
B. The Requirement ofNonsubjection
This last thought demands clarification. Imagine a marriage in which one person makes most of or even all the important decisions: a husband
355
Democratic Equality and Political Authority
who decides where the children go to school, where the family goes on vacation, what will be served for dinner, and so on. This strongly suggests that there is something problematic about the relationship. Perhaps the problem is only apparent. Perhaps the wife simply allows the husband to make these decisions because she does not care, though if she cared, she could exercise equal control over these issues. But if she could not exercise equal control even where it mattered to her, then we must conclude that the relationship is problematically unequal, even hierarchical.
Imagine next a relationship in which one person has disproportionate power to determine not only what the parties do together, but also what they can expect of one another. To the extent that these expectations are partly constitutive of what their relationship actually is, that person also has the power to determine the shape of the relationship: the norms and rules governing it. (Is ours the kind of relationship where one person can simply turn up at another's door and expect to be let in? Where one person can touch another without first getting explicit consent? Where one person can borrow another's property without asking?) This poses an even more pervasive threat to the ideal of relational equality.
These examples suggest that an egalitarian relationship requires, among other things, rough equality of power over the interactions that make up the relationship.29 Relational equality is, in turn, threatened where the parties to the relationship have significantly different power over how they interact with and relate to one another. We might call this the threat of subjection, and refer to equal power as a matter of nonsubjection.30 Where the parties to a relationship lack equal power, the relationship lacks the distinctive value associated with relating to one another as equals.
But this is not all. Just as equal concern requires not only that the relationship in fact advances the parties' interests equally, but also that the parties see each others' interests as equally significant in the
29. In principle, power can be equal even if, in fact, only one person ever makes decisions. Conversely, it can be unequal even if both make decisions, as long as the weaker party's decisions are already conditioned by the stronger party's preferences. I do not dwell on these complications here.
30. I borrow these terms from John Stuart Mill. See Mill, "The Subjection of Women," in On Liberty, and Other Essays, ed. John Gray (Oxford: Oxford University Press, 1991), pp. 471-582; as well as Mill, "Considerations on Representative Government," in On Liberty, and Other Essays, pp. 205-467.
356
Philosophy & Public Affairs
relationship, so nonsubjection requires not only that the parties in fact have roughly equal power, but also that the parties be committed to having equal power. Normally, having such a commitment is a matter of being guided, in our deliberation and actions, by norms that sustain and protect equal power among us—norms that bar us from relying in our dealings with one another on advantages that are in fact distributed unequally among us, or that we know are quite likely to be so distributed.31
Power differentials can in principle be purely internal to the relationship, and need not reflect any differences in power outside of it: I might submit to your decisions simply because I promised I would, not because you have greater economic, social, or physical power. You might not even have the capacity to enforce my promissory obligation. Still, as long as I consider myself bound by the promise, our power in the relationship differs. Yet at least in relationships that are committed to egalitarian ideals, such internal inequalities are relatively rare (and indeed suggest incompatible or even incoherent commitments on the part of at least one of the parties). The more common threat to nonsubjection arises from power differentials outside of the relationship that can be translated into unequal control within it: the spouse who can use his greater economic power to determine who takes time off from work after a child is born; the friend who can use her greater social power—her willingness to abandon this friendship because she has plenty of other friends—to force another to accommodate her preferences about what to do together; and so on. It is on these that I will focus here.32
31. Commitment may shape deliberation even where it does not shape action, for instance, because the intention we form is conditional, and the condition is never met. But even where deliberation does not affect action, it is relevant for evaluating the relationship in which we stand to others. Imagine, for instance, that you are stronger than me, and that you are committed to nonsubjection in our relationship, but I am not. So we make decisions in accordance with egalitarian decision procedures. But the only reason why I go along with them is because I know that you are ultimately strong enough to enforce them. And I intend to do away with the procedure, and impose my own judgment of what would be best, as soon as I have sufficient power to do so. Even if I never acquire enough power, the fact that I intend to act in this way sets back the value of our relationship.
32. That the existence of external power differentials (and in particular persistent inequality in income, wealth, and economic security among men and women) can affect, and even undermine, personal relationships is insightfully argued in Susan Moller Okin, Justice, Gender, and the Family (New York: Basic Books, 1989).
357
Democratic Equality and Political Authority
The concern with equal control over the relationship is, I believe, a distinctive element of what it means to stand in an egalitarian relationship to one another. It must not be mistaken for two other ideals with which we are perhaps more familiar. First, equal control is distinct from the requirement of equal freedom that is often considered a demand of justice.33 Equal freedom requires comparisons across persons simplic-iter. By contrast, equal control is concerned with comparisons across members of a particular relationship within that relationship. Thus, the two ideals suggest different assessments of cases where the same two people are (unwittingly) part of two different relationships that are each unequal but in which different persons have greater control. Assume A employs B, and the laws in their society ensure that B has much less control over this relationship than A. Simultaneously, A and B are active participants on an anonymous messaging board, where, ignorant of each other's real identity, they strike up a friendship. B has much greater control over that friendship, for instance, because B, but not A, has plenty of other friends. Across the two relationships, A and B exercise equal power over each, and have equal freedom to shape their circumstances. But while the second unequal relationship may help alleviate the concern of equal freedom, it merely multiplies the problem of equal control: since that ideal is internal to each relationship, A and B now stand in two relationships that are each unattractively inegalitarian.
Second, equal control is distinct from the ideal of self-rule, either individual or collective, that is often deemed central to liberal-democratic political morality. Individual autonomy is threatened where I allow others to make decisions over my life. It is threatened, in particular, even where they in turn let me make similar decisions over theirs. When it comes to individual autonomy, losing control over my own life cannot normally be made up for by gaining control over another's. Yet such an exchange is perfectly compatible with the requirement of equal control that is central to egalitarian relationships. Collective autonomy requires that the parties to the relationship shape the norms and rules governing their interactions. It is threatened where these decisions are
33. See, for example, Anna Stilz, Liberal Loyalty: Freedom, Obligation, and the State (Princeton, N.J.: Princeton University Press, 2009). I thus agree with David Miller that the ideal of equality makes demands on us that are separate from, and sometimes even in tension with, those of justice: Miller, "Equality and Justice."
358
Philosophy & Public Affairs
instead made by someone external to the relationship. But submission to an impartial third party, acting as an authoritative arbitrator settling the parties' disputes, may be one way of achieving the equal control that egalitarian relationships demand. And since the arbitrator is not herself party to the relationship that gives rise to the dispute, any inequality in power between her and the disputants does not affect the egalitarian character of that relationship.
Emphasizing this distinction between autonomy (collective or individual) and nonsubjection serves first and foremost to sharpen our understanding of the bases of democratic authority. My suggestion is that it is the egalitarian character of democratic procedures, rather than the fact that they involve self-rule, that does the crucial justificatory work in establishing our duty to obey their outcomes. This is evidently compatible with recognizing that self-rule too has value in our relationships. That it has value may explain why, where we have a choice among various egalitarian procedures, we have a preference for democratic decision making over submission to the authority of an impartial third party. (I return to this point in the conclusion.)
Someone may insist that collective autonomy in particular is not just another value that can be traded off against nonsubjection, but rather a precondition of our having an intrinsically valuable relationship—and thus a relationship in which nonsubjection matters—at all. If that were so, then the argument for the authority of egalitarian procedures I offer here would be restricted to procedures that are also compatible with collective autonomy. (It would not, for instance, apply to the authoritative decisions of an impartial third party acting as an arbitrator to our dispute.) Two observations are worth making in response. First, even if this were true, the important conceptual point stands: nonsubjection, and not self-rule, does the central justificatory work in the account of democratic authority offered here. Second, it is in fact unlikely to be true. Friendships among children are often greatly constrained by parents or teachers. The children thus lack autonomous control over their relationship. And yet we would be reluctant to say that they have no relationship, that their relationship has no value, or that it does not matter if the little bit of control they have over their relationship is distributed unequally among them. Nor is this thought limited to children. A law that bars prenuptial agreements may restrict a couple's capacity to shape their own relationship, and thus also their collective autonomy. Yet we ought
359
Democratic Equality and Political Authority
to support it nonetheless if the law were necessary to protect one party from asymmetrical dependence on the other. Such constraints are thus incompatible neither with having a relationship nor with the relationship's being intrinsically valuable.
C. Excluding Unequal Power
One way of achieving nonsubjection is to enter into relationships only with those who are our equals across all domains of life (at least in the aggregate). But this is unattractive for several reasons. Not only would it significantly limit with whom we could be friends. It would also require us to give up friendships when one or the other party's circumstances changed—a common enough occurrence in modern societies with a significant degree of social mobility.
But limiting our relationships in this way is (luckily) not the only way of avoiding subjection. Even though our egalitarian relationships cannot be hierarchical, this does not mean that, when the parties are ranked along some dimension, they must come out on a par. You and I can relate to one another as equals and yet both think that you are smarter than I am. It need not even be the case that there is some aggregation of judgments of merit that makes us come out even. You and I can agree that you are smarter, stronger, better-looking, physically more fit, and richer than I am; and yet we can relate to one another as equals—as long as these differences do not translate into unequal control over the relationship.
They may not translate into unequal control because external restrictions bar the use of the asymmetrically distributed power within the relationship. Thus, laws against marital violence serve, among many other important ends, to prevent men from exploiting their greater physical strength in shaping their relationship to their female partner. And laws about the distribution of marital property may play an important role in reducing an economically weaker spouse's dependence on a wealthier, or better-paid, partner. But just as, if not more, important than such external restrictions are norms that the parties have internalized, and that shape not just their actions, but their very reasoning and deliberation about how to relate to one another. Thus, partners committed to relating to one another as equals will exclude from deliberation facts of unequal physical power, or unequal economic prospects, in dealing with one another. It is not just that, having considered whether
360
Philosophy & Public Affairs
my partner is physically weaker or otherwise incapable of punishing me for certain transgressions, I conclude that the risk of punishment remains too high, and thus stick to whatever rules we have agreed on; or that, having considered how much we each earn, I decide that mentioning our relative economic standing will not be helpful in advancing my cause when we disagree about how to split the housework. Rather, I never even consider whether to make use of my greater physical strength or economic success when deliberating about how to renegotiate the details of our relationship.
Often there is no good reason why particular differences would translate into unequal control in the first place. (Why should the fact that you earn more money than I do give you greater power over how we lead our life together?) Still, such differences may translate into greater power. (That you earn more money means that you would do just fine if you left me, whereas I would struggle financially if I left you.) And that power can be used as a means for advancing valuable ends. (That you have this power over our relationship allows you to insist that we divide the housework, and other responsibilities, equally among us, and not in the unequal—and unjust—way I would prefer.) At other times, the differences may even provide at least prima facie reasons for granting us unequal control: If my spouse and I are committed to leading a Christian marriage, or raising a Christian family, and my spouse happens to be a church deacon who knows much more about the demands of a Christian life than I do, then we would in some respects be better off if she decided the most important matters that arise in our relationship. But even though there are benefits to her having such authority—we come closer to our marital ideal—it also comes at a significant cost: unless there are other, roughly equally important domains in which I can decide and she cannot, my spouse's greater authority threatens the egalitarian character of the marriage.
If I happen to have greater power, and I do think that justice requires a particular scheme that you are reluctant to participate in, then I have some reason to use my power to impose that scheme. And if we could lead a more Christian marriage if we let my spouse make all of the important decisions, then we also have some reason to do that. But the concern for upholding equality among us may require me to forgo the use of the means I have to achieve the establishment of the more just scheme; and it may also require us to forgo the benefits that we could
361
Democratic Equality and Political Authority
achieve if my spouse were to rule unequally over our marriage. We have reason to share our decision making—by taking turns, assigning decision making about different domains of our common life to each partner, and so on—even if this means that we may make less good decisions, or end up settling on a less just scheme, than we could have.34 The value of relational equality thus makes intelligible why we often set aside and do not act on what are otherwise perfectly good reasons for action: the fact, for instance, that one of us generally knows better, and thus ought to be obeyed; or, alternatively, the fact that I have sufficient power to impose a better, more just scheme of rights and rules among us than would otherwise be established. In each of these cases, our concern with relating to one another as equals bars us from using certain means for achieving our ends even where these ends—which may include justice or the common good—are of great value. (And where the value of equal control is outweighed by competing concerns for the quality of our decisions, there the commitment to relational equality makes sense of our regret that these means were necessary to establish the better scheme.) It is the significance of barring considerations of unequal power from affecting our relationship, and undermining the equal control that we have over it, that (I argue in the next section) ultimately justifies the special authority of egalitarian political procedures. But before we can turn to that argument, we must take one last step and explain, or at least make plausible, why the egalitarian ideal also applies to the relationship in which we stand to our fellow citizens.
D. From Personal to Political Relationships
Someone might accept this sketch of the value of equality as it applies to the personal relationships I have considered, and yet resist the further claim that this value also applies to our political relationships among fellow citizens. She need not say that the value of relational equality is irrelevant for politics. But it would matter only insofar as politics could instrumentally sustain that value in the private relationships in which it is properly at home. If that requires distributing control over lawmaking
34. Sometimes we may be unable or unwilling to do this, most obviously if the better outcomes are of exceeding significance. There may thus be limits to the friendship I can hope to sustain with a moral sage.
362
Philosophy & Public Affairs
unequally—say, because unequal distributions of political power lead to laws that better protect equal relationships in private life—then so be it.
This position is certainly not incoherent. But it strikes me as implausible nonetheless. Part of this is simply an interpretive matter with regard to our social ethos: while a commitment to relational equality may have its origins in close interpersonal relationships, I think there can be little doubt that it is now widely understood in our culture to also apply to the large-scale relationships among fellow citizens.35 But part of it is also just the practical difficulty associated with cabining the ideal in the way the position suggests.
The ideal of nonsubjection that we find in many private relationships can be brought to bear on the relationship among fellow citizens in two different but mutually reinforcing ways. First, the ideal itself may have traveled from the private realm to the political realm: the ideals that govern our friendships are also meant to govern our relationships to our neighbors and ultimately to our fellow citizens. Something like this is suggested by George Kateb when he writes that
some of the moral energy behind the push for representative democracy has come from people . . . who sought to hold political authority to standards that to some degree had their original home in nonpo-litical life, private and social. The modern birth of representative democracy, in England, North America, and France, was itself facilitated by the urgencies of the private or domestic or neighborly voice, or the voice of friendship or brotherhood or religious devotion. There was a passion to repudiate the claimed immunity of the political sphere from the exacting requirements of the best morality of everyday life.36
35. We might not even appreciate quite how deeply the ideal of relational equality is engrained in our culture until we remember what life was like when society denied it. For a reminder, it is worth reading Gordon Wood's account of the fundamentally hierarchical character of social relations in the English-speaking world prior to the American Revolution, and how it compares to the egalitarian ethos that has arisen (and prevailed) since: Gordon S. Wood, The Radicalism of the American Revolution (New York: A. A. Knopf, 1992), pp. 24-42, chap. 1.
36. George Kateb, "The Moral Distinctiveness of Representative Democracy," Ethics 91 (1981): 365-
363
Democratic Equality and Political Authority
But this is not the only way in which the private ideal of nonsubjection can make its demands even in the political sphere. We might also think that the political sphere is subject to these demands not because political relationships themselves are committed to relational equality, but because unequal control in politics would by itself undermine private relationships that are committed to this ideal. This may happen where two conditions are met. First, politics regulates many if not all private relationships in one way or another. Second, it cannot be assumed that (and indeed would be considered a shame if) those who make the law will not have private relationships with those who do not and are nonetheless ruled by it. But these private relationships would be threatened if some of us had greater power over the rules governing them than others do.37 It could thus also be for the sake of protecting and upholding the ideal of equal control in the private relationships in which we anyway stand (or could stand) to each other that we require protection from unequal control over political procedures.38
None of this is to say that there could not be trade-offs between equality among fellow citizens and equality within particular relationships. If we democratically make laws that foster the dependency of wives on their husbands, how should we think about the competing claims of
37. Something like this thought is central to Mill's argument for female suffrage. See Mill, "The Subjection of Women," in On Liberty, and Other Essays, pp. 471-582. One might reasonably ask how this is compatible with the special power that our elected representatives have over the laws governing us. I cannot consider this issue here. For an account of democratic representation that goes some way toward answering this concern, see Eric Beerbohm, In Our Name: The Ethics of Democracy (Princeton, N.J.: Princeton University Press, 2012).
38. It might be suggested that the egalitarian ideal does not extend from domestic to political relationships because the latter, but not the former, are largely nonvoluntary. One reason for resisting this objection is that not all domestic relationships to which the egalitarian ideal applies are voluntary: I should think that siblings ought to relate to each other as equals, for instance, as do neighbors who have each inherited their homes and lack the resources to move away. And though arranged marriages may be distinctly problematic because they lack voluntariness, an arranged marriage in which the spouses relate to one another as equals is greatly superior to one in which they do not. So the commitment to relational equality does not track the voluntary/nonvoluntary distinction in our thinking about relationships. Second, even if it were true that the ideal of relational equality applied strictly speaking only to voluntary relationships, this would not show that political relationships are not properly shaped by this ideal: as the second argument for extending egalitarian demands to politics shows, the ideal may make demands even on nonvoluntary relationships if these in turn significantly affect (the possibility of) valuable voluntary relationships.
364
Philosophy & Public Affairs
relational equality that arise here? I will return to this question in Section VI, when I consider the limits to the authority of democratic decision making on this egalitarian account. But first, I must complete the justification of such authority.
V THE AUTHORITY OF EGALITARIAN PROCEDURES
In the previous section, we discussed what it means to relate to one another as equals, and found among its requirements that the parties have equal control over their relationship and be committed to norms that bar them from relying on considerations of unequal power in their dealings with one another. In this section, I argue that relational equality justifies the authority of an egalitarian decision procedure like democracy. The commitment to nonsubjection, together with the need to coordinate our actions and expectations and the fact that we disagree about the best way of doing so, poses a problem to which the authority of egalitarian procedures—procedures over which neither party has greater control—provides the solution.
A. Disagreement and Coordination
To understand the authority of egalitarian procedures, we must begin by recognizing two central features of our political life: the fact of disagreement, and the need for coordination.39 Members of our community regularly disagree about which rights and rules ought to govern our relationships. They hold different views of justice and the common good, leading, for instance, to different views of which scheme of property rights or taxation would be best or most just. Yet for both moral and practical purposes, we must settle on a common set of rights and rules to structure our interactions. Without settled common rules governing the acquisition, possession, and transfer of objects, competing claims to property rights are likely to lead to conflict and (if agents believe they have a right to coercively enforce their rights) perhaps violence; and unless property rights are widely respected, they cannot secure interests and choices in the way they are meant to do if they are to advance autonomy. Without widely recognized rules assigning tax burdens, we
39. For a more detailed discussion, see Waldron, Law and Disagreement, p. 102.
365
Democratic Equality and Political Authority
cannot establish a fair scheme of contributions to the pursuit of collective endeavors.40 If each of us simply acts on her own best understanding of justice and the rights and rules it requires, we will often end up worse off than we would have been had we acted in some coordinated fashion—even if the coordination solution to which we are jointly contributing is in principle less just than the alternative either of us prefers.41
Our concern here is ultimately with the role that the egalitarian character of political procedures plays in justifying our duty to obey their outcomes. But before we can even begin talking about procedural virtues, we must consider a worry that arises from the invocation of disagreement and coordination. Together, one may think, these two are fully sufficient to justify authority, making otiose appeals to democracy (or any other egalitarian procedure) in the justification of a duty to obey (though perhaps not in discussions of how authoritative institutions ought to be organized) ,42
The argument for authority based on coordination and disagreement alone takes something like the following form: If we have (and know we have) reason to coordinate our activities in the pursuit of valuable ends, but hold divergent views of what coordinated action will best advance these ends (or even of what these ends precisely are), then we cannot conform to this reason by each simply acting on our respective view of the best coordination solution. Given that our views on this matter diverge, our actions would almost inevitably fail to fit together in the way
40. In addition, many rights stand in systematic relations to one another: a certain view of property acquisition may go hand in hand with a certain view of transfer, inheritance, and taxation; and even if you and I disagree about the best set of rules, we can agree that converging on one such set is better than exposing others to the competing pressures of, say, my conception of rights in acquisition but your conception of rights in inheritance and taxation.
41. This is obviously only true if that shared view falls within a certain range of plausible conceptions and is not radically unjust. For the remainder of the discussion, I will not generally add this caveat but rather assume that the condition is met. Disputes so understood are, as Waldron has pointed out, a particular instance of what game theorists call "partial conflict games": each party prefers coordination to noncoordination; yet they disagree about which among the possible coordination solutions to pursue. Waldron, Law and Disagreement, pp. i03ff.
42. This possibility is adumbrated by Jeremy Waldron when he writes that "an account of majority decision cannot itself explain the authority of legislation. . . . That explanation must instead come primarily from our sense of the moral urgency and importance of the problems that it is necessary for us to address—the things that (morally) need to be done and must be done by us, in our millions, together, if they are to be done at all." Ibid., p. 117.
366
Philosophy & Public Affairs
required for successful coordination. By contrast, if we each obey the orders of a common authority, and that authority directs us toward a particular coordination solution, then coordination will likely be achieved. So (this argument concludes) we have instrumental reasons to obey a common authority rather than act directly on our own assessment of the best coordination solution; and that reason does not depend on any egalitarian features of the authoritative decision procedure.
I accept that this argument succeeds in justifying authority for some people who have reason to coordinate. But I think the argument leaves significant gaps in the authority of our laws, including laws concerned with valuable coordination. Even if each member of our group has reason to coordinate her actions with those of others, it does not follow that each member of the group has coordination-based reasons to obey the directives of a common authority. If those members who lack such coordination-based reasons nonetheless have a duty to obey, that duty is based not on the need for coordination alone, but on egalitarian considerations of the sort discussed in the previous section.
The coordination argument assumes that, once a de facto authority lays down rules for coordinating our activities, each of us will best contribute to the coordination outcome by following these rules even if they are suboptimal. But there are at least two kinds of cases where some of those to whom the rules apply may nonetheless help bring about a better coordination outcome by disobeying rather than obeying. First, at least sometimes we converge on a common course of action, not because we each agree that it is the best available to us, nor because we each have submitted to a common authority, but because one of us has the power to coerce others to go along with what these others deem a suboptimal solution to their common problem. So if a purported authority directs us toward a particular coordination outcome, yet I (a) in fact know that another coordination solution is superior and (b) have the means to coerce others to converge on that superior solution, then as long as our concern is solely with achieving the best coordination outcome, I have no good reason to obey the directives of the purported authority rather than pursue this alternative strategy for achieving coordination. (And notice that this alternative strategy does not require me to have authority. Authority and the capacity to coerce are quite distinct.) Such cases may be relatively rare in politics, since most of us lack the capacity to coerce the many people whose activities our political decisions seek to
367
Democratic Equality and Political Authority
coordinate. But then again, some of us—some public officials, and perhaps leaders of some corporations or religious groups—may have this power; and it would surely matter if it turned out that they lacked good moral reasons to obey the democratic laws that apply to them.
Second, even those of us who lack the power to threaten thousands of people may still not have a good coordination-based reason to abide by the rules imposed by a common authority. Up to now we have assumed that different coordination solutions lie so far apart—that is, moving from one to another requires changing, in a coordinated fashion, the actions of so many people—that only the de facto authority or an almost equally powerful body would be capable of successfully moving us from one solution to another. But in politics, coordination solutions are often quite close, and it takes a lot less than that to shift from one to another.
Think, for instance, of the current tax code (call it C), a scheme of rules for coordinating the activities of thousands, or millions, of citizens. There can be little doubt that many other alternative schemes (C*, C**, and so on) could also fulfill this coordinating function, and that some of these alternatives differ from C only with regard to an exemption granted to a small class of people—at the extreme, a class with a single member. If C* includes such an exemption for a class of which I am the only member, then I can single-handedly (or perhaps at most with the help, or rather forbearance, of my tax inspector) move us from C to C*. And if the outcome of C* is superior to that achieved by C, and the only reason for following the rules of C is that it brings about instrumentally valuable coordination outcomes, then I lack reason to abide by C rather than act in accordance with C*. In fact, if our only concern is with the quality of outcomes, I ought to disobey the law laying down C, and instead single-handedly move us toward the superior coordination solution that is C*.
B. Coordination without Subjection
The examples just discussed show that the need to settle on a coordination solution despite our disagreements is at least at times (and I think reasonably frequently) insufficient to justify authority over all those to whom the law applies. The example also serves another purpose. It reminds us that in many relationships we are not driven by an exclusive concern with coordination outcomes. We also care how coordination is achieved. In particular, we have reason to ensure that considerations of
368
Philosophy & Public Affairs
unequal power that would threaten the valuable egalitarian character of our relationship not play a central role in determining what coordination solution we adopt. If we are committed to nonsubjection, we must avoid relying on power that is clearly distributed unequally, for instance, on the kind of economic or physical power that enables us to bribe or threaten others to converge on our preferred solution. Nor must we rely on power advantages (such as the fact that I can, but others cannot, single-handedly move us from C to C*) the distribution of which is very likely unequal.
And once we consider not just our need for resolving disputes and implementing a common scheme of rights and duties but also the value of doing so without invoking considerations of unequal power—once we are concerned not with achieving coordination simpliciter, but with achieving coordination without subjection—we have the resources for justifying the practical authority of egalitarian procedures. Obeying their outcomes is both necessary and sufficient for the parties to settle on common rules in spite of disagreement without letting asymmetrical power play an important role in determining the settlement. Crucially, the concern with coordination without subjection does not merely give us reason to establish egalitarian decision procedures. Rather, it shows why the parties must treat the outcome of an egalitarian decision procedure as a content-independent and preemptive reason for action, and thus as authoritative.
Let me spell out in more detail the problem that obedience to egalitarian procedures solves. We could achieve coordination without invoking any unequal power advantages if we simply agreed on what solution was best as a matter of justice and the common good. We would not need to invoke unequal power even if we disagreed as long as that disagreement did not require resolution, that is, if there were no need for us to coordinate. And we could even settle on a coordination outcome despite our disagreement if we let unequal power determine the settlement. But if we are barred from invoking unequal power advantages, and yet continue to disagree about the best solution, then coordination—and with it the benefits that only coordination can provide—eludes us.
But it need not elude us if we have available an egalitarian decision procedure that we can treat as a source of authoritative directives. Accepting the decisions of an egalitarian procedure as authoritative is sufficient to achieve coordination without subjection. A procedure
369
Democratic Equality and Political Authority
counts as egalitarian in the relevant sense if the parties bound by its outcome have an equal say over the decisions it makes.43 If all parties to the dispute treat as authoritative the outcome of the procedure, then they will each act on the authoritative determination rather than on their own (possibly divergent) view of what scheme would be best, and thus achieve the benefits of coordination. And in doing so, they will avoid the problem of subjection because the arrangement they settle on is determined not by unequal power advantages, but by the equal power over the decision that the procedure grants to each.
That the authority of the procedure is sufficient to achieve coordination without subjection is already a reason in favor of obeying the procedure. In fact, I want to defend a stronger claim: the egalitarian character of the procedure and the authoritative force of the directive are each necessary for this achievement. That the former is necessary is, I think, obvious enough: unless the procedure that determines how we relate to one another is egalitarian in the distribution of power over its outcomes, following its directives will not help us achieve the equal control over our relationship to which nonsubjection is committed.
Does achieving coordination without subjection also require treating as authoritative—as a preemptive and content-independent reason— the outcome of an egalitarian procedure? Preemptive reasons to do x are both reasons in favor of doing x and reasons against considering (that is, for excluding from deliberation) certain other reasons against doing x. Unless the disputants set aside the considerations of greater and lesser power that they otherwise take into account in determining how best to advance justice or the common good, they cannot hope to realize nonsubjection in their relationship. So the concern for nonsubjection explains the exclusionary demands of the directive. But if considerations of unequal power are excluded, then the disputants' actions are underdetermined by reason: While they may still permissibly act on their judgments of justice and the common good, these no longer tell them what to do when achieving justice or the common good requires coordination with others who hold different views of these values and who
43. This can, but need not, be a matter of giving them an equal say over each decision. It might instead involve taking turns, flipping coins, or even submitting our disagreement to arbitration by an impartial third party. What matters is equal control over the relationship as extended in time, not over any particular slice of it.
370
Philosophy & Public Affairs
are equally barred from relying on considerations of unequal power. So it is not enough for each disputant to simply set aside considerations of power, as this leaves them without determinate grounds on which to act. The disputants also need a first-order reason, acting on which will enable them to converge on a common course of action. The reason each has to set aside other reasons, and to act instead on the reason provided by the directive, accounts for the preemptive structure typical of authoritative directives. The outcome of the egalitarian procedure must, finally, also be a content-independent reason: It is the fact that the directive picks out scheme A rather than scheme B, not the merit scheme A has as such, that gives each a reason to act in accordance with A. (For if it were the merit of the scheme that guided their actions, the parties' disagreement would once again upset their attempts at coordination.) So, under conditions of disagreement, disputants must treat as authoritative the outcomes of egalitarian procedures if they are to achieve coordination without subjection.
C. The Distinctiveness of Egalitarian Authority
How does this justification of authority relate to more familiar accounts of our duty to obey?44 First, it would be a mistake to insist that egalitarian procedures are authoritative only if the disputants consent to being bound by them. Relying on consent would reintroduce the very threat to nonsubjection that the egalitarian procedure is meant to avoid: if a disputant can freely reject the outcome of the procedure, or refuse to let that procedure settle the matter in the first place, he can take advantage of his power and hope to hold out for a better position in a different court.45
Second, the authority of the procedure does not rest on its reliability or expertise. Expertise-based authority depends on the assumption that the bearer of authority is more reliable than the subject. But the decision of an egalitarian procedure can be binding on a subject even if its
44. Here I emphasize differences between the egalitarian justification of authority and others. Elsewhere I have discussed what the egalitarian justification shares with better-known accounts of authority: Daniel Viehoff, "Procedure and Outcome in the Justification of Authority," Journal of Political Philosophy 19 (2011).
45. While no performative agreement is required for such authority, another form of agreement is necessary: if the parties do not share beliefs on who the decision maker is, the decision maker cannot enable settlement on a common scheme.
371
Democratic Equality and Political Authority
judgments of the matter in dispute are less reliable than the subject's own, because obeying the outcome of the procedure (but not acting on his own, more reliable, judgment) will enable the subject to achieve coordination without subjection.46
Third, while the authority of egalitarian procedures is based on a concern for egalitarian relationships, it does not fall prey to the objections such accounts often encounter. Leslie Green has suggested that symmetrical relationships—captured in the notion of "fraternal" as opposed to "paternal" obligations—do not intuitively lend themselves to a justification of authority because "the content of fraternal or sororal obligations lies in the neighbourhood of mutual aid or respect, not obedience." Green continues: "That is why the usual associative model for obedience is not in fact the horizontal association among siblings but the vertical hierarchy of parent and child."47 Green is surely correct that the paradigmatically symmetrical relationships among siblings do not normally include a duty to obey one's brother or sister. Yet this does not show that the value of symmetrical relationships cannot provide the basis for authority—if the authority is attributed not to one of the symmetrically situated parties to the relationship, but rather to another party or procedure, obedience to which enables disputants to uphold their symmetrical relationship in spite of their disagreement.48
VI. THE LIMITS OF EGALITARIAN AUTHORITY: INEQUALITY, INJUSTICE, TYRANNY OF THE MAJORITY
If we aspire to uphold the ideal of relational equality among us as fellow citizens, then we have reason to ensure that economic, physical,
46. This is compatible with the thought that the procedure's reliability bears on the scope of its authority. I discuss this point in the next section.
47. Leslie Green, "Law and Obligations," in The Oxford Handbook of Jurisprudence & Philosophy of Law, ed. Jules L. Coleman and Scott Shapiro (New York: Oxford University Press, 2002), p. 534. Green's target here is the account of associative obligations in Ronald Dworkin, Law's Empire (Cambridge, Mass.: Belknap Press of Harvard University Press, 1986).
48. On the egalitarian view I defend, it is the intrinsic value of the symmetrical relationship among the subjects (and the corresponding need to exclude those considerations that would threaten such symmetry), rather than that of the hierarchical relation the subjects have to the bearer of authority, that justifies the subjects' duty to obey. One way to bring home this point is by highlighting that authority of the sort I defend here depends on the existence of a dispute among the subjects. On traditional paternalist accounts of authority, by contrast, the parent's (or king's) decision would be binding even if the subjects agreed among themselves on how to act or live together.
372
Philosophy & Public Affairs
or other advantages do not enable some of us to exercise asymmetrical control over the rules governing our common life. This, in turn, explains the authority that an egalitarian decision procedure like democracy has for us: When we disagree with one another and must find ways of settling our disagreement without invoking one person's greater physical power or lesser economic need, then we must be willing to accept as authoritative the outcome of an egalitarian procedure that makes decisions independently of just these considerations. We must accept its decision as authoritative since the alternative is either that we do not come to a settlement or that the settlement is determined by considerations we want to exclude from our relationship. The egalitarian discipline that democracy imposes on our decision making can thus play a central role in justifying the practical authority of our laws.
By grasping the structure of the argument, we can in turn understand the limits to justified authority that follow from it. First, and perhaps most obviously, an institution or a procedure can lack authority because it is insufficiently egalitarian. As I pointed out earlier, all actually existing democratic states fall quite short of the requirement of equalizing political control over their citizens' common life. Does this mean that they all lack the authority that this argument would otherwise establish? No. Where the shortfall is too great, their decisions lack the authority they could have had were political institutions more egalitarian. But even institutions that are not perfectly egalitarian can possess authority based on the argument developed here. The argument rests on the thought that we better realize the ideal of relational equality by following democratic decisions than by seeking to advance justice directly; and this does not require that democratic decisions are perfectly equal or enable us to achieve perfect relational equality. We have reason to come closer to the ideal even if we are still quite far away from it; and this we may be able to achieve even if our actually existing political institutions are unequal in various ways.
Second, an egalitarian procedure may lack authority because it is insufficiently likely to reach the correct conclusion. This is not to say that submission to the authority is justified by the concern for the quality of the outcome. It is, rather, to say that the benefits of submission to the democratic authority, measured in terms of relational equality, may be insufficient to justify the expected costs measured in justice or the common good. That the quality of the outcomes bears on the authority
373
Democratic Equality and Political Authority
of the procedure does not, however, entail that the purported subject can ask on a case-by-case basis whether the decision is insufficiently just or good. The value of relational equality is threatened not only where I in fact disobey the decision, but even where I threaten to disobey it if it takes (what I believe to be) a wrong turn and no such turn is taken. So to determine whether disobedience is justifiable, we must, in the first instance, ask not how disobeying in any particular instance will advance justice, but rather whether a general moral rule of disobeying under certain circumstances will advance justice sufficiently to make up for the threat this poses to relational equality.49
Finally, even where a decision is only somewhat unjust, and the procedures themselves distribute political power relatively equally, a decision may lack authority because the citizens vote or act on the basis of the wrong kinds of reasons. This may seem like a surprising claim. It follows, however, straightforwardly from the earlier discussion of relational equality. The intrinsic value of egalitarian relationships depends (among other things) on the equal concern that the parties have for one another. Where I treat the equal power that I have merely as an instrument for advancing my own good, without concern for the well-being of my friend or colleague, or the common good or internal justice of our group or community, submitting to our collective decision does not uphold a valuable relationship among equals. So treating it as authoritative is not justified on the basis of the argument presented here.
This observation helps us better understand the role that appeals to the "tyranny of the majority" play in thinking about democracy and its authority. Often complaints about majority tyranny merely highlight that a minority has lost on an issue dear to them (perhaps, in particular, an issue involving rights claims). But losing is part of democratic
49. Jeremy Waldron has argued that "there is something of a 'category-mistake' in treating justice and fairness as co-ordinate principles, competing on the same level." Waldron, Law and Disagreement, p. 196. This is because justice and fairness are not independent of one another, but rather "functionally related": "It is the task of political fairness to address the situation that arises when people in a society cannot agree about justice, and thus cannot act unequivocally as a society on the basis of an appeal to justice alone." Ibid. But even if political fairness comes into its own as a demand only where we disagree about what justice requires, this does not show that the value that fairness helps us realize—the value of relational equality—is not distinct from, and cannot conflict with, justice. All it shows is that, as long as we agree about what justice requires of our relationship and act on that shared view, we will normally have also satisfied the demands of relational equality.
374
Philosophy & Public Affairs
politics. Sometimes such complaints are meant to be limited to those losses that create severe injustice, and thus perhaps to cases where the procedure lacks authority on the basis of the argument from injustice we considered a moment ago. But perhaps a better way of understanding this claim is a third: Tyranny, as Aristotle explained, arises where those who rule aim at their private benefit.50 This is not, in the first instance, a matter of who ultimately benefits from the laws (though it will evidently often have an effect on that). It is, rather, a matter of the attitude with which public power is exercised. Tyranny is always problematic. But it is especially problematic if the authority the rulers wield is based on the value of a relationship that depends on an attitude of equal concern and the good faith pursuit of the common good. Thus, where democratic institutions are knowingly set up to unjustly benefit some members of society and not others, or where citizens vote solely to advance their own interest and without an eye toward how their vote contributes to the common good, the distinctive authority of egalitarian political procedures is undermined. The flip side of this account of the duty to obey democratic laws is thus a rather demanding account of the duties that citizens, and their representatives, have to exercise their vote conscientiously.
VII. CONCLUSION
I have argued that democratic decisions at least sometimes have special authority because they are made by procedures in which all citizens have an equal say. But, contrary to what one might expect (and what the argument from public equal respect assumes), this is not because equality requires us to treat other people's judgments as reasons for action. It is, rather, because, by treating as binding the outcome of an egalitarian decision procedure, we can avoid acting on various considerations—in particular, unequal power—that we have reason to exclude from our relationship. If democracy has special authority based on procedural equality, this is not because it enables us to rule ourselves, but because it protects us from the threat of being ruled over unequally.
Let me conclude with one final observation. We can avoid unequal rule not only by making decisions through democratic procedures that
50. Aristotle, Politics, ed. C.D.C. Reeve (Indianapolis: Hackett, 1998), p. 77 (book 3, chap. 7).
375
Democratic Equality and Political Authority
are positively (and equally) responsive to our votes. We can also do so by making decisions in ways that are completely unresponsive to anyone's judgment—for instance, by flipping coins. So coin flipping too could be a source of authoritative directives on the account just offered. Yet it is clear that we do not usually make political decisions by flipping coins. Should it thus concern us that the account justifies the authority of both these procedures?511 do not think so. Even if both procedures could have practical authority over us, there are plenty of other considerations that may justify our preference for democracy: Democracy may, for instance, be epistemically superior. Or there may be independent value in the exercise of political agency involved in deliberating and voting that would be lost if we flipped coins instead. These considerations could give us reason to establish democratic over other egalitarian procedures even if our duty to obey the procedure (once established) depends not on these considerations, but on the procedure's egalitarian character.
51. David Estlund worries that this is a problematic feature of theories of democracy that emphasize its procedurally egalitarian character: Estlund, Democratic Authority, p. 82.