To hold political authority is to have a right to rule, and those subject to the authority have an obligation to obey its directives. That is the core of the standard or traditional account of the relationship between authority and political obligation. To be sure, few philosophers, living or dead, would endorse this account without elaboration or qualification. Among other things, most would insist that the authority in question must be genuine or legitimate rather than merely apparent, that its directives do not exceed the authority's proper bounds, and that the obligation to obey is in some way defeasible rather than unconditionally binding. With such qualifications in mind, however, it has long been thought that there is a direct connection between the existence of political authority and a general obligation to obey its laws.
This has been the view, moreover, not only of those who maintain that there is a general obligation of obedience but also of those who do not. Anarchists have long denied that anyone has an obligation to obey any laws, of course, but they have done so because they reject all claims to political and legal authority. In recent decades, these political anarchists have been joined by others whose doubts about the claims of political authority stop short of calling for the abolition of the state. The most radical of these are the philosophical anarchists, such as Robert Paul Wolff and A. John Simmons, who deny the legitimacy of political authority and the existence of a general obligation to obey the law while being content to let the state remain in place.Footnote 1 In Wolff's case the argument is that authority is incompatible with our fundamental moral duty of autonomy; in Simmons's, it is that all attempts to show how the citizens of even a reasonably just state have a general obligation to obey its laws have failed. Neither Wolff nor Simmons, however, nor their counterparts among the political anarchists, have questioned the connection between political authority and political obligation.Footnote 2 That step has been left to the diverse group of revisionists who are the principal concern of this essay.
According to these revisionists, there are no good grounds for doubting the possibility and existence of political authority, but that is largely because the connection between authority and obligation is either weaker than the traditional account holds or because there is no necessary connection between them at all. In their view, the fact that a group of people holds legitimate political authority does not entail a correlative obligation of obedience on the part of those subject to that group's rule. Those who are subject to authority may well find themselves under obligation, according to the revisionists, but it will not be an obligation to obey the laws issuing from the authorities simply because they are laws. The existence of political authority thus does not ensure that its subjects will have a general obligation to obey the law as such.
My purpose in this paper is to defend the traditional account of political authority and obligation by resisting this revisionist retreat. The arguments advanced by those who want to sever the connection between authority and obligation are somewhat various, however, which means that the defense will have to proceed by responding seriatim to the revisionists. Following these responses, to be set out in Section II of this paper, I turn in Section III to the positive task of sketching a satisfactory version of the traditional account—one that relies on the fair-play theory of political obligation. I then relate this theory, by way of conclusion, to Scott Shapiro's advocacy of an “Arbitration Model” of political authority.Footnote 3 I begin, however, with a brief review of some of the standard concepts and distinctions in the debates over authority and political obligation.
I. CONCEPTUAL CONSIDERATIONS
Discussions of political and legal authority typically acknowledge the difference between two kinds of authority. On the one hand, there is the kind of authority that someone can have; on the other hand, there is the kind that someone can be. The two are not unrelated, of course. To have or hold a position of authority in the Society for American Baseball Research, for example, one almost certainly will need to be an authority on baseball. But there is no necessary connection between the position and the expertise, and no reason to think that someone who is an authority on a subject will be in authority over others. Nor, conversely, is there reason to think that someone who is in a position of authority must be an authority on anything at all. Although there frequently are competency standards that judges, police officers, and others in positions of authority must meet, these need not be so strict as to qualify anyone who meets the standards as an authority on the subject in question. Moreover, there are some ways of gaining positions of authority—inheritance, election, becoming a parent, and random selection among them—that require little if anything by way of authoritative mastery of a subject.
This distinction is often put in terms of a difference between theoretical (or epistemic) and practical authority, with political authority generally taken to be a form of the latter. There are those, though, who hold that political authority, while clearly practical in its concern for how people are to act, must—or should, at any rate—also partake of a considerable amount of theoretical authority. Plato's apparent argument in The Republic for rule by philosophers is a clear example, but so too is David Estlund's recent brief for the epistemic authority of democracy.Footnote 4 Even so, few philosophers have been inclined to collapse practical into theoretical authority, for doing so would require us to limit some commonly acknowledged forms of practical authority, such as parental authority, to those who possess the relevant theoretical qualifications.
Another familiar distinction in discussions of political and legal authority is that between de facto and de jure authority. In this case the point is that someone may exercise authority to which he or she is not entitled, on the one hand, or hold title to authority that he or she is not effectively able to wield, on the other. On some accounts de facto authority seems to be little more than brute power, but the more common view is that it differs from power because some degree of acceptance on the part of the subjects, rather than mere acquiescence in the face of threats and violence, must be present. There may even be cases of complete acceptance on the part of the subjects, as when an impostor surreptitiously deposes the rightful ruler and assumes her place, thereby exercising de facto an authority to which the impostor has no title. Perhaps the most telling difference between power and de facto authority, though, is that the latter necessarily implies an appeal to de jure authority. Those who rule by sheer might need not concern themselves with questions of right or title; they can simply rule, as long as they are able, through terror and intimidation. But when they lay claim to authority, they are necessarily appealing to some idea of rightful rule or ruling by right, which typically includes the claim that the subjects have a corresponding obligation. As Joseph Raz puts the point, “Having de facto authority is not just having an ability to influence people. It is coupled with a claim that those people are bound to obey.”Footnote 5
The other side of the coin here is that any claim to de jure authority also implies some degree of de facto authority. That is, no matter how clear a title to authority a group may have, it must be able to exercise authority to some extent if it is to retain its authority. If usurpers drive the group in question out of office and into a remote territory, so that no one in the homeland is in a position to follow the directives of those whom they continue to regard as their rightful governors, then the government in exile must either find a way to displace the usurpers or watch helplessly as its authority dwindles away. Authority must be effective not only to be de facto, in other words, but also to be de jure.Footnote 6
According to many accounts, the distinction between de facto and de jure authority is congruent with the distinction between authority and legitimacy. Some writers do treat “authority” and “legitimacy” as virtual synonyms when they equate “those in authority” with “the legitimate government,” but doing so makes a redundancy of the widely used phrase “legitimate authority.” Other writers try to separate legitimacy from authority, as when Estlund defines authority as “the moral power to require or forbid action” and legitimacy as the morally permitted use of coercion, so that “a state's uses of power are legitimate if and only if they are morally permitted owing to the political process that produced them.”Footnote 7 For Estlund, then, legitimate authority combines the right to rule with permission to enforce that right by means of coercion. But that is a point one can agree to without accepting Estlund's distinction. If one takes “authority” to mean having a moral right to rule, with that right including a right to use coercion, then to talk of “legitimate authority” is simply to reinforce the point that the authority in question is not merely de facto but also de jure—that is, not only effective but morally justified authority. That, anyhow, is the sense in which I will use “legitimate” and its cognates in this paper.
Two final points to note with regard to political authority are that its directives are generally taken to be preemptive (or peremptory)Footnote 8 in nature and content independent. According to the first point, laws preempt or exclude any other reasons for action someone subject to their authority may have. If my parents or employers or church leaders tell me to do something that is contrary to the law, they give me reasons to break the law; but from the law's point of view, its directives preempt the others. That is not to say that people may never have decisive reasons, moral or prudential, for disobeying the law; it is to say that the law does not typically recognize these reasons.Footnote 9 Moreover, the authority of laws, considered individually, does not depend on their content. Law is binding because it issues from the proper authorities regardless of what its content may be. There are some limits here associated with the idea of legality, such as strictures against directives that command the impossible or are not made public, but otherwise a law—at least on the standard or traditional account—is binding simply because it is the law.
This quality of content independence is usually taken to be a feature of political obligations, too. One's obligation or duty to obey does not depend on what it is that one is required to do but simply on the fact that the law requires it. Again, there may be circumstances that justify disobedience, but in the ordinary course of affairs—and from the law's point of view—laws are binding regardless of their content. The subject's obligation is content independent, then, in that it is an obligation to obey the law as such. Whether obligation or duty is the proper word to use in these contexts sometimes has been a matter of dispute, with some writers using “duty” for broad moral requirements and reserving “obligation” for narrower ones. We shall encounter one such writer, David Copp, in Section II of this paper. The general idea is that we need to take some action in order to acquire or incur an obligation—to make a promise, for example, or take part in a cooperative activity—whereas duties to contribute to charity, to aid those in distress, and support just institutions, among others, are moral responsibilities that bear on us regardless of what we have or have not done. Most of those who discuss political obligation these days treat “obligation” and “duty” as equivalent terms, however, and I shall do the same in this paper. Almost all regard these obligations or duties as moral in nature, too, so that a political obligation is understood to be a moral duty to obey the law.Footnote 10 From this point of view, which I share, the problem of political obligation thus takes the form of the question, Do the citizens of any polity, even one that is reasonably just, really have a general and moral obligation to obey the law?
As I have noted, the revisionists join the anarchists, both political and philosophical, in answering this question negatively. The revisionists differ from the anarchists, though, in holding that a rejection of political obligation need not entail a rejection of political and legal authority. But what are their arguments? And are they sound? To these questions I now turn.
II. RESPONDING TO THE REVISIONISTS
The arguments of the political and legal philosophers I am calling the revisionists are various, but they tend to fall into three categories. The first two categories comprise philosophers who reject the claim, or assumption, that the right to rule or govern is, in terms made familiar by W. N. Hohfeld, a claim-right—that is, a right that entails a correlative obligation or duty on at least one other person's part.Footnote 11 According to the first set of revisionists, political and legal authority is better understood as a justification-right; according to the second set, as a power (or power-right). The third category of revisionists comprises philosophers, most notably Joseph Raz, who also conceive of political authority as a moral power. Their emphasis, however, is less on authority as power than on Raz's influential “service conception of the function of authorities.”Footnote 12 There are differences of focus, emphasis, and arguments among those in these three categories, in short, but the revisionists all agree that the presence of legitimate authority is no guarantee of a general obligation to obey the law.
A. Authority as Justification-Right
Robert Ladenson appears to have been the first of the revisionists even though his much-discussed essay, “In Defense of a Hobbesian Conception of Law,” does not directly address the topic of political obligation.Footnote 13 Nevertheless, Ladenson's account of the authority of law clearly implies, as he says, that “no neat logical connection holds between the right to rule and the duties of subjects with regard to allegiance to the state and compliance with the law.”Footnote 14 Nor does Ladenson mention Hohfeld in his article, which leads to the question of exactly where his identification of the right to rule with a “justification right” fits into Hohfeld's scheme of four kinds of rights (or “jural relations”).Footnote 15 It is clearly not, in Ladenson's view, a claim-right, nor is it an immunity; but is it a liberty-right (or privilege) or a power (or power-right)? Or is it a kind of right that Hohfeld's schema fails to capture? Ladenson offers self-defense as a paradigmatic example of a justification-right, which seems to suggest it must be a Hohfeldian liberty-right—that is, a right imposing no duties on others and consistent with everyone else having the same right (in the sense of liberty or privilege). However, Ladenson also suggests it is akin to a power-right when he says, “strong reasons can be advanced for holding that possession of the governmental power and acceptance by those one presumes to govern of its exercise jointly constitute a justification for coercive acts which would otherwise be immoral.”Footnote 16 Whether it rests on a liberty-right, a power-right, or something outside the boundaries of Hohfeldian analysis, Ladenson's claim is that “the authorities, and the authorities alone, are morally justified in visiting evil upon individuals who perpetrate acts of a kind that tend to disrupt the social peace.”Footnote 17 To be justified in exercising authority in this way is, in his view, to hold a justification-right.
This conception of political authority as a justification-right is open to serious criticism. As Raz notes, there is a difference between “the justified use of coercive power”—as in Ladenson's paradigmatic justification-right, self-defense—and authority. The difference is that those who hold authority must at least claim or imply that they have a right to compliance: “[t]he exercise of coercive or any other form of power is no exercise of authority unless it includes an appeal for compliance by the person(s) subject to the authority.”Footnote 18 By obliterating the distinction between justified coercion and authority, Ladenson's argument leads to the unacceptable conclusion that “all de facto authorities are legitimate.”Footnote 19
To be fair, Ladenson does say that acceptance by the governed is one of two conditions that define authority qua justification-right; the other is “effectively uncontested governmental power.”Footnote 20 Nor is acceptance nothing more than prudential acquiescence to those who hold power. On Ladenson's account, those who are subject to effective governmental power have reason to be grateful for the security and stability it provides, especially as governmental authority is not an unlimited right to rule in any way that the governor chooses. Exactly what its limits are is not clear, but the point seems to be that the subjects of “effectively uncontested governmental power” invest it with moral authority when they come to regard it as having a justification-right to rule them.
The problem with this argument is that Ladenson does not make clear what counts as acceptance—or perhaps authorization—on the part of the governed. To make matters worse, it appears that any attempt to provide clarity here would place him on the horns of a dilemma. If he defines acceptance as something like consent to or open acknowledgment of the rulers’ right to rule, then he seems to be committed to the view that authority rests on an obligation that those subject to the authority have somehow undertaken or imposed upon themselves. But that is to make a claim-right of authority, with a correlative obligation on the part of the subjects, despite Ladenson's assertions about justification-rights “neither presupposing any institutional background nor correlating with duties.”Footnote 21 This leaves Ladenson with the option of defining acceptance as something like sensible acquiescence in the face of overwhelming power. In that case, however, acceptance is a hollow notion and political authority is simply “effectively uncontested governmental power.” Understood as a justification-right, then, political authority is nothing more than de facto authority—and possibly nothing more than sheer power.
Similar problems beset Rolf Sartorius's attempt to explicate authority as a justification-right, but there are two respects in which his argument in “Political Authority and Political Obligation” differs significantly from Ladenson's.Footnote 22 First, as the title indicates, Sartorius is directly concerned with the relationship between political authority and obligation, and he aims to overturn the traditional belief that one entails the other. “On this [traditional] view,” he protests, “the notion of a fully general moral obligation to obey the law, whatever its content and without regard to the consequences of obeying it, represents a moral blank check that government is free to fill in at will or at least within very broad limits.”Footnote 23 Second, Sartorius differs from Ladenson in relying on the idea of political authority as a trust. Ladenson may hint at this idea, but Sartorius makes it explicit: “[g]overnment … holds a trust; it has both the right and the responsibility to act in certain ways for the benefit of its citizens, but these beneficiaries have no correlative obligation to do what it requires in the course of its exercising its putative beneficence.”Footnote 24 To have political authority is, therefore, not only to have a justification-right to rule but also a responsibility to rule in the interests of those subject to its authority.
Invoking the idea of a trust may help to clarify the justification-rights account of authority, but it does not free the account from its difficulties. In fact, it leads to questions about how trust is to be understood, and attempts to answer these questions simply raise further difficulties. To begin with, it is not at all obvious that conceiving of authority as a trust is at odds with the belief that authority is a claim-right. After all, John Locke insists in his Second Treatise that government is a trust of this kind, yet he also holds that there is a general obligation to obey the law when those in power exercise legitimate authority. Moreover, typical examples of trusts do not support the conclusion that the relationship between those who establish the trust and the trustees is one that invests the trustees with nothing more than justification-rights. If you hire me to manage a fund or account for you, you are presumably justifying me in acting on your behalf. But I also have a duty to you to act within the bounds of your charge, and you have a right to hold me to account. More to the point, you also have a duty, so long as the trust remains in effect, to allow me to exercise my right to act as I think will best promote your interests. All of which is to say that the rights of the trustee appear to be claim- rather than justification-rights; or they may be, at most, a bundle of rights that include both claim- and justification-rights. But they are not simply justification-rights.
The kind of trust to which I have been referring is, of course, one that involves two or more adults. This is not the only kind of trust, however, and Sartorius relies heavily on the idea of parental authority as a relationship analogous to political authority. In his words, “the foundation of political (and parental) authority arises from the necessity of a task whose successful performance requires customary compliance on the part of those for whom the task is supposed to be done.”Footnote 25 Whether this analogy is helpful to Sartorius's argument, though, is doubtful at best. For one thing, parental authority is commonly taken not only to require customary compliance on the part of the children in their care but also to entail a duty of obedience on the part of the children, at least when the children are old enough for talk of duties to be meaningful to them. Parental authority is thus not a clear example of a justification, and only a justification, right. But even if it were, the implications for Sartorius's conception of political authority are not likely to be acceptable. For if political authority is analogous to parental, as he says, then the governing authorities stand in the same relation to those they govern as parents do to children; and when the governed ask why they should comply with the directives of the authorities, the answer is straightforward: “[b]ecause it's for your own good.” Conceiving of political authority in this way seems to be at odds with Sartorius's previously quoted worry about theories of political obligation that give government “a moral blank check.”
For Sartorius as for Ladenson, then, the appeal to a conception of political authority as a justification-right leads either back to the traditional view that political authority entails an obligation of obedience on the part of subjects or to the reduction of legitimate authority to justified coercion—to “effectively uncontested governmental power,” in Ladenson's words. They give us no reason, as a result, to abandon the belief that the right to rule is a claim-right.
B. Authority as Power-Right
The second category of revisionists comprises those who hold that political authority is best understood as a Hohfeldian power (or power-right). If successful, their arguments will allow them to justify political authority without having to confront the apparently devastating attacks Wolff, Simmons, and others have directed against the belief in a general obligation to obey the law. If the right to rule is a power-right, in other words, it entails only that those subject to authority are under a liability with regard to those in authority, and not under an obligation or duty of obedience. Belief in political obligation may be cast aside, then, without taking political authority with it.
Stephen Perry provides a clear example of this position in a long and subtle essay in which he raises what he calls the “reverse-entailment problem.”Footnote 26 The problem, as Perry formulates it, is that “the existence of legitimate authority logically entails an obligation to obey, but an obligation to obey does not logically entail the existence of legitimate authority.”Footnote 27 This is true, furthermore, even when the obligation in question is a general obligation to obey the law, and the problem “can arise even if legitimate authority—the right to rule—is understood as … a claim-right.”Footnote 28 Indeed, the problem must be devastating for those who understand political authority to be a claim-right, for they are committed to the reverse-entailment thesis—that is, political authority entails political obligation and political obligation entails political authority. Because the reverse-entailment problem undercuts the reverse-entailment thesis, in short, the standard account of the relationship between political authority and political obligation must fail.
To support this claim, Perry produces two examples of hypothetical legal systems in which general obligations to obey the law appear to be in place but in which “we nevertheless remain, at best, uncertain as to whether or not the state has a right to rule.”Footnote 29 In one example, every directive of the state accords with something we have an independent moral duty to do or not do: do not commit murder, rape, robbery, fraud, and so on. We would have a general obligation to obey the law in such a circumstance, according to Perry, but that obligation would tell us nothing about the putative authority of the state or the law. In the second example, Perry envisions a legal system in which “any instance of law-breaking whatsoever will inevitably take place under conditions that will set a bad example for others, and for that very reason every subject of the legal system has at a least a pro tanto obligation to obey each and every law.”Footnote 30 Again, Perry concludes, the presence of a general obligation to obey the law in these circumstances is not enough to establish the presence of legitimate authority. In this case the problem of the independent moral merit of the law again arises; why else would we worry about someone's disobedience setting a bad example? Moreover, the argument from not setting a bad example presupposes what needs to be demonstrated, which is that “there is some good to be derived, at least potentially, from some persons having the moral power to subject other persons to obligations.”Footnote 31
These are clever examples, but I think they fail, for three reasons, to substantiate Perry's identification of a reverse-entailment problem. First, and most obviously, the examples are far-removed from actual legal systems. To imagine a legal system with no laws that do not coincide with independently grounded moral injunctions is to imagine one with no laws that address coordination and collective-action problems or that take a stand with regard to matters of moral controversy, such as capital punishment and abortion. Second, even if we grant the value of Perry's hypotheticals, the most that they can establish is that “a reverse-entailment problem can arise even when the argument purporting to establish the existence of a general obligation to obey the law has impeccable conceptual credentials.”Footnote 32 Can is not sufficient, though. What Perry needs to show is that a reverse-entailment problem must arise in these circumstances. Finally, we should note that the kind of “general obligation to obey the laws” that Perry provides in his examples is not the kind that advocates of the traditional account of political authority and obligation have in mind, for it is not an obligation to obey the law as such—not an obligation to obey the laws because they are laws. The upshot is that he has not shown that there truly is a problem that undermines the position of those who believe in a reverse entailment between political authority and obligation.
More generally, the problem for those who define political authority as a power rather than a claim-right is that it is not obvious that they gain anything other than a sense of fidelity to Hohfeld's scheme. But even that gain comes at the cost of a departure from ordinary understandings of authority. Indeed, it is noteworthy that none of the revisionists has proposed, so far as I am aware, that we abandon talk of authority as the right to rule, or that we refer to our “civic duties” as “civic liabilities.” Nor does shifting attention from obligations to liabilities do more than postpone the challenge of providing a justification for political obligation. I say this because power and its “jural correlative,” liability, are both prospective terms. That is, each characterizes a current condition in terms of its potential—of what can happen. You may have power over me, leaving me liable to gain or suffer from your actions, without your ever exercising that power. In the case of legitimate authority, of course, we are not talking about sheer power but about a moral capacity, so that your authority over me is the power to impose moral requirements on me or to release me from duties I currently have. Power and liability are still forward-looking concepts, however, with liability to be understood not as a duty itself but as a liability to be subjected to duties. Indeed, Perry takes this to be an advantage of the authority-as-power approach:
once we recognize that the Hohfeldian correlate of a power to impose duties is a liability, not a duty, it is easy to see that, in principle, the power can exist without ever being exercised. The normative status of the person over whom the power is held is that he or she is liable to be subjected to a duty, not that he or she is in fact under an existing duty.Footnote 33
To make the case for political authority, then, we need not show that those who are subject to the authority have an obligation or duty to obey its laws, but only that they are under a liability to have such obligations or duties imposed on them.
But is this truly an advantage of the authority-as-power approach? In particular, what is the advantage of shifting attention from the subjects’ obligation to obey the law to the subjects’ liability to have obligations imposed on them? In either case one will still have to establish that there is a general responsibility on the part of subjects, whether we call it a duty or a liability, in order to make a case for the legitimacy of the authority over them. We may also construe a liability as a kind of duty in itself. That is, those who hold authority-as-power and those who are subject to having duties imposed on them stand in a certain relationship to one another, and it seems reasonable to say that those who are liable to have duties imposed on them already have a duty to obey while they are waiting for their orders to come. In any case, the obligation or duty to obey has to enter the picture at some point, whether presently in the standard account of political authority and obligation or prospectively in the authority-as-power approach. Revisionism of this kind seems to do nothing more than postpone the day of reckoning.Footnote 34
This objection will be beside the point, however, if the revisionists can provide an account of political authority qua power that is clearly divorced from a general obligation to obey the law. Those who fall into my second category of revisionists have tried to do this, and their arguments are worthy of closer attention than I can give them here. Brief consideration of two essays, though, is in order.
The first of these essays is Arthur Isak Applbaum's appropriately entitled “Legitimacy Without the Duty to Obey.”Footnote 35 According to Applbaum:
to judge an authority legitimate simply is to judge that the subjects of that authority are morally liable—that is, not morally immune—from the exercise of a moral power to impose and enforce conventional duties and change relevant social facts in ways that change the subject's normative situation. Whether subjects face a moral duty as well remains an open question.Footnote 36
One of the advantages of this view, Applbaum argues, is that it enables us to make better sense of the belief that civil disobedience may be warranted even when the authority in question is legitimate. For if we take legitimate authority to entail an obligation of obedience on the part of the citizens, we face the following quandary: “[i]f the authority is legitimate, disobedience is not justified. If, by assumption, disobedience is justified, then the authority that is disobeyed cannot have been legitimate.”Footnote 37 So much is true, at least, if legitimate authority entails “a dispositive duty to obey,” in which case “civil disobedience disappears as a poignant moral phenomenon.”Footnote 38 The problem evaporates, however, if we conceive of authority as a power that entails a correlative liability on the part of the citizens subject to the law.
Applbaum's argument rests in part on two controversial claims about duties. One is that genuine conflicts of duty cannot occur because genuine duties are necessarily dispositive; the other is that there is a clear distinction between moral duties and those that are merely legal, conventional, or institutional. If we grant these claims, then it is easy to see how the duty to disobey a law may be dispositive—that is, the only moral duty in the case at hand—because the duty to obey the law is at most an institutional duty. The question, of course, is whether we should accept either of these controversial claims. Accepting the first does square with the desire that in every moral case there must be one all-things-considered right course of action, but it also precludes the possibility of truly tragic situations and genuine moral dilemmas. In the case of the second claim, there does seem to be a difference between such clear examples of moral duty as the duties of charity, rescue, and supporting just institutions, on the one hand, and the institutional duties involved in occupying a position and doing one's job, on the other. But the distinction is easily overdrawn, and we should not lose sight of the moral element involved in attending to “my station and its duties,” as Plato and F. H. Bradley have observed. So far as I can see, there is no compelling reason to agree to either of Applbaum's claims about duties. Nor does disagreeing with Applbaum entail the disappearance of the “poignant moral phenomenon” of civil disobedience. For we can continue to regard civil disobedience as justified, all things considered, when the moral grounds for disobedience outweigh the prima facie or pro tanto obligation to obey the law. Such a stance is even open to someone who shares Applbaum's conviction that legitimate authority is better understood as a moral power than as a claim-right.Footnote 39
With regard to civil disobedience, then, Applbaum's version of authority-as-power revisionism is not at all superior to the traditional understanding of authority-as-claim-right. Applbaum proceeds, though, to set out two examples in support of his contention that legitimacy does not entail a duty of obedience. I will confine myself here to the first example, in which a court's ruling in favor of Beachowner legally reverses the relationship between those who own beachfront property and those, such as Clamdigger, who have exercised a customary right to follow a path through Beachowner's property to the shore. As a result of the ruling, Beachowner now has a right to exclude others from his property and Clamdigger now has a duty to stay off the path. Clamdigger acknowledges the legitimacy of the court's ruling, if not its correctness, and further acknowledges that he now has a legal duty to use the path only with Beachowner's permission. But he denies that he has a moral duty to obey the ruling, skips over the chain that Beachowner has stretched across the path, and “walks down the path to dig clams as he always has done.”Footnote 40 Nor, Applbaum contends, is Clamdigger making a conceptual mistake regarding obligation and authority when he does this. In fact, Clamdigger can quite reasonably conclude that his normative situation has changed, as has Beachowner's, but changed so that each now has a moral privilege with regard to the other. That is, he can reasonably conclude that Beachowner now is morally at liberty to try to block entrance to the path while he, Clamdigger, is morally at liberty to continue to use the path. He no longer has a claim-right to do so, but neither does he have a moral obligation to stay off the path, despite the ruling of a legitimate authority.
Whatever support this example lends to Applbaum's claim follows, I suspect, from his sharp distinction between legal and moral duties. If we are suspicious of this distinction, as I have suggested we should be, the plausibility of the example fades away. It is one thing to grant that an all-things-considered moral duty outweighs a (merely) legal or institutional duty, but quite another to hold that the latter is a not-at-all-moral duty. It is also possible to grant that two people may have moral privileges that put them at cross-purposes, as would be the case if Beachowner were morally permitted to (try to) exclude Clamdigger from a path that Clamdigger is morally permitted to (try to) use. In such circumstances, though, either one side will give in or conflict is bound to arise, in which case some way to resolve the conflict—probably further recourse to the legal authorities—must be found. Indeed, Applbaum says that he is “well aware of the instability and uncertainty that positing such misfirings of authority creates, of the difficulties that arise from letting us be judges in our own cases, of the need for procedural finality to settle substantive disagreement, and the like.” But he insists that Clamdigger makes no “conceptual error” in holding that he has no moral duty to obey the court's legitimate ruling.Footnote 41 We may concede Applbaum's point, I suppose, if we are willing to grant that we, like Clamdigger, can remain conceptually correct while contributing to the instability and uncertainty that positing misfirings of authority creates. But that is quite a lot to grant.
It seems too much to grant, in fact, even for revisionists who share Applbaum's desire to replace the traditional account with one that treats authority as a Hohfeldian power. For Perry and other writers in this category, the correlate to this power is a liability, as Applbaum says, but this liability is at most one remove, as I have argued, from a duty or obligation. For David Copp, the relation seems to be even more direct. In “The Idea of a Legitimate State,” Copp proposes “that a legitimate state would have the power to put its residents under a pro tanto duty to do something simply by enacting a law, provided that the law is morally innocent.”Footnote 42 Copp also argues in this essay that this pro tanto duty is part of a “cluster” of Hohfeldian advantages that serve to distinguish legitimate from illegitimate states. This cluster contains the following features:
(1) a sphere within which it has a privilege to enact and enforce laws applying to the residents of its territory; (2) a power to put people residing in its territory under a pro tanto duty to do something simply by enacting a law that requires them to do that thing, provided that the law falls within its sphere of privilege and is otherwise morally innocent; (3) a privilege to control access to its territory by people who are not residents and have no moral claim to live or travel there; (4) a claim against other states that they not interfere with its governing its territory; (5) an immunity to having any of these rights extinguished by any action of any other state or person.Footnote 43
As the last three items in this cluster indicate, Copp's concern with the legitimacy of states extends beyond the usual bounds of discussions of political authority and obligation, which typically confine themselves to questions concerning the relationship, if any, between authority and the obligation to obey its laws. I shall accordingly consider only the first two elements of Copp's cluster here.
Together with their companions, these two elements are supposed to have the Hohfeldian virtues of disentangling and clarifying what has been long obscured by the tendency to conceive of political authority simply as the right—or more precisely, the claim-right—to rule. In addition, the two elements are supposed to provide an understanding of legitimate authority that escapes the challenge posed by Wolff, Simmons, and others who deny or doubt the possibility of a satisfactory account of political obligation. These would be notable accomplishments for Copp's cluster-of-advantages theory of authority if his arguments were successful, but there are problems with both of the relevant elements in his cluster.
With regard to the first element, the problem concerns Copp's claim that “a sphere” of “privilege” is part of the right to rule. There are, in fact, two problems here. The first is whether privilege is the proper term; the second is how a sphere of privilege comes into the picture. According to Hohfeld's schema, the correlate of a privilege is not a duty or a liability but a no-right.Footnote 44 If legitimate governments or states have the privilege of enacting and enforcing laws, as Copp says, then it seems to follow that everyone else has no right to interfere with their doing so, but no duty to obey these putative laws, and no liabilities to bear as a result of them. Moreover, privileges—or permissions or liberty-rights—need not be exclusive. This means that an account of how a state or government can acquire an exclusive privilege to enact and enforce laws, thereby ruling out competition from others who might also have the privilege of legislation and enforcement, is necessary. Without such an account, it will be impossible to make sense of the idea of a sphere of privilege. And even with such an account, the question remains, Why is this a privilege rather than a power or a claim-right?
Copp does invoke Hohfeldian power in the second element, of course, but his explanation of this power raises further problems. Like other revisionists, Copp turns from the authority-as-claim-right understanding to authority-as-power in part because of concerns about the possibility of providing an adequate defense of the general obligation to obey that the traditional account entails. Unlike the others, however, Copp invokes the distinction between obligation and duty that I mentioned in Section I of this paper. The difference, he says, is that an “obligation is owed to some agent, and it corresponds in a precise way to a right possessed by that agent,” whereas other “moral requirements, including duties, are not owed to any agent and do not correspond in this way to rights.”Footnote 45 He subsequently concludes, “If an obligation to obey the law would be sufficient for the legitimacy of a state, then surely it would be sufficient as well if people had a duty to obey the law, even if they did not owe their obedience to the state.”Footnote 46
This conclusion is troublesome in at least three ways, two of which may be set aside here. One is that Copp needs to provide a defense of the distinction between obligations and duties that will persuade those who regard the terms as virtual synonyms to see the error of their ways. The second is that he will need to explain how legitimate authority, understood as power, entails a duty, albeit a pro tanto duty, on the part of those subject to the authority when the proper Hohfeldian correlate is a liability. Again, I am prepared to grant Copp this point, especially in light of my criticism of Applbaum's supposedly duty-free account of legitimate power; but other revisionists should demand more of him.Footnote 47 This leaves the third and most important problem, however, which is that Copp's approach does not provide a genuine replacement for the standard account of political authority and obligation.
This problem itself has two aspects. The first concerns Copp's claim that a duty to obey the law should be sufficient to warrant the legitimacy of the state even if the subjects did not owe their obedience to the state. This claim raises questions about the nature of a political obligation or duty: Is it something owed to a state, a government, or one's fellow citizens? I will take up this point in Section III of this paper but set it aside for now.
The other troublesome aspect is Copp's assertion that a duty will be sufficient just as an obligation would be, despite the distinction he draws between them, to sustain a claim to legitimacy. The problem here is that Copp shifts in the course of his discussion from a duty to obey the law, as he says in the quotations at the top of this page, to the pro tanto duty to do something simply because the legitimate state has enacted a law, as he says in the second element of his Hohfeldian cluster. He cannot, however, have it both ways. If the duty in question is the former—that is, a duty to obey the law as such—then it is indeed likely to be as sufficient a ground for legitimacy as an obligation to obey. But it will also be open to the objections brought against the claim that there is a general obligation to obey the law in a reasonably just, or legitimate, polity. Taking this path, then, gives Copp's authority-as-power approach no advantage over the standard account of political authority and obligation. If the duty in question is to be conceived in the latter sense, however, it is not a single duty but a series of duties—a duty to obey this law, and that law, and this other law—that is supposed to be, according to the revisionist power-entails-liability account, markedly different from the obligation to obey the law as such. But if the latter is the proper reading of Copp's claim, it is no longer a plausible substitute for an authority-grounding general obligation to obey the law. Whichever way he turns, Copp will have to provide further argument to support his claim about a duty or duties to obey providing a sufficient basis for legitimate authority. But he will have to turn one way or the other and develop his arguments accordingly.
In the end, Copp's treatment of the legitimacy of a state as a bundle or cluster of Hohfeldian advantages may well prove valuable. Its value, though, will not derive from a demonstration of its superiority to the traditional understanding of the relationship between political authority and obligation. In this respect, Copp is no more successful than than the other revisionists in this category have been. It is possible, of course, that some other attempt to supplant the traditional conception of political authority as the (claim-)right to rule with one that conceives it as the power(-right) to do so may prove compelling. The three sophisticated attempts by Perry, Applbaum, and Copp, however, are not themselves compelling, nor do they warrant the conclusion that the traditional account is itself untenable.
C. Authority as Service
Although he does not draw explicitly on Hohfeld in this regard, Joseph Raz also conceives of authority as power. As he says in The Morality of Freedom:
The obligation to obey a person which is commonly regarded as entailed by the assertion that he has legitimate authority is nothing but the imputation to him of a power to bind. For the obligation to obey is an obligation to obey if and when the authority commands, and this is the same as a power or capacity in the authority to issue valid or binding directives.Footnote 48
In this regard, Raz could easily be included in the second category of revisionists. His criticism of the traditional account has much less to do with powers and liabilities, however, than with his service conception of authority, and that is reason enough to give him and those who follow his service conception a category of their own.
As the term suggests, the core of Raz's conception of authority is the belief that authority exists not for its own sake but to serve those subject to it. To those who ask, why should I consider myself subject to some other person's authority?, Raz's answer is that a genuine authority is more likely than you are “to act correctly for the right reasons.”Footnote 49 In this way, as Scott Shapiro points out, Raz has a response to the two autonomy-based challenges anarchists raise against those who believe in the existence of practical authority: Why should I surrender my autonomy, in the sense of thinking and acting for myself, to someone else? And why should I violate the moral duty of autonomy by allowing someone else to think and act for me?Footnote 50 The answer, in short, is that you are not really surrendering your autonomy in these cases—not, that is, so long as the authority you are following is a genuine authority that is more likely “to act correctly for the right reasons” than you are likely to do on your own. Someone can have authority over another, then, “only if there are sufficient reasons for the latter to be subject to duties at the say-so of the former.”Footnote 51
Raz's explication of the service conception of authority in The Morality of Freedom proceeds by way of three theses, the first of which—the preemptive thesis, familiar from Section I of this paper—rests on the other two. The second is the dependence thesis, which states: “all authoritative directives should be based on reasons which already independently apply to the subjects of the directives and are relevant to their action in the circumstances covered by the directive.”Footnote 52 In subsequent work, Raz replaces this thesis with “the independence condition,” but he also indicates that it is the remaining thesis that “provides the key to the justification of authority,” so I shall follow common practice by concentrating on this third thesis.Footnote 53 According to this normal justification thesis,
the normal way to establish that a person has authority over another person involves showing that the alleged subject is likely better to comply with reasons which apply to him (other than the alleged authoritative directives) if he accepts the directives of the alleged authority as authoritatively binding and tries to follow them, rather than by trying to follow the reasons which apply to him directly. Footnote 54
When applied to specifically political authority, the service conception leads to a highly flexible and discriminating understanding of the scope of authority. “It all depends,” as Raz says, “on the person over whom authority is supposed to be exercised: his knowledge, strength of will, his reliability in various aspects of life, and on the government in question.”Footnote 55 Thus, the expertise of those who work for our government's drug regulation agency will place most of us under its authority, but not those who themselves are expert pharmacologists; the expert pharmacologists will be under the government's authority with regard to the roadworthiness of their cars, but expert mechanics will not; and so on. The result is “a very discriminating approach to the question” of how far the authority of the government may extend: “[t]he government may have only some of the authority it claims, it may have more authority over one person than another.”Footnote 56 Instead of providing an account of unified political authority, “the normal justification thesis invites a piecemeal approach to the question of the authority of governments, which yields the conclusion that the extent of governmental authority varies from individual to individual and is more limited than the authority governments claim for themselves in the case of most people.”Footnote 57
Raz's normal justification thesis, with its “piecemeal approach,” thus has significant implications for the relationship of political authority to political obligation. Like the other revisionists, Raz denies that even the citizens of a reasonably just society have a general obligation to obey the law. Unlike the others, though, Raz does not derive this conclusion from the denial that authority entails obligation but from the limited extent of political authority understood in light of the service conception. Even a “qualified recognition of authority” as “the authority of just governments to impose prima facie obligations on their subjects cannot be supported by the argument of the normal justification thesis.”Footnote 58 To be sure, Raz does acknowledge that there is “probably a common core of cases regarding which the obligation exists and applies equally to all,” such as the duty to pay taxes and other duties that follow from the coordinative functions of government.Footnote 59 Otherwise, the piecemeal effect of the normal justification thesis blocks the possibility of even a pro tanto obligation on the part of every citizen to obey every law in even a reasonably just polity.
Raz's service conception has proven to be both highly influential and highly controversial, with critics claiming that it captures some important aspects of authority but not others. Three criticisms are of particular significance here. The first is that the service conception scants the official—that is, office-holding, or institutional—aspect of political authority. As Jeremy Waldron says, taking the normal justification thesis “as a sufficient condition of A's having authority over C … would imply that millions of people have authority over each one of us.” It might even follow that the U.S. Conference of Catholic Bishops, which “produced several powerful and illuminating statements” on welfare reform in the 1980s and 1990s, held authority over the U.S. Congress, which made “disastrously unjust decisions about welfare reform” in that period.Footnote 60 Such a conclusion might be acceptable if we were talking only of moral authority, or if we were to say that the Catholic bishops ought to have authority over Congress. No doubt there are many who would make such judgments today with regard to the Intergovernmental Panel on Climate Change (IPCC) and the U.S. Congress. But that is quite far from holding that the Conference of Catholic Bishops or that the IPCC really does have authority over Congress.
The second criticism concerns the piecemeal nature of political authority as the service conception defines it. As we have seen, Raz takes this to be a strength of his conception of political authority, but others regard it as a serious weakness. One critic, Thomas Christiano, argues that “the instrumentalist and piecemeal nature of authority on this account allows it to attribute legitimate authority to ferociously unjust regimes.”Footnote 61 Another, Christopher Bennett, objects that “the most basic idea of a legitimate authority is that of a governing body whose subjects have a duty to obey it by virtue of its position rather than because of the piecemeal helpfulness of following its dictates.”Footnote 62 To put the point a bit differently, those who hold political authority are supposed to exercise authority over the members of a polity qua members, not in their several capacities as pharmacologists, automobile mechanics, computer programmers, plumbers, and so on. To appreciate how the service conception deviates from this systemic understanding of political authority, one need only envision Citizen A explaining to Citizen B why B has a duty to obey a law that A is free to ignore.
The third criticism is that the service conception fails to take disagreement and democracy seriously. If we follow the normal justification thesis, the purpose of political authority is not so much to reconcile those who disagree, or to bring order out of conflicting opinions, as it is to find the right answers to questions of practical reason. On this view, Christiano objects, disagreement may simply be taken to indicate “that people's false political views have no direct relation to what they or we have in fact reason to do.”Footnote 63 Nor is there room for the view, as Scott Shapiro says, that “democratic procedures are capable of possessing legitimate authority because they represent power-sharing arrangements that are fair.”Footnote 64 If democratic procedures have any value at all in the service conception of authority, it is because they serve practical reason by helping to elucidate the reasons that rightly apply to people who must live together. But there are reasons for thinking that there are features of democracy—features relating to equality, autonomy, and fair play—that give it not only the instrumental value that the service conception recognizes but also endow it with what Christiano calls “inherent authority.”Footnote 65 This, however, is a dimension of authority that the service conception fails to capture.
None of this is to say that Raz has completely misconstrued authority, political or otherwise. On the contrary, the service conception probably does more to clarify the relationship of authority to practical reason than any other conception of authority has done. It also goes far toward explaining why possession of some degree of de facto authority is necessary to the possession of de jure authority. But the service conception has serious shortcomings too, as the preceding criticisms attest. Chief among them is its inability to provide a satisfactory account of the relationship between political authority and the polity itself. Because that account is unsatisfactory, there is also reason to suspect Raz's dismissive conclusions about the possibility of a general obligation to obey the law. Like the other revisionists, in short, Raz has not made his case for the separation of political authority from political obligation. I conclude, then, that it is not yet time to give up on the traditional account of their relationship.
III. DEFENDING THE TRADITIONAL ACCOUNT VIA FAIR PLAY
Defects in the revisionists’ arguments, of course, do not immediately translate into virtues of the traditional account. On their own, in fact, the criticisms raised in the preceding section might serve only to encourage the anarchists, philosophical and political, in their wholesale rejection of political authority and obligation. Mounting a proper defense of the traditional account thus requires that something positive be added to the criticisms of the various revisionist arguments; and that something positive must be a demonstration of how consent, natural duty, membership, or some other theory of political obligation can vindicate the traditional account by linking such obligations to political authority. In this section, I sketch such a demonstration by indicating how the fair-play theory of political obligation establishes a satisfactory foundation for the traditional account and, in doing so, upholds what Perry calls the reverse-entailment thesis.Footnote 66
According to the principle of fairness, or fair play, anyone who takes part in a cooperative practice and accepts the benefits it provides is obligated to bear a fair share of the burdens of the practice. In H. L. A. Hart's canonical formulation, “when a number of persons conduct any joint enterprise according to rules and thus restrict their liberty, those who have submitted to these restrictions when required have a right to a similar submission from those who have benefited by their submission.”Footnote 67 For this principle to bear on political obligation and authority, it will be necessary to conceive of a body politic as a “joint enterprise according to rules”—that is, a cooperative practice—and that is a standard many states have not achieved. As the term “fair play” itself suggests, only polities that give something approaching equal consideration to all of their members under the rule of law can be considered cooperative practices. In such polities, according to the principle, there is a general obligation to obey the law—that is, an obligation to obey the “rules” according to which the “joint enterprise” operates.
Most commentators agree that the principle of fair play does provide the basis for a duty of fair play in some circumstances. Among them, however, are critics who raise two main objections against the attempt to extend the principle to the political domain. The first is that the principle requires voluntary acceptance of benefits, and it is unreasonable to think that most members of even a reasonably just polity have knowingly and willingly accepted the benefits it provides, especially when those benefits are public goods that one will receive whether one asks for them or not.Footnote 68 The second objection is that the duty of fair play applies only to small groups, because only in such groups can a social practice be truly cooperative. In the gigantic states of the modern world, it strains credulity to think that people will perceive one another as fellow members of a cooperative practice.Footnote 69
Proponents of the fair-play theory of political obligation have generally responded in one of two ways. The first line of response is to try to show that acceptance of the polity's benefits is more widespread than the critics acknowledge, and particularly so if one allows that voluntary acceptance need not be as deliberate as the critics insist.Footnote 70 The second response is to argue that receipt of benefits in the proper circumstances is sufficient to generate fair-play obligations, even when there is no knowing or willing acceptance of those benefits. Thus, George Klosko has pointed to “presumptive benefits,” such as national defense, as grounds for believing that the citizens of large-scale polities have an obligation or duty of fairness that entails an obligation to obey the law.Footnote 71 In the remainder of this section, my sketch presupposes that either or both lines of response are successful.
For present purposes, the leading feature of this sketch is that fair-play theory adheres to the traditional understanding of the relationship between political authority and obligation. A reasonably just polity is a cooperative practice that centers on the rule of law, understood as a public good that is not only presumptively beneficial, as Klosko says, but also generally accepted in various ways in the daily lives of the polity's members. For the rule of law to take form and to persist, however, authority is necessary. Those who hold and exercise this authority have the right to rule, subject to the limitations of the rule of law and considerations of fair play, with this right understood as a claim-right entailing a corresponding obligation of obedience on the part of those who are subject to their authority.Footnote 72 This obligation, in turn, is a defeasible but general obligation to obey the law as such.
In fair-play theory, as in the standard account of political authority and obligation, this obligation is preemptive, content-independent, and of morally binding force. How strong that force is will vary from one law to another, and we may occasionally need to examine the content of individual laws to determine whether there are moral reasons that justify or even require disobedience rather than obedience. That is why the obligation to obey particular laws is prima facie or pro tanto. But political obligation is the obligation to obey the law as such of a reasonably just polity, and that obligation is grounded in a single principle: the principle of fair play. This is an obligation, furthermore, that is owed not to the state or government or the authorities but to the cooperating members of the polity. Nor is it a free-floating duty—that is, a moral requirement “not owed to any agent” that does not “correspond … to rights”—of the kind that Copp distinguishes from an obligation.Footnote 73 Whether we speak of an obligation or a duty to these other entities, it is only as a shorthand way of acknowledging that offices, institutions, and authority are necessary to sustain a cooperative practice under the rule of law.
Sustaining the polity, finally, requires securing it from the threats of those who would take its benefits while shirking the burden of obeying the law when they would rather not. That is why the polity accords some of its members the authority to detect, arrest, and punish those who do not respect the persons, property, and promises of the other members of the cooperative practice. Insofar as the offender enjoys the benefits of the cooperative enterprise without fully contributing to their provision by bearing her share of the burdens of obeying the law, the offender in effect justifies the cooperative, burden-bearing members in punishing her. In doing so, the polity communicates its censure to the offender by means of punishment and aims at the maintenance of cooperative fair play. How severely to punish, or whether a mere remonstrance will be sufficient, is one of several judgments that will have to vary with the severity of the crime and other considerations. Still, the authority to punish rests on the polity's commitment to fair play under the rule of law.
The foregoing remarks constitute a sketch, of course, and not a fully developed statement of how the principle of fair play justifies, in some circumstances, both a general obligation to obey the law and the punishment of those who fail to meet this obligation.Footnote 74 For present purposes, though, the sketch should suffice to show that a plausible theory of political obligation in accordance with the standard account of political authority and obligation is available. Assuming that it does suffice, we now have two reasons to resist those who believe the standard account is in need of thorough revision; first, the defects of the revisionist arguments noted in Section II of this paper, and second, the strength of the standard account itself.
IV. FAIR PLAY, ARBITRATION, AND AUTHORITY
In a long and valuable essay on authority, Scott Shapiro distinguishes between two ways in which authorities can serve their subjects. One is by mediation between reasons and persons—that is, “by enabling subjects to achieve benefits that they would not have been able to achieve without the [authorities’] directives.” The other is by arbitration between rival parties—that is, serving subjects “by providing them with a way to resolve their disputes on normative matters.”Footnote 75 Raz's service conception falls into the former category, but Shapiro concludes that the arbitration model is the more satisfactory of the two. He does so, moreover, for reasons that support the traditional account of political authority and obligation. Indeed, he draws the following summary contrast between the two models: “[i]n the Mediation Model, obedience itself is instrumentally valuable. In the Arbitration Model, the parties do not benefit through their obedience. Obedience, rather, is the moral price that parties must pay in order to secure the compliance of others.”Footnote 76
As this contrast suggests, and as he explicitly acknowledges on the same page, Shapiro takes fair-play theory to fall under the arbitration model. Within the limits of his contrast, I believe he is right to do so. I also think that the connection he goes on to draw between the arbitration model and democracy is correct, insofar as “deference to democratically elected authority under conditions of meaningful freedom is deference to a power-sharing arrangement that is socially necessary, empowering, and fair.”Footnote 77 I quarrel, though, with two features of Shapiro's assessment: first, that arbitration must be understood as arbitration between “rival parties”; and second, that “the parties do not benefit through their obedience.”
According to fair-play theory, the parties engaged in the cooperative practice are rivals on occasion, to be sure. If they were not, they would have no need for laws and authorities to settle their disputes. But their rivalry occurs within the framework of a cooperative practice in which there will be reason to see one another as fellow participants. Moreover, the parties do benefit through their obedience, even if not in the narrow sense of benefiting through each and every act of obedience. They will no doubt see themselves as paying a price from time to time, as Shapiro says. But they should also understand, upon reflection, that if they want to receive the benefits of the cooperative practice, they will have to undertake the duty to bear a fair share of its burdens. Suitably modified, however, Shapiro's arbitration model is a convenient way to conceive of the traditional account of political authority and obligation that is also more than hospitable to the argument from fair play.
Exactly what form this suitable modification should take is not clear. One possibility would be to add a third model to Shapiro's pair, so that Arbitration and Mediation would be joined by, say, the Collaboration Model of authority. Or perhaps the Arbitration Model should be reconceived, and renamed, to encompass not only dispute-settling among rivals but also cooperation among fellow members of a political or legal system. Either way the appropriate name seems to be the Collaboration Model, or perhaps the Cooperation Model, to reflect the need for established authority in settling coordination and collective-action problems. Whether it is part of a dichotomy or trichotomy, however, and regardless of the name it bears, the important point is that this is a model of authority that preserves the standard account of political authority and obligation.
