I. INTRODUCTION
Philosophy, despite its typical attitude of detachment and abstraction, has for most of its long history been engaged with the practical and mundane-seeming question of whether there is a duty to obey the law. As Matthew Kramer has recently summarized: “For centuries, political and legal theorists have pondered whether each person is under a general obligation of obedience to the legal norms of the society wherein he or she lives. The obligation at issue in those theorists' discussions is usually taken to be prima-facie, comprehensively applicable, universally borne, and content-independent.”1
Matthew H. Kramer, Moral and Legal Obligation, in BLACKWELL GUIDE TO THE PHILOSOPHY OF LAW AND LEGAL THEORY 179 (Martin P. Golding and William A. Edmundson, eds., forthcoming).
This essay is a commentary on the current state of discussion of this perennial philosophical topic.
Each of the elements in Kramer's nice formulation deserves a short explanation. The venerable expression “prima facie duty” is carried over from the work of David Ross,2
W.D. ROSS, THE RIGHT AND THE GOOD (1930).
but I will use instead the less misleading and recently preferred term “pro tanto duty” to emphasize that the duty in contention is not absolute but subject to being defeated or outweighed by countervailing moral considerations.
3 SHELLY KAGAN, THE LIMITS OF MORALITY 17 (1989).
As such, the duty is not, however, regarded merely as one morally relevant consideration among others but as one that is ordinarily decisive and as one toward which deference is worth cultivating. Moreover, violations of the duty to obey the law are regarded as subject to justifiable censure and punishment. In what follows, the qualifier “pro tanto” should be assumed to attach wherever the term “duty” or “obligation” occurs.
The duty is “comprehensively applicable” in the sense that it attaches to all of a jurisdiction's mandatory laws, though of course not to those purporting merely to create legal powers or permissions and perhaps not to those carrying nonstigmatizing sanctions, such as modest fines. With respect to Meir Dan-Cohen's distinction4
Meir Dan-Cohen, Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law, 97 HARV. L. REV. 625–677 (1984).
between conduct rules addressed to citizens and decision rules addressed only to officials, the duty in question here pertains preeminently to conduct rules.
The duty is “universally borne” in the sense that it purports to apply to each and every one of those to whom the jurisdiction's mandatory laws are directed and who would be exposed to the possibility of sanction for noncompliance. Disagreement with the wisdom or morality of a given mandatory law would not, in other words, exclude an actor from the reach of a universally borne duty to obey.
The duty is “content-independent” in the sense that the existence of the duty is not a direct function of the moral merit of the particular law in question. This aspect is sometimes expressed by saying that one has a duty to obey the law qua law, regardless of whether there are independent moral or other reasons to do as the law mandates—unless, perhaps, those reasons are extraordinarily powerful or emanate from a source not in the lawmaker's contemplation. A content-independent duty effectively preempts the subject's individual assessment of the merits of the action required by law and is categorical in the sense that it is not contingent upon any motivating end or goal of the subject.5
The combined effect of the above elements is summarized in John Finnis's vivid dictum: “The law presents itself as a seamless web. Its subjects are not [morally] permitted to pick and choose.”
6 John Finnis, The Authority of Law in the Predicament of Contemporary Social Theory, 1 NOTRE DAME J.L. ETHICS & PUB. POL'Y 115–137 (1984), at 120.
Some further terminological stipulations and subject-matter
limitations are in order. The expressions “duty to obey the
law” and “obligation to obey the law” are
etymologically and perhaps idiomatically distinguishable, but current
philosophical usage tends to downplay the significance of any deeper
“conceptual” distinction between the two—even among
writers who otherwise emphasize the need to work within a voluntaristic
framework. Accordingly, the two terms, “duty” and
“obligation,” will be used more or less interchangeably
here, as equivalent both mutually and with the more cumbersome term,
“moral requirement,” which has become current in specialist
discussions.
The duty to obey has been contrasted with what might be termed a duty merely to comply with the law—the difference being that the former, unlike the latter, purports to capture the idea that the law presents itself as authoritative.7
Green, supra note 5, at 3.
Obedience, unlike compliance, is at least in part a matter of “doing what [someone] tells you to do because he tells you to do it”
8 ROBERT PAUL WOLFF, IN DEFENSE OF ANARCHISM 9 (1971).
and not of doing solely for reasons of one's own what another happens to have commanded or recommended. This idiomatic point taken, discussion of a
general duty to comply is in most instances naturally enough understood as pertaining to obedience and authority in the stringent sense, needing only to be contrasted to a mere jumble of particular duties that happens to overlap what the law requires.
Recent usage has tended also to conflate the duty (or obligation) to obey the law with what is referred to as “political obligation,” but in this case the tendency, though understandable, will be resisted. Political obligation is a more compendious term that sweeps in the duty of obedience concerned here together with, for instance, a more overarching and diffuse duty to support and defend the state of which one is a citizen, a duty to give preference to compatriots over foreigners, and to duties assigned to political offices or other positions of advantage within a territory.9
DAVID MILLER, ON NATIONALITY 68, 71 (1995).
Recent debates about whether patriotism is a virtue
10 See, e.g., Alasdair MacIntyre, Is Patriotism a Virtue? The Lindley Lecture, University of Kansas (1984) in 3 LIBERALISM 246–263 (Richard J. Arneson, ed., 1992); MILLER SUPRA note 9; MARTHA C. NUSSBAUM, FOR LOVE OF COUNTRY? (2002).
and whether relations among compatriots are of special moral significance
11 Samuel Scheffler, Families, Nations, and Strangers, the Lindley Lecture, Department of Philosophy, University of Kansas (1995) in SAMUEL SCHEFFLER, BOUNDARIES AND ALLEGIANCES (2001); Diane Jeske, Associative Obligations, Voluntarism, and Equality, 77 PAC. PHIL. Q. 289–309 (1996); RICHARD DAGGER, CIVIC VIRTUES chap. 5 (1997); Samuel Scheffler, Relationships and Responsibilities, 26 PHIL. & PUB. AFF. 189–209 (1997); Andrew Mason, Special Obligations to Compatriots, 107 ETHICS 429–437 (1997); Christopher H. Wellman, Relational Facts in Liberal Theory: Is There Magic in the Pronoun ‘My’? 110 ETHICS 537–562 (2000); Diane Jeske, Special Relationships and the Problem of Political Obligations 27 SOC. THEORY & PRAC. 19–40 (2001); David Copp, Social Unity and the Identity of Persons, 10 J. POL. PHIL. 365–391 (2002); Pauline Kleingeld & Eric Brown, Cosmopolitanism, in STANFORD ENCYCLOPEDIA OF PHILOSOPHY (Edward N. Zalta, ed., Fall 2002), available at http://plato.stanford.edu/archives/fall2002/entries/cosmopolitanism/.
will therefore be explored here only to the extent necessary to clarify the state of discussion with regard to the duty to obey the law qua law. Accordingly, the term “political obligation” as used here should be understood to refer preeminently to the duty to obey.
II. LOCATING THE QUESTION IN THE CONTEXT OF POLITICAL THEORY: AUTHORITY AND LEGITIMACY
Efforts intended to support or to challenge the claim that there is a duty to obey the law often go hand in hand with efforts to expose the conceptual connections between political authority and a duty of obedience. The following set of propositions may help to exhibit the relationship between these two lines of inquiry:
- Authority. X is an authority only if X at least implicitly makes claims distinctive of authority.
- Warranty. An authority X is a legitimate authority only if X's distinctively authoritative claims are true.
- Political authority. Political authority is a species of
authority whose distinctive claim is that persons subject to it have a
general moral duty to obey its commands.
The above three propositions are of interest because of their bearing
upon the following:
- Inseparability. Political authority is legitimate only if it imposes a general moral duty of obedience on those subject to it. (from 2 and 3)
- Denial of duty. There is no general duty to obey political authority X, even if X is (nearly) just.
- State legitimacy. Legitimate political authorities are possible and even actual.12
William A. Edmundson, Introduction, in THE DUTY TO OBEY THE LAW: SELECTED PHILOSOPHICAL READINGS (William A. Edmundson, ed., 1999b).
The members of the subset {4, 5, 6} cannot be true together: logically one or more has to be rejected as false. The question: Is there a duty to obey the law? seems particularly urgent insofar as a No answer calls into question the very possibility of a legitimate state. If there is no general duty to obey the law (as 5 states) then either legitimacy does not depend upon the existence of such a duty, or there are no legitimate states, actual or possible. The legitimacy of the state is a “high-stakes” issue because it is conceptually linked in turn to the moral permissibility of its administering punishment and monopolizing the use and threat of force.
In recent writings, so many philosophers have aligned themselves with proposition 5—the denial of a duty to obey—that it has been referred to, perhaps with dramatic intention, as stating the currently fashionable view.13
See, e.g., Phillip Soper, Legal Theory and the Claim of Authority, 18 PHIL. & PUB. AFF. 209–237 (1989), at 211; Leslie Green, Who Believes in Political Obligation? in FOR AND AGAINST THE STATE, 1–17 (John T. Sanders and Jan Narveson, eds., 1996), at 28.
A number of those who accept proposition 5, the denial of duty, have gone on to deny proposition 6, state legitimacy, thus embracing a view that has been termed
philosophical anarchism.14 Wolff, supra note 8; A. John Simmons, The Anarchist Position: A Reply to Klosko and Senor, 16 PHIL. & PUB. AFF. 269–279 (1987); Simmons, Philosophical Anarchism, in FOR AND AGAINST THE STATE, 19–39 (John T. Sanders and Jan Narveson, eds., 1996a).
For an “a priori” philosophical anarchist legitimate authority is a conceptual impossibility, while for an “a posteriori” philosophical anarchist it is not: properly given universal consent could indeed create a universally borne duty. But a posteriori philosophical anarchism sees universal “actual” consent as a deep practical impossibility, given the size and scope of modern states and the irreducible plurality of views and values among their populations. Whether a priori or a posteriori, philosophical anarchism entails no commitment to the tradition of
political anarchism associated with Godwin, Proudhon, St. Simon, and Marx and it is consistent with an attitude of watchful acquiescence to the demands of law—as long as they are independently justified (as where a legal prohibition happens to coincide with an independently justified moral one, such as the prohibition of battery) or not worth resisting.
15 Simmons, The Anarchist Position, supra note 14.
Even so, philosophical anarchism has “teeth” insofar as it warrants ignoring laws that impose “distinctively political requirements,” such as “certain taxes…military service [and] many paternalistic and moralistic laws.”
16 Simmons, Philosophical Anarchism, supra note 14, at 29.
Other philosophers have repudiated philosophical anarchism and have tried to defend state legitimacy (i.e., proposition 6) by building a case against the denial of duty (i.e., proposition 5). These efforts to rehabilitate the duty to obey the law have variously drawn upon a principle of fair play,17
Richard Arneson, The Principle of Fairness and Free-Rider Problems, 92 ETHICS 624–626(1982); DAGGER, supra note 11; George Klosko, The Principle of Fairness and Political Obligation, 97 ETHICS 353–362 (1987a); Klosko, Presumptive Benefit, Fairness, and Political Obligation, 16 PHIL. & PUB. AFF. 241–259 (1987b); KLOSKO, THE PRINCIPLE OF FAIRNESS AND POLITICAL OBLIGATION (1992).
or natural duty,
18 JOHN RAWLS, A THEORY OF JUSTICE (1971); Jeremy Waldron, Special Ties and Natural Duties, 22 PHIL. & PUB. AFF. 3–30 (1993); Nancy J. Hirschmann, Freedom, Recognition, and Obligation: A Feminist Approach to Political Theory, 83 AM. POL. SCI. REV. 1227–1244 (1989); Hirschmann, Rethinking Obligation for Feminism, in REVISIONING THE POLITICAL: FEMINIST RECONSTRUCTIONS OF TRADITIONAL CONCEPTS IN WESTERN POLITICAL THEORY (Nancy J. Hirschmann and Christine Di Stefano, eds., 1996).
or consent,
19 HARRY BERAN, THE CONSENT THEORY OF POLITICAL OBLIGATION (1987); Mark C. Murphy, Surrender of Judgment and the Consent Theory of Political Authority, 16 LAW & PHIL. 115–143 (1997a).
or associative obligation,
20 RONALD DWORKIN, LAW'S EMPIRE (1986); Michael Hardimon, Role Obligations, 91 J. PHIL. 333–363 (1994); Scheffler, Families, Nations, and Strangers, supra note 11; Scheffler, Relationships and Responsibilities, supra note 11; Mason, supra note 11.
or beneficence.
21 Christopher H. Wellman, Liberalism, Samaritanism, and Political Legitimacy, 25 PHIL. & PUB. AFF. 211–237 (1996).
Mixed theories have also come increasingly into play that defend the duty to obey the law by appeal to multiple principles, such as duties of beneficence and fairness.
22 Wellman, supra note 21; George Klosko, Multiple Principles of Political Obligation, POL. THEORY (Forthcoming).
A refutation of the denial of duty would enable one to maintain that there is a strong link between legitimate authority and a duty to obey—as proposition 4 states, and which Durning23
Patrick Durning, Political Legitimacy and the Duty to Obey the Law, 33 CAN. J. PHIL. 373–390 (2003).
refers to as “
the inseparability thesis” (a.k.a. the “strong legitimacy thesis” in my
Three Anarchical Fallacies)
24 WILLIAM A. EDMUNDSON, THREE ANARCHICAL FALLACIES: AN ESSAY ON POLITICAL AUTHORITY (1998).
—without having to give up on the possibility of a legitimate state. But the rehabilitative work on behalf of the duty to obey the law has been subjected to sustained criticism,
25 See, e.g., Green, supra note 13; A. JOHN SIMMONS, JUSTIFICATION AND LEGITIMACY: ESSAYS ON RIGHTS AND OBLIGATIONS (2001a).
and to that extent the affirmation of state legitimacy (i.e., proposition 6) is perhaps more precarious than ever. Others have acquiesced in the denial of duty but save state legitimacy by denying the inseparability thesis.
26 M.B.E. Smith, Is There a Prima Facie Obligation to Obey the Law? 82 YALE L.J. 950–976 (1973); Rolf Sartorius, Political Authority and Political Obligation, 67 VA. L. REV. 3–17 (1981); KENT GREENAWALT, CONFLICTS OF LAW AND MORALITY (1987); Edmundson, supra note 24; David Copp, The Idea of a Legitimate State 28 PHIL. & PUB. AFF. 3–45 (1999).
For these philosophers the absence of a general duty to obey does not undermine the legitimacy of the state because there is no necessary correlation—at least, none as strong as what the inseparability thesis states—between legitimate authority and a general duty to obey.
Denying the inseparability thesis seems appealing chiefly because such a move would straight away reconcile the denial of duty with state legitimacy.27
PHILLIP SOPER, THE ETHICS OF DEFERENCE (2002); David A. Lefkowitz, Legitimate Political Authority and the Duty of Those Subject to It: A Critique of Edmundson, 23 LAW & PHIL. 399–435 (2004).
But the inseparability thesis expresses a deeply held view of the nature of political authority.
28 Joseph Raz, Authority and Justification 14 PHIL. & PUB. AFF. 3–29 (1985); DWORKIN, supra note 20, at 191–192; Soper, supra note 13.
Moreover, if inseparability is denied, then either the warranty thesis (i.e., proposition 2) or the political authority thesis (i.e., proposition 3) must be denied also, because the conjunction of the warranty thesis and the political authority thesis entails the inseparability thesis. Joseph Raz, for example, affirms the authority thesis (i.e., proposition 1) and the political authority thesis (i.e., proposition 3) but denies the inseparability thesis and thus, by implication, denies proposition 2, the warranty thesis. Raz argues that the legitimacy of state authority is not as closely tied to the imposition of a duty to obey as the inseparability thesis states, but he does not explain his implicit denial of the warranty thesis, which is unsatisfying given Raz's general claim that practical authority (of which political authority is a species) and scientific authority are structurally similar, and the fact that the warranty thesis seems plausible in the case of scientific authority.
29 EDMUNDSON, supra note 24.
Raz defends a “normal justification” thesis—that the standard way to justify authority is by showing that actors are likelier to act on reasons that apply to them by following the authority's directive than by acting on their own—in a way that appears to be consistent with his implicit denial of the warranty thesis.30
JOSEPH RAZ, THE MORALITY OF FREEDOM (1986), at 53–57.
I have elsewhere argued that the warranty thesis is false in the cases of both scientific and practical authority and that proposition 4—the inseparability thesis—should therefore be denied as well.
31 EDMUNDSON, supra note 24.
But in place of inseparability, some account has to be offered of the connection between state legitimacy and the duties of citizens, and such an account should establish rather than simply assume that legitimacy is a global, all-or-nothing characteristic of legal systems rather than a family of characteristics, perhaps possessed in degrees.
32 Durning, supra note 23; cf. Simmons 2001a, 130.
And as Murphy points out,
33 Mark C. Murphy, Moral Legitimacy and Political Obligation, 99 APA NEWSL. ON PHIL. & L. 77–80 (1999).
denying the inseparability thesis is unsatisfactory in the absence of an alternative conception of legitimacy that is proof against the arguments that have driven so many to embrace the denial of duty. I have argued that a legal system's legitimacy depends upon its subjects having a duty not to interfere with the administrative prerogatives of the state, that is, with the relatively direct orders by which disputes are channeled into the legal system and there resolved.
34 EDMUNDSON, supra note 24.
This duty of noninterference entails a duty of obedience only to a rather narrow range of specific, focused directives—“move along,” “you are hereby summoned,” “you are ordered to pay,” and so on—and none at all to the law
proprio vigore. I have further argued that the traits of comprehensive applicability and content independence are less objectionably attributed to this duty to obey administrative prerogatives than to the duty to obey the law as traditionally conceived.
35 EDMUNDSON, supra note 24; but see Lefkowitz, supra note 27, at 415–427, for an argument to the contrary.
Another issue arises once the authority thesis, which identifies an authority as an entity that makes distinctive claims, is disjoined from the warranty thesis, which conditions the legitimacy of an authority upon the truth of those distinctive claims. I have denied the warranty thesis and proposed instead the “proximity thesis” that the legitimacy of an authority be understood in terms of its sincerity and the approximate truth of its distinctive claims.36
EDMUNDSON, supra note 24, at 44–70; Raz's normal justification thesis can, I believe, be construed as a specification of the proximity thesis, but cf. Lefkowitz, supra note 27, at 406–412.
Such a move avoids the inseparability thesis but in so doing raises the following questions about political authority: How can an authority sincerely claim to possess a moral power that it has insufficient reason to believe it in fact possesses? And how can such a claim be even approximately true, given the widespread acceptance of the denial of duty by those who have pondered it most carefully? I have attempted an answer in terms of the law's ability to “seed” norm-conforming behavior in a way that satisfies the compliance conditions that must be satisfied before fair-play duties come into existence.
37 William A. Edmundson, Social Meaning, Compliance Conditions, and Law's Claim to Authority, 15 CAN. J. L. & JURIS. 51–67 (2002).
This answer may, however, be irreconcilable with strong formulations of a principle of publicity that would link state legitimacy to the public transparency of its rationale.
38 Cf. Larry A. Alexander, Pursuing the Good—Indirectly 95 ETHICS 315–332 (1985); David Luban, The Publicity Principle, in THE THEORY OF INSTITUTIONAL DESIGN 154–198 (1996); Lefkowitz, supra note 27, at 427–434.
The elusiveness of any stable position within this narrow-seeming conceptual domain is illustrated by one author's desultory path over the last two decades. At one time, Philip Soper affirmed the political authority thesis and the inseparability thesis and defended the warranty thesis in the case of scientific, though not political, authority.39
Thus Soper's commitment to the inseparability thesis was not compelled by adherence to the warranty thesis and the political authority thesis, for he rejected the warranty thesis in its generality because, in his view, it failed to apply to practical authority. Rather, Soper embraced inseparability because it captures the views of “insiders,” whose common convictions and practices must not be rejected lightly. Since insiders (and others) are at least as wedded to state legitimacy as to inseparability, Soper's conclusion was that political philosophy has to work harder at refuting the denial of duty. More recently, Soper has defended the somewhat paradoxical view that although the law does not claim practical authority, it nonetheless possesses it.
40 Thus Soper now rejects both the political authority thesis and the denial of duty, but has not repudiated his earlier adherence to the inseparability thesis and state legitimacy. Soper's position is remarkable in that he denies both the warranty and political authority theses (propositions 2 and 3)—which if jointly held entail inseparability—but maintains inseparability anyway, because insiders do, and the considered views of insiders must be taken as data for the legal theorist. Soper was surely correct that political philosophers have to work harder in opposition to the denial of duty, but with equal justice it must be said that legal philosophers have to work harder in support of the inseparability thesis. A bald appeal to common or insider opinion cannot furnish a satisfactory basis for either inseparability or the assertion of the duty to obey.
41 Thus the tension within the set {4, 5, 6} could be resolved in a number of ways (most simply by denying the most dubious of its members), but for purposes of this essay the focus will be on efforts to refute proposition 5—that is, to rehabilitate the duty to obey the law—rather than on, for example, revisionary work intended to replace the inseparability thesis. Throughout, John Simmons's warning42
against conflating the tasks of legitimating the state and of justifying it will be scrupulously observed.
43 Cf. Robert Ladenson, Legitimate Authority, 9 AM. PHIL. Q. 335–341 (1972); Ladenson, In Defense of a Hobbesian Conception of Law, 9 PHIL. & PUB. AFF. 134–159 (1980).
Legitimacy crucially involves the imposition of
duties upon citizens, whereas a justification need only furnish state actors with a moral permission to perform their characteristic offices. Owing to its focus upon duty, this discussion will be haunted by a more general difficulty in moral theory: that of closing the logical gap between moral reasons favoring types of acts and institutions and moral requirements—duties—of individuals that they perform such acts and submit to the commands of such institutions.
Some have complained that duty and obligation have been overemphasized in contemporary moral theory to the reciprocal detriment of the moral virtues,44
See, e.g., G.E.M. Anscombe, Modern Moral Philosophy, 33 PHILOSOPHY 1–19 (1958).
and there has been a revival of interest in the question of whether patriotism is a virtue
45 MacIntyre, supra note 10.
as well as in the virtues generally.
46 ROSELIND HURSTHOUSE, ON VIRTUE ETHICS (1999); but cf. John M. Doris, Persons, Situations and Virtue Ethics, 32 NOUS 504–530 (1998).
Nonetheless, I am not aware of any detailed proposal that we construe law-abidingness as a virtue rather than a duty. An aretaic account of law-abidingness might prove to be awkward because the law presents itself as a set of rules of conduct that do not generally purport to improve the individual and may not dependably have any such tendency. Holmes's “bad man's” perspective may not be the last word on how to determine the law's content but it is not obviously incompatible with a disposition of punctilious obedience to the law whatever its content may be. Moreover, obedience seems a virtue only in the attenuated sense that being disposed generally to discharge one's duties is a virtue. If there were no duty to obey some particular legal command and the performance it required were valuable on no other ground, many would be disinclined to praise supererogatory compliance as an exhibition of any virtue and might in fact criticize it as sycophantic.
III. ADEQUACY CONDITIONS AND ANTITHEORIES
The bulk of recent literature on the duty to obey the law consists of discussions intended either to advance or to undermine one or another account of or justification of the duty. Before turning to that literature, however, it will be useful to survey recent writings that stand at one remove. These writings largely prescind from the merits of, say, a hypothetical-consent theory or of arguments invoking the principle of fair play, and instead seek to make more general points. The first set of these points can be characterized as adequacy conditions. Members of this set do not take a position on particular accounts of the duty to obey but instead state some (partial) criterion of success for such accounts. The second set consists of arguments that purport to show that the nonexistence of the duty to obey can be derived from general reflections upon the nature of law or upon the nature of moral duty. The second set, in other words, consists of arguments that simultaneously set up an adequacy condition and deny that that very condition can be satisfied. The two sets can conveniently be discussed together.
It is now generally agreed that the duty to obey arises, if at all, only if a “threshold condition of justice is met”47
Green, supra note 5, at 5.
by the legal system in question. That this threshold is within easy reach is usually assumed but has rightly been challenged.
48 See, e.g., Jeffrie G. Murphy, Marxism and Retribution, 2 PHIL. & PUB. AFF. 217–243 (1973).
A number of writers have insisted that any successful defense of the duty to obey the law must be consistent with certain other entrenched moral principles.
49 RONALD DWORKIN, TAKING RIGHTS SERIOUSLY (1977), at 184–196.
Consistency with these other principles, then, is posited as a necessary but not sufficient adequacy condition that theories of the duty to obey the law must meet. One such condition is that the duty to obey the law must be consistent with liberalism and more specifically with liberal or libertarian tenets such as self-ownership and the presumption of liberty. Because persons are presumptively free, on this view, they are presumptively unencumbered with moral duties that would, as a simple matter of Hohfeldian logic, curtail their moral liberty. This presumption is most readily overcome in the cases of illegitimate force or fraud, which are harmful to others, and so general moral duties to refrain from force and fraud must be admitted. The defense of the duty to obey the law has likewise to overcome the liberty presumption. If the defense does not go well, then the matter is not left in suspense, but a conclusion may be drawn that there is no such duty.
The presumption of liberty also favors certain types of defense of the duty to obey the law, a duty that, like all others, is to be seen as a departure from a baseline of natural liberty.50
A. JOHN SIMMONS, ON THE EDGE OF ANARCHY (1993); Robert Paul Wolff, Political Obligation, Fairness, and Independence, 8 RATIO 87–99 (1995); Christopher H. Wellman, 2001a; Simon Cushing, Justification, Legitimacy and Social Embeddedness: Locke and Rawls on Society and the State, 37 J. VALUE INQUIRY 217–231 (2003).
Most desirable are consent theories, which represent the subject's moral duty to obey as a free undertaking, more or less along the lines laid down by Grotius, Hobbes, and Locke. Next in order of desirability would be an account of a voluntaristic nature, which would ground the duty to obey in some voluntary undertaking—such as political participation or residence. Less desirable, from a liberal perspective, would be an account that grounded the duty on counterfactual conditions—such as what a reasonable person would consent to or would accept or undertake. Least desirable of all are accounts that dispense altogether with voluntaristic conditions—actual or hypothetical—and ground the duty on the bare existence of unchosen relationships to others, as by analogy with filial duty, or the bare receipt of unsought benefits.
A presumptive case against the duty to obey may similarly be founded not on liberty but on competing duties, such as a duty of autonomy or of individual responsibility.51
Suppose it is my clear moral duty to do what, after careful reflection, seems to me to be the right thing to do. If that is supposed, no room seems to be left for a duty to obey, for the law either clashes with what conscience directs me to do, or it does not. If it does not, the duty to obey does no work, and if it does clash, it must yield unless it overcomes the presumption in favor of the duty to act autonomously. Whether or not I possess moral liberty, I, on this line, must presume that the law has no authority. A defense of a duty to obey must overcome this presumption, and to the extent that disputes about the duty to obey stubbornly remain, the proper response is not to suspend judgment but to proceed as though there were no such duty.
The presumption of liberty and the duty of autonomy have been invoked in this fashion to circumscribe the terms in which the debate over the duty to obey the law is to be conducted. So circumscribed, the defense of the duty to obey is saddled with the burden of persuasion, and the denial of the duty to obey enjoys a presumption of correctness. But this way of setting the terms of engagement has been challenged. The challenges have taken two forms. The first form directly counters the liberal presumption of liberty and the Wolffian duty of autonomy. The second proceeds by constructing a counterpresumption of correctness favoring the duty to obey—a form typified by Mark Murphy's appeal52
M. Murphy, 1997b; M. Murphy, Philosophical Anarchisms, Moral and Epistemological (n.d.).
to a “conscience principle” and perhaps also by recent invocations of “associative” obligations.
53 DWORKIN, supra note 20; Hardimon, supra note 20.
The presumption of liberty has been embraced by many liberal philosophers54
See, e.g., Feinberg 1973, at 22; RANDY E. BARNETT, RESTORING THE LOST CONSTITUTION: THE PRESUMPTION OF LIBERTY (2004).
but has been eschewed by Rawls and Ronald Dworkin, the most influential among recent liberal theorists,
55 JOHN RAWLS, POLITICAL LIBERALISM 291–292 (1993); DWORKIN, supra note 49, at 266–278; and see EDMUNDSON, supra note 24, at 91–93.
and has also been criticized by feminists and communitarians.
56 Hirschmann, Freedom, Recognition, and Obligation, supra note 18; Hirschmann, Rethinking Obligation for Feminism, supra note 18; cf. JOHN HORTON, POLITICAL OBLIGATION (1992).
The presumption of liberty has, however, enjoyed a second youth in the guise of the doctrine that law is coercive. What is coercive is, on many accounts, presumptively morally unjustified and thus presumptively without moral authority.
57 Hans Oberdiek, The Role of Sanctions and Coercion in Understanding Law and Legal Systems, 1975 AM. J. JURIS. 71–94 (1975).
To the extent that law is coercive, these presumptions would attach to the purported duty to obey. The prereflective view that law
is coercive rests, however, upon contestable assumptions about the proper baseline against which to measure the law's effect, and the need to defend these assumptions—however the relevant baseline is conceived—ultimately reopens the issue of the moral character of law.
58 William A. Edmundson, Is Law Coercive? 1 LEGAL THEORY 81–111 (1995); and cf. Grant Lamond, Coercion and the Nature of Law, 7 LEGAL THEORY 35–58 (2001); Mitchell Berman, The Normative Functions of Coercion Claims, 8 LEGAL THEORY 45–89 (2002).
The duty of autonomy has also been subjected to skeptical scrutiny
59 JEFFREY H. REIMAN, IN DEFENSE OF POLITICAL PHILOSOPHY (1972).
that has emphasized the difficulty of reconciling such a duty with the widespread and uncontested practice of deferring to epistemic authorities, such as scientists, public health officials, and medical experts. If there is no strong general duty to exercise autonomy in epistemic contexts, it becomes less plausible to maintain that there is a sufficiently strong duty of autonomy in moral and political contexts to establish a presumptive case against the duty to obey the law.
60 Green, supra note 5, at 5.
Nonetheless, even a weakened duty of autonomy may call into question the “content independence” said to mark the duty to obey the law. Both the coherence
61 P. Markwick, Law and Content-Independent Reasons, 20 OXFORD J. LEGAL STUD. 579–596 (2000).
and the moral palatability of content-independent rules of conduct
62 HEIDI HURD, MORAL COMBAT (1999); Michael S. Moore, Authority, Law, and Razian Reasons, 62 S. CAL. L. REV. 827–896 (1989); LARRY A. ALEXANDER & EMILY SHERWIN, THE RULE OF RULES: MORALITY, RULES, AND THE DILEMMAS OF LAW (2001).
have been challenged; but the dialectic of this dispute cannot even be outlined here. Suffice it to say that a complete account of the duty to obey the law will remain hostage to these more fundamental disputes.
The venerable debate between natural-law theorists and legal positivists has also had a bearing on the question of the existence of a duty to obey the law. If, following Aquinas, we define law as an ordinance of reason issued by one in authority to promote the common good,63
Mark C. Murphy, Natural Law Theory, in BLACKWELL GUIDE TO THE PHILOSOPHY OF LAW AND LEGAL THEORY 179 (Martin P. Golding and William A. Edmundson, eds., forthcoming).
it would seem that a duty to obey the law is demonstrable as a simple specification of a (pro tanto) duty to promote the good. Or law itself might directly be defined in terms including a duty of obedience. Derivations of this kind have been grouped under the label “the conceptual argument” for the duty to obey. Although such conceptual arguments were seriously proposed in the early 1960s, they are seldom heard today.
64 SIMMONS, supra note 25, at 72–73.
This may be due to the ascendancy of sophisticated versions of legal positivism, which maintains the “separability thesis” that there is no necessary conceptual connection between legal and moral duties.
65 RAZ, supra note 30; Jules Coleman, On the Relationship between Law and Morality, 2 RATIO JURIS 66–78 (1989).
Present-day natural lawyers avoid the “conceptual argument” in that they acknowledge the existence of valid-but-defective laws, that is, validly promulgated positive law that is contrary to what moral reason in fact requires.66
Finnis, supra note 6; M. Murphy, supra note 63; Green, supra note 5, at 5.
A modern natural lawyer thus has to provide both: (1) an account of the duty to obey valid-but-defective law; and (2) an account of the authoritativeness of nondefective law. For a legal positivist the nonexistence of a duty to obey the law might be seen as a straightforward consequence of the separability thesis,
67 Coleman, supra note 65, at 66.
but such an inference would be too quick since it would fail to have ruled out the possibility of other arguments not of a narrowly conceptual nature to establish the existence of the duty.
68 Kenneth Einar Himma, Positivism, Naturalism, and the Obligation to Obey Law, 36 S. J. PHIL. 145–161 (1998).
Accordingly, legal positivists might in a more modest vein insist that the separability thesis at least dissolves any presumption favoring the existence of a duty to obey the law, and this is one way of taking the distinctively legal positivistic admonition that the existence of law is one thing; its moral merit is another.
69 Cf. John Gardner, Legal Positivism: 51/2 Myths, 46 AM. J. JURIS. 199–227 (2001), at 206– 207).
A legal positivist who also adheres to a liberal political philosophy might thus have two grounds for doubting the existence of the duty: the separability thesis and the presumption of liberty (perhaps
sub nomine the coerciveness of law).
The first volley in the “presumptions wars” arguably was fired by apologists for the state at least as early as 1930, with David Ross's appeal to the intuitive obviousness that there is a duty to obey. More recently, the existence of a duty to obey the law is often said to share the presumptive correctness of any such deliverance of common sense and to represent (to use a term popularized by Rawls) a “provisional fixed point” in our moral reasoning. Moreover, many hold as a general matter that tenacious commonsense beliefs are presumptive victors over competing propositions whose only footing is in conceptual inquiry or political theorizing. There has been empirical work to confirm the commonsense status of the doctrine that there is a duty to obey the law,70
TOM R. TYLER, WHY PEOPLE OBEY THE LAW (1990).
but the analysis has been challenged as resting upon responses to ill-formed survey questions.
71 According to Green, Tyler's work fails to disclose what ordinary respondents would have to say about the hypothetical cases that are grist for the philosopher's mill. Green concludes that the duty to obey the law has not justified its claim to share the presumptive correctness of commonsense and so must be viewed as but one political theoretic proposition among others—and an implausible one at that.
The duty to obey is typically called into doubt by an elenchus which is perhaps but an instance of a wider pattern whose general availability undermines the very idea that there are any duties as the term “duty” is commonly understood—that is, as a type of act having a significant and invariably positive moral value. Any putative duty is vulnerable to critical challenges that are designed to extract the concession that in certain circumstances the duty would have to yield. Ross's terminology of “prima facie” duty was coined in order to reconcile the possibility of such concessions with the existence of act-types that are normally obligatory even though they are not obligatory in certain extraordinary circumstances. All may seem well after the “defeasible” nature of our duties is duly noted. But recent decades have witnessed an insurgency of moral particularism, which denies that obligatoriness applies to act-types at all, in pro tanto fashion or otherwise.72
JONATHAN DANCY, MORAL REASONS (1993).
According to particularism, the fact that an act-token would be an instance of law-abidingness is not invariably a significant reason in favor of performing it. Moreover, that fact is not invariably of positive valence—sometimes the fact that an action would be one of obedience is a reason against performing it—as, for example, when the particular law or the legal regime happens to be unjust.
73 Jonathan Dancy, The Particularist's Progress, In MORAL PARTICULARISM, 130–156 (Brad Hooker and Margaret Olivia Little, eds., 2000).
It all depends.
74 Cf. Mark Tunick, The Moral Obligation to Obey Law, 33 J. SOC. PHIL. 464–482 (2002).
The defender of duties as act-types will respond with the familiar Rossian move to the pro tanto, but a determined particularist will insist that the list of necessary conditions and qualifications is potentially endless and that the lesson to be drawn is that obligatoriness attaches not to act-types but to particular act-tokens. The particularist challenge has been noticed, and attempts have been made to answer it,
75 See, e.g., Frank Jackson, Philip Pettit, & Michael Smith, Ethical Particularism and Patterns, in MORAL PARTICULARISM, 79–99 (Brad Hooker and Margaret Olivia Little, eds., 2000).
but space does not permit further exploration here. It will suffice to state the obvious point that should it turn out that there are no obligatory act-types at all, then there is no duty to obey the law.
IV. POSITIVE ACCOUNTS AND REBUTTALS
The preponderance of the recent literature on the duty to obey the law consists of positive accounts of the duty, responsive criticism, and rejoinders. The positive accounts typically acknowledge the influence of John Simmons's 1979 book, Moral Principles and Political Obligations, which built so impressive a negative case that it is no exaggeration to say that the literature of the intervening quarter-century has largely consisted of efforts to overcome or deflect Simmons's objections.
It may be helpful to contrast “primitive” and “derived” positive accounts. Derived accounts seek to justify the duty by tying it to a wider and perhaps less controversial moral principle or cluster of moral principles. Primitive accounts, on the other hand, seek to defend the duty as morally freestanding: Primitive accounts may but need not locate the duty within a wider constellation of moral principles, with which they may conflict. Instances of primitive accounts include the “conceptual argument” mentioned earlier, which represents moral obligatoriness as essential to the existence of a legal system.76
Certain hierarchical accounts of society and law's role in giving society its proper shape could be considered as primitive accounts of the duty to obey the law—I have in mind F.H. Bradley's “My Station and Its Duties” and resonances of Edmund Burke's or Hegel's organic visions of civil society. Also perhaps belonging in the primitive category are defenses that appeal to the presumptive correctness of moral intuition: Mark Murphy's recent defense of the duty by appeal to what he calls the “conscience principle” is an example.
77 M. Murphy, Philosophical Anarchisms, supra note 52.
Another way to take such accounts is as amplifications of an adequacy condition applicable to other, derived accounts.
Any discussion of a derived theory naturally falls into two parts. The first is an elaboration of the more general principle. The second is an evaluation of the prospects of assimilating political obligation to the general principle. Within the category of the derived, it may also be helpful to observe a distinction between “unary” and “mixed” accounts. Unary derived accounts approach the problem of defending the duty as a problem of deriving the duty from a single, more general moral principle whose validity is less dubious than that of the duty to obey the law. Derived mixed accounts do not restrict themselves to a single moral principle. Unary derived accounts can be further divided into three subcategories: natural-duty accounts, volitional accounts, and associative accounts. Briefly, natural duties are predicated upon nothing more than the personhood of the duty-bearer; volitional duties are predicated upon some voluntary act or preference of the duty-bearer; and associative duties are predicated upon the duty-bearer's perhaps unchosen and unwanted association with some proper subset of all persons. Other divisions are possible. Simmons and Wellman distinguish natural-duty accounts, associative accounts, and transactional accounts, each having further subdivisions.78
A. John Simmons, Political Obligation and Authority, in BLACKWELL GUIDE TO SOCIAL AND POLITICAL PHILOSOPHY (Robert L. Simon, ed., 2002); Christopher H. Wellman, Political Obligation and the Particularity Requirement, 10 LEGAL THEORY 97–115 (2004).
Green divides the leading positive accounts into two major categories: voluntarist theories and nonvoluntarist theories.
79 I claim no particular advantage for the scheme of classification I use here, other than the possibility that organizing the field in a different way may highlight important linkages.
80 See also RUTH HIGGINS, THE MORAL LIMITS OF LAW: OBEDIENCE, RESPECT, AND LEGITIMACY (2004).
Mixed derived accounts might conceivably draw upon elements from any combination within or across the three subcategories, but in fact advocates of mixed theories have not found it fruitful to exploit all such possibilities. In what immediately follows, I will outline some of the principal unary derived accounts. I will conclude by considering several noteworthy mixed theories.
A. Natural Duty Accounts
Rawls described natural duties in this way:
in contrast with obligations, it is characteristic of natural duties that they apply to us without regard to our voluntary acts. Moreover, they have no necessary connection with institutions or social practices; their content is not, in general, defined by the rules of these arrangements…. A further feature of natural duties is that they hold between people irrespective of their institutional relationships; they obtain between all as equal moral persons. In this sense the natural duties are owed not only to definite individuals, say to those cooperating together in a particular social arrangement, but to persons generally. This feature in particular suggests the propriety of the adjective “natural.”81
RAWLS, supra note 18, at 114–115.
I have discussed Rawls's shifting view of political obligation elsewhere;
82 William A. Edmundson, Introduction, in THE DUTY TO OBEY THE LAW: SELECTED PHILOSOPHICAL READINGS (William A. Edmundson, ed., 1999b).
here it will suffice to say that his mature view was that political obligation, insofar as it purports to be universally borne, must be defended as deriving from a natural duty to support and comply with just institutions.
83 RAWLS, supra note 18, at 114–117, 333–355.
There are other natural-duty theories as well, but recent discussion has been dominated by a general objection, termed the “particularity problem,” which has been made to any natural-duty account of political obligation and which some view as the most serious difficulty for any defense of political obligation.
84 John Simmons, who has pressed the objection, puts it this way:
Political obligations are felt to be obligations of obedience and support owed to one particular government or community (our own), above all others. Citizens' obligations are special ties, involving loyalty or commitment to the political community in which they were born or in which they reside. More general duties with possible political content, such as duties to promote justice, equality, or utility, cannot explain (or justify, or be) our political obligations, for such duties do not necessarily tie us either to one particular community or to our own community.85
A. JOHN SIMMONS, MORAL PRINCIPLES AND POLITICAL OBLIGATIONS (1979), at 250.
Rawls's natural duty to support just institutions, for example, is one that everyone—wherever and however situated—is supposed to owe toward existing, sufficiently just institutions—wherever and however situated. But the fact is that people believe they have special ties to their own states (however flawed) and not to others (however just). Natural-duty theory cannot dismiss this attitude without doing violence to what is a settled conviction of many reflective and reasonable people, and therefore the theory must instead somehow reconstruct it. The most promising lines of reconstruction for the natural duty theorist, however, are ones that seem to invoke things like consent or receipt of benefits—invocations that would lead to a mixed theory. The pursuit of a unary natural-duty theory of political obligation, the objection concludes, is a dead end. The objection is equally apt with respect to the duty to obey or to more encompassing notions of political obligation; it seems odd to say that one owes to persons generally a duty to obey the law, and odder still if we were to understand that duty to encompass the laws of (sufficiently just) states generally.
Jeremy Waldron has offered a general answer to the particularity objection on behalf of natural-duty accounts: he begins by noting a qualification in Rawls's statement of the natural duty and a related objection arising from that qualification.86
Rawls's duty to support and comply with just institutions is restricted to those that “apply to us,” and the objection
87 SIMMONS, supra note 85, at 151.
is that no sense can be given to the restriction without turning to ideas of consent or fair play. (The companion duty to help
create just institutions is not qualified in this way but is instead qualified by the condition that the cost to the actor be minimal.) Preliminarily, Waldron notes that all will agree that there is a natural duty not to undermine just institutions existing elsewhere and argues that theories of “acquired” obligation (including consent and fair-play theories) cannot well explain why this should be so.
Waldron distinguishes between “range-limited” and range-unlimited principles of justice, and, as to the former, between two categories of person—insiders and outsiders. Range-limited principles are principles intended to do justice between a limited set of persons, for example, Hobbes's children or New Zealanders. “Insiders” are simply those within the range of “conduct, claims, and interests” with which the relevant principle—or its administering institution—purports to deal88
Waldron, supra note 18, at 279–280.
—they need not be volunteers or in any other sense beneficiaries. Waldron cites three conditions that must be satisfied if range-limited principles of justice are to be effective: (1) insiders must accept the demand of the relevant range-limited principle; (2) insiders must accept the demand that they accept the administering institution's administration of the principle; and (3) insiders and outsiders must refrain from undermining the administering institution. Waldron claims that his account of these demands makes sense of the Rawlsian proviso that the actor's duty to support just institutions is limited to those that “apply to him” and at the same time explains “much of” the specialness of an actor's relationship to his own national institutions—conceding that there is a somewhat atavistic residue of “patriotic affect” that escapes. Waldron's position is that “an organization that is just, effective, and legitimate (in the sense of being singled out as
the salient organization for this territory) has
eo ipso a claim on our allegiance.”
89 With reference to Simmons's objection that others “cannot simply force institutions on me, no matter how just, and force on me a moral bond to do my part,”
90 SIMMONS, supra note 85, at 148.
Waldron responds that because the pursuit of substantive justice is morally imperative, at some point “the theorist of natural duty must stop treating [the] question, ‘Can an organization simply impose itself on us, morally…?’ as an objection and simply insist that the answer is yes.”
91 Waldron, supra note 18, at 27.
Although the particularity worry has dominated much recent discussion, the intuition from which it springs is not beyond question, and the absence after a decade of a compelling rejoinder to Waldron suggests that the worry may be overwrought.92
Cf. Mason, supra note 20, at 436–437; Wellman, supra note 78, at 101–105.
John Simmons insists that “political obligations, properly understood, must bind us to one particular community or government in a way that is special [i.e., a way that arises from special relationships with certain others and are owed to these others and not to humankind generally]; if an obligation or duty is not ‘particularized’ in this way, it cannot be what we ordinarily think of as a political obligation.”
93 Simmons, supra note 78, at 29.
But it is not obvious why the particularity intuition should be any more sacrosanct than the prereflective intuition that political obligations exist—an intuition that Simmons himself repeatedly warns against taking at face value.
94 It may turn out that our political obligations are more cosmopolitan than we suppose prior to reflection. Simmons concedes that dual and multiple citizenship are held by many but doubts “whether one can satisfy all of the
possible demands of obedience and support to more than one state simultaneously.”
95 Id. at 29; emphasis in original.
The doubt seems hyperbolic, however; it is a commonplace occurrence for pro tanto duties to come into conflict without ceasing
ipso facto to be genuine duties.
On Waldron's account, the range of a range-limited principle is determined with reference to its “point and justification,”96
Waldron, supra note 18, at 280 of reprint.
and so need not be limited in range to a geographical area; and a “territory,” as he uses the term, need not correspond to any conventional boundary but may be “any area within which conflicts must be settled if any stable system of resource use is to be possible among the inhabitants.”
97 The possibility arises that more than one institution may impose rules that apply to persons within a geographical area, and—because conflicting rules may be equally just (e.g., rules of the road)—persons may find themselves subject (as Simmons notes) to conflicting rules that apply to them. In other words, people might find themselves to be insiders with respect to multiple institutions administering incompatible principles. Waldron's response is that no natural duty arises to obey the rules of a just, effective institution unless the institution also possess an additional, exclusive characteristic which he calls
legitimacy, namely, “a good reason to recognize
this organization, as opposed to any rival organization, as
the one to do justice in the given territory or with regard to the claims that are at issue.”
98 Id. at 287 of reprint; emphasis in original.
Thus, for Waldron as for Simmons, conflicting demands result in not two prima facie duties but either no duty or a single duty.
The question arises: Why must legitimate political authority divide into mutually exclusive geographical territories? Being simultaneously subject to multiple quasi-sovereign authorities is not only a conceptual possibility but for many an everyday reality. In the United States, for example, there is a constitutional division of sovereignty between the states and the federal government, but additionally there are interstate authorities (e.g., the Port Authority of New York and New Jersey), and within the states there are further divisions between counties, municipalities, special-use districts, special-purpose regional land-use and resource management authorities, and so forth (e.g., tribal reservations that lie within or straddle state boundaries). The potential for conflict and duplication inherent in this variety is of course reduced by rules that tend to impose a manageable hierarchy, but it is a hierarchy that is shifting and contingent and one whose existence should not obscure the fact that a unitary, apical Leviathan is more a creature of theory than of experience—just as the “Westphalian” nation-state itself is a creature of historical contingency.99
F. OST & M. VAN DE KERCHOVE, DE LA PYRAMIDE AU RÉSEAU? POUR UNE THÉORIE DIALECTIQUE DU DROIT (2002); Cushing, supra note 50, at 230 n.41; ANNE-MARIE SLAUGHTER, A NEW WORLD ORDER (2004); cf. LESLIE GREEN, THE AUTHORITY OF THE STATE (1988).
Waldron's defense of a geographically exclusive conception of legitimacy extends Kant's story about the moral imperative to originate civil society: “Since no one can afford to wait until all possible conflicts arise so that all can be definitively settled at once, the Kantian approach implies that I should enter quickly into a form of society with those immediately adjacent to me, those with whose interests my resource use is likely to pose the most frequent and dangerous conflicts.”100
Waldron, supra note 18, at 15.
Thus principles of justice may be limited “at least on a
pro tem basis,” and we may understand the justice of a system of nation-states in this way. But, as Waldron notes, “as the sphere of human interaction expands, further conflicts may arise, and the scope of the legal framework must be extended and if necessary re-thought.”
101 Might not such rethinking extend to the question of the proper deference owed to the notion of territorial sovereignty? In the two centuries since Kant wrote, the circle of our concerns for justice has expanded at least as far as the ever-expanding spheres of human interaction. But as circles expand, they begin to overlap: the Westphalian world, neatly partitioned into nation-states, has already given way to a complex pattern of overlapping local, sub-, and supranational regional and international organizations, nongovernmental and quasi-nongovernmental organizations (NGOs and quangos) and multinational treaty organizations such as the UN, the EEU, and the International Court of Justice. The importance of fashioning transnational political institutions capable of doing justice has already proven to be greater than that of preserving exclusive sovereignties. Waldron rightly insists that the “backbone” of a natural-duty position is the moral imperative of justice;
102 but that is precisely why territorial exclusivity—however deep the patriotic affect associated with it—is ultimately a contingent administrative expedient rather than an element of legitimacy. Waldron's natural-duty response to Simmons could be amplified by stating, in effect, “Yes, an organization can indeed impose itself upon us morally,
and without regard to locality.”
103 William A. Edmundson, Introduction: Some Recent Work on Political Obligation, 99 APA NEWSL. ON PHIL. & L. 62–67 (1999c).
Particular natural-duty accounts and their special difficulties include the following:
Utilitarianism and Other Consequentialist Theories
The duty to obey the law is insupportable from an act-consequentialist or direct-consequentialist perspective.104
See George Klosko, Parfit's Moral Arithmetic and the Obligation to Obey the Law, 20 CAN. J. PHIL. 191–214 (1990b); and cf. Christopher MacMahon, Authority and Autonomy, 16 PHIL. & PUB. AFF. 315–328 (1987).
Therefore consequentialist defenses of the duty to obey depend upon the working-out of plausible versions of rule or indirect consequentialism—as indeed was Rawls's initial attempt.
105 John Rawls, Two Concepts of Rules, 64 PHIL. REV. 3–13 (1955).
The rules concerned here are not conventional rules tied inextricably to social practices but rather are ideal rules that stand apart—and so understood may indeed correspond to what are natural duties according to Rawls's criteria. There has been an enormous amount of work on the tenability of rule consequentialism, but the duty to obey the law is rarely singled out for special attention. There has also been extensive discussion of rules, their nature, and their moral appeal.
106 See, e.g., Alexander, supra note 38.
This discussion, too, though pregnant with implications for the duty to obey, is beyond my scope; the reader should consult Shapiro.
107 Scott Shapiro, Authority, in THE OXFORD HANDBOOK OF JURISPRUDENCE AND PHILOSOPHY OF LAW (Jules Coleman, Scott Shapiro, and Kenneth Einar Himma, eds., 2002).
Instrumentalist or Necessitarian Theories
The general form of such theories can be presented syllogistically: (P1) whatever is typically a necessary means to a morally compelling end is at least a pro tanto duty; (P2) law-abidingness is typically a necessary means to a morally compelling end; therefore (C) law-abidingness is at least a pro tanto duty. John Finnis defends a natural-law account of this general form.108
One point of contention invited by such theories is the specification of morally compelling ends. Finnis, acknowledging the plurality of ends sought by citizens, emphasizes law's unique ability to secure the common good, namely, “the good of individuals living together and depending upon one another in ways that tend to favour the well-being of each.”
109 John Finnis, Law as Co-ordination, 2 RATIO JURIS 97–104 (1989), at 103.
Similarly, Mason invokes the intrinsic value of citizenship.
110 These are values that others of a more individualistic persuasion would contest or deflate. This difficulty might be finessed by observing (à la Raz) that the normal way to justify legal coercion is by establishing that citizens better comply with the reasons that apply to them (whatever they may be) by obeying than by determining for themselves what reason requires.
111 However, Raz himself is skeptical that states possess the competence to establish a comprehensively applicable and universally borne duty of obedience.
112 Joseph Raz The Obligation to Obey: Revision and Tradition, 1 NOTRE DAME J.L. ETHICS & PUB. POL'Y 1: 139–155 (1984).
Yet another problem for instrumentalist theories is what could be called the “harmless disobedience” difficulty, often put with reference to what have been termed “stop-sign-in-the-desert” examples, which are devised to show that there is nothing even pro tanto wrong with disobeying the law when there is a vanishingly low chance of harm and a palpable benefit to be gained.113
Unless obedience is itself a morally compelling end, such theories are open to the objection that perfect and universal obedience is as a matter of fact not necessary to achieve plausible social ends—such as order, harmony, or substantive justice. A standard first move made to avoid the harmless-disobedience difficulty is to insist upon the necessity of social coordination to achieve a range of morally compelling ends
114 Cf. Finnis, supra note 109; TONY HONORé, MAKING LAW BIND (1987), at 56–66.
—but this move is generally agreed to fail for the simple reason that schemes of social coordination are typically able to tolerate nonconformity in small amounts. What has become the standard fallback move for such theories is the invocation of a fair-play duty, which condemns even harmless noncompliance as unfair to (or disrespectful of the equal worth of) those who do comply with socially beneficial rules. Theories that avail themselves of this move could be considered as mixed rather than pure theories or possibly as pure fair-play theories.
Contractarian or Hypothetical-Consent Theories
Despite their voluntaristic flavor, accounts of this type are now widely thought to be properly classified as natural-duty theories.115
Simmons, supra note 78; David Schmidtz, Justifying the State, 101 ETHICS 89–102 (1990); Cynthia Stark, Hypothetical Consent and Justification, 97 J. PHIL. 313–334 (2000); Cushing, supra note 50.
Because such accounts are designedly insensitive to what Simmons calls “the eccentricities of individual uptake,”
116 their emphasis falls heavily upon what one ought to consent to rather than upon what one has in fact consented to. Moreover, because such theories typically invoke a highly idealized choice situation—such as Rawls's celebrated “original position”—their relevance to the constrained circumstances in which citizens would actually exercise or withhold their consent is questionable.
117 Rawls's contractarian defense of the natural duty to support and comply with just institutions in A Theory of Justice has been so influential as to occupy the field.118
Rawls, supra note 18; cf. Simon Cushing, Rawls and “Duty-Based” Accounts of Political Obligation, 99 APA NEWSL. ON PHIL. & L. 71–77 (1999); Lefkowitz, supra note 27, at 412–415.
Space will not permit a detailed examination of Rawls's view here; but the difficulties attending a contractarian account of the duty to obey are fairly obvious. For one, it is unclear what could rationally compel assent to a comprehensively applicable duty when it is well known that even just states may adopt silly and unjust laws. For another, the merits of the content independence that is supposed to characterize the duty to obey is controversial,
119 See, e.g., Alexander, supra note 38; Moore, supra note 62; Hurd, supra note 62.
and thus it is difficult to understand why rational assent to the duty would be compelled, especially in the absence of any assurance that the state's judgment is generally superior to the citizen's.
120 Raz, supra note 112; CHRISTOPHER MORRIS, AN ESSAY ON THE MODERN STATE (1998).
To the extent that contractarian arguments take on a consequentialist flavor, they must overcome the “harmless disobedience” difficulty that attends the latter; and to the extent that the principle of fairness is invoked to avoid the difficulty, that principle must be reconstructed on a contractarian foundation if the overall account is to be unary rather than mixed.
121 Cf. WILLIAM A. EDMUNDSON, AN INTRODUCTION TO RIGHTS (2004), at 114–118.
Fair-Play Accounts
These accounts are inspired by H.L.A. Hart's celebrated duty of fair play, that is, the duty to cooperate that falls upon those who benefit from the cooperative sacrifices of others.122
H.L.A. Hart, Are There Any Natural Rights? 64 PHIL. REV. 175–191 (1955).
As Rawls expressed the idea:
when a number of persons engage in a mutually advantageous cooperative venture according to rules, and thus restrict their liberty in ways necessary to yield advantages to all, those who have submitted to these restrictions have a right to a similar acquiescence on the part of those who have benefitted from their submission.123
Rawls, supra note 18, at 112.
Coordinate to the “right to acquiescence” is a duty on the part of the beneficiaries to submit to the rules. In Rawls's contractarian scheme, the principle of fair play would be adopted by suitably situated and qualified choosers and would hence count as a natural duty. Nonetheless, particular fair-play duties do not count as natural duties, Rawls concluded, because applications of the principle presuppose the knowing and presumably voluntary acceptance of benefits. The qualification led Rawls to abandon his erstwhile
124 John Rawls, Legal Obligation and the Duty of Fair Play, in LAW AND PHILOSOPHY, 3–18 (Sidney Hook, ed., 1964).
hope of deriving universally borne political obligations from the principle of fairness. Simmons has elaborated this qualification and its implications for a fair-play account of the duty to obey.
125 Simmons, supra note 85, Simmons, supra note 25.
A fairness principle that presupposes voluntary participation in a cooperative social venture and the willing acceptance of benefits deriving therefrom would be of limited use in the defense of a duty to obey the law. Although such a defense could fairly easily satisfy particularity worries (transborder “spillover” effects aside), it would fail to establish a duty universally borne except in the rare instance of smaller, well-integrated communities. In larger states, it is unlikely that all will regard themselves as willing cooperators.
126 Even if a cooperation condition were dispensed with, there remains the difficulty that public goods such as police protection and national defense are ones whose receipt noncooperators have no real choice about; and as to such “nonexcludable” goods, a principle of fair play that requires acceptance of benefits over and above mere receipt is unsatisfiable.
127 Id.; Wolff, supra note 50.
A fair-play defense of the duty to obey thus faces a dilemma. A broader, nonvoluntaristic conception of the principle is vulnerable to Robert Nozick's notorious “classical music” counterexample:128
Robert Nozick, ANARCHY, STATE, AND UTOPIA (1974).
Even if I enjoy the classical music my neighbors cooperatively broadcast by sacrificing one day a year at the community turntable, why should I have to pitch in? But a narrower, voluntaristic conception fails to generate a universally borne duty.
129 Simmons, supra note 85; Wolff, supra note 50.
A number of philosophers
130 Klosko, supra note 17; Arneson, supra note 17; Dagger, supra note 11.
have treated the dilemma. Klosko, for example, contesting Rawls's and Simmons's voluntaristic formulation of the fairness principle, has emphasized the role of “presumptively beneficial goods”—a concept intended to track Rawls's primary goods, which are (unlike piped music) irrebuttably taken to be valued by any rational agent.
131 But cf. Alan Carter, Presumptive Benefits and Political Obligation, 18 J. APPLIED PHIL. 229–243. (2001).
I have discussed these efforts and Simmons's counters
132 A. John Simmons, Fair Play and Political Obligation: Twenty Years Later, in JUSTIFICATION AND LEGITIMACY: ESSAYS ON RIGHTS AND OBLIGATIONS (2001a).
to them elsewhere
133 Edmundson 1999b, 1999c; Edmundson, Locke and Load, 22 LAW & PHILOSOPHY 195–216 (2003).
and will not recapitulate that discussion here. It will perhaps be enough to say that the running dispute between Klosko and Simmons as to the proper general formulation of a binding principle of fairness turns upon appeals to one's intuitive response to a number of hypothetical state-of-nature situations that involve undoubted benefits that cooperators bestow upon an “outsider” who prefers self-provision. Even if (as I think) Klosko has the better of the argument as to the general formulation of the fair-play principle, the problem remains that far from all of the goods in the bundle the state provides are presumptively beneficial, and thus the duty generated fails to be comprehensively applicable.
134 George Klosko, The Obligation to Contribute to Discretionary Public Goods, 38 POL. STUD. 196–214 (1990a); Edmundson, supra note 103.
B. Volitional Accounts
A volitional account is one that renders the duty to obey as the moral consequence of some act or course of action undertaken by the party who is bound—or, at the barest minimum, by that party's actually welcoming or valuing something, whether or not it is the product of her own action. In contrast, an associative account renders the duty to obey as deriving in the last analysis from transactions among persons with whom the party bound is associated, and perhaps involuntarily so. Grotius and Hobbes gave accounts that are rather explicitly volitional—invoking analogies to contracting and promising—while Locke's theory of the obligation to obey invokes the idea of a tacit consent manifested by continued residence within a political territory. Volitional accounts seem relatively untroubled by particularity worries, at least insofar as the relevant volitional predicate is assumed to be inherently “range-limited”—as Simmons has insisted they are135
Simmons, supra note 78, at 28.
—but with respect to the less explicit predicates, complications can be expected here along the same lines as those that worry natural-duty accounts. Volitional accounts as a group invite the standard criticism that the duty to obey is not universally borne but, rather, borne only by those who have satisfied some volitional condition. Volitional accounts thus face a dilemma: Either the volitional condition is a strong one—for example, that an express promise to obey has been given—in which case the binding power of the condition seems strong, but the condition is not universally satisfied, or the volitional condition is weaker—for example, mere presence in a territory—in which case the condition is likely to be universally satisfied within a territory, but its binding force is questionable. Hypothetical consent and contractarian theories now tend to be classified with natural-duty theories, as noted above, insofar as their volitional predicates are satisfiable by reference to what an idealized actor would want or do, rather than what the obligee (i.e., duty-bearer) in fact wants or has done.
Fair Play (with Acceptance Condition)
Fair-play accounts, as noted above, hover in the twilight between the volitional and the associative. Rawls and Simmons understand the background moral principle of fairness as requiring a volitional element, namely, willing acceptance of benefits. Klosko rejects the voluntarist insistence upon willing acceptance rather than “mere” receipt and thus advocates an associative understanding of the principle of fairness. H.L.A. Hart, who first tendered the principle of fair play, as he termed it, did so in response to the eponymous question posed in the 1955 article “Are There Any Natural Rights?”—suggesting that the correlative duties would themselves be “natural.” As explained above, the general principle of fairness or fair play is better understood as stating a natural duty, but one whose applications have (here contention begins) either an associative or a volitional predicate. If, as Simmons insists, a volitional predicate must be satisfied—that is, that benefits be willingly accepted rather than merely received—then the prospects of defending a universally borne duty of obedience are drastically diminished. As Simmons points out and Klosko concedes, an important range of benefits conveyed by the state are ones that cannot readily be refused (e.g., public goods such as territorial defense) and so are not in any comfortable sense willingly accepted. Moreover, recipients who would prefer self-provision or market provision of such benefits cannot be said to have willingly accepted benefits that have been, as it were, thrust upon them willy-nilly.
Consent Theories
The Achilles heel of consent and (“actual”) social-contract theories is, as has been notorious since Hume, the fact that the necessary predicate of consent or agreement has not been universally satisfied.136
Although occasional doubts have been raised as to whether a comprehensively applicable duty could arise consistently with the duty of autonomy,
137 the claim that universal consent would adequately ground the duty to obey is usually unchallenged, except by way of emphasizing that the required act of consent must be uncoerced and informed and that the legal system consented to not be grossly unjust.
138 Margaret Gilbert, Reconsidering the “Actual Contract' Theory of Political Obligation, 109 ETHICS 236–260 (1999); Simmons, supra note 85.
The empirical fact of insufficiently widespread consent or expression of consent has been referred to as the problem of descriptive adequacy
139 or as the “ ‘no agreement’ objection.”
140 Gilbert, supra note 138, at 240.
Responses to the “no agreement” objection have been of three types. One is to confess the problem and the consequent nonexistence of a universally borne duty in modern states while calling for reforms intended to secure wider consent.141
Beran, supra note 19; Bernard R. Boxill, On Some Criticisms of Consent Theory, 24 J. SOC. PHIL. 81–102 (1993); Simmons 2001a; cf. historical studies by John Dunn, POLITICAL OBLIGATION IN ITS HISTORICAL CONTEXT (1980); Michael Davis, ACTUAL SOCIAL CONTRACT: A PHILOSOPHER'S HISTORY THROUGH LOCKE (2002).
Another is to dispense with a requirement of actual consent and to construct instead a contractarian theory featuring hypothetical consent (see above). The third is to explore the possibility of rehabilitating the descriptive adequacy of consent theory. This third way is exemplified by the work of Mark Murphy and Margaret Gilbert.
Gilbert assimilates political obligation to a wider category of “joint commitments” formed when persons “mutually express their readiness to be so committed, in conditions of common knowledge.”142
Margaret Gilbert, Group Membership and Political Obligation, 76 THE MONIST 119–131 (1993), at 123.
Such persons “cannot unilaterally remove” themselves from the commitment and, to that extent, form a “plural subject.”
143 The relevant mutual expression need not be a datable event, and the content of the commitment may be vague, but a central conversational use of the first-person plural pronoun typically invokes or presupposes a plural subject. Gilbert calls this idiom the “plural subject sense of ‘we.’”
144 Gilbert, supra note 138, at 250; see also Michael E. Bratman, Shared Intention and Mutual Obligation, In FACES OF INTENTION (1999).
Mutual acquiescence in this idiom presumptively forms a plural subject, and this phenomenon is identifiable on larger, less direct scales, as where, for example, islanders use the plural-subject idiom to express their readiness to be jointly committed to other islanders, whoever they may be. “Social groups are plural subjects,” she argues, “constituted by joint commitments which immediately generate obligations.”
145 Gilbert, supra note 142, at 126.
Promises and agreements generally are, for Gilbert, but instances of joint commitment.
With respect to the problem of political obligation, Gilbert discounts Simmons's objection to actual consent theory: that agreements do not obligate when made under coercion nor when their content is morally flawed; and Simmons's “no agreement” objection—which is damaging to actual consent theory—does not touch plural-subject theory, which does not rest upon agreement. Gilbert argues that the widespread practice of referring to “our country,” “our constitution,” and “our law” is interpretable as a use of the “plural subject sense of ‘we,”’ and that persons who employ such idioms are to be so understood. Such an interpretation would explain the sociological fact that there is a widespread and deep sense of attachment to “our country” and an obligation to obey its laws.
Gilbert is, however, elaborately tentative in her conclusions, and rightly so. Although the phrase, “We, the People,” ringingly begins the U.S. Constitution, and references to “our constitution” are commonplace, it is doubtful that these usages are capable of bearing the full weight Gilbert asks of them. In any case, the colloquial phrase is “duty to obey the law” and not “duty to obey our law,” which to this writer's ear has an ominously cabalistic ring. As Gilbert readily acknowledges, a plural-subject account requires an express joint commitment bespeaking “an intention to obligate oneself.”146
Gilbert, supra note 138, at 254.
As such, plural-subject theory stands in contrast to those types of voluntarist theory that require something intentional but short of an intention to obligate oneself. Because of the relative specificity of the intention it requires, a plural-subject account seems less likely to be universally borne than voluntarist accounts that require a more diffuse intentional element.
Mark Murphy's rehabilitation of consent theory is such an account. Murphy's point of departure is the observation that acting on abstract moral principles often involves what Aquinas called a determinatio or choice among acceptable means, no one of which is morally compelled.147
M. Murphy, supra note 33.
This is especially so with respect to ends such as justice, which are achievable only by cooperative action. (Honoré
148 Honoré, supra note 114, at 115–138.
similarly exploits morality's incompleteness but does so as part of a necessitarian argument.) There is consent “in the acceptance sense” to the rules of a cooperative scheme to achieve a choice-worthy end when the actor practically treats the group rule as her own determination of principle. Consent in this sense requires no “occurrent” or “attitudinal” event or disposition in order to bind morally. Rather, it requires no more than that the actor have made a determination of general principles of justice that apply to her; and its moral bindingness flows from the fact that we are morally bound to act according to determinations of general moral principles that apply to us if failing to so act would frustrate the point of our having so determined.
As a matter of empirical fact, it seems plausible to suppose that citizens employ their knowledge of the law in this way much more frequently than they declare their consent to its authority. As an illustration, Murphy points to drunk-driving laws that specify (“determine”) safe blood-alcohol levels. Citizens' surrender of judgment to the state's determinations on a range of such subjects is what constitutes the political authority of the state, and to the extent that we are morally required to accept salient determinations of moral principle, we are not morally at liberty to revoke our acceptance of law by the simple expedient of discontinuing our use of social rules in our practical decision-making.
Murphy thus proposes to close the gap that exists in a natural-law account between a moral principle requiring all to promote the common good and the several provisions of the positive law of the particular jurisdiction in whose domain individuals find themselves. Murphy faults John Finnis's reliance149
JOHN FINNIS, NATURAL LAW AND NATURAL RIGHTS (1980), at 231–232; Finnis, supra note 6.
upon the law's mere salience to close this gap; for the law's salience by itself is insufficient to endow it with authority, that is, with the moral power to impose duties. At one point Murphy suggests that the common-good principle and the law's status as “salient coordinator” might
morally require that “one submit to a political authority by performing some relevant obligation-generating act.”
150 Mark C. Murphy, Natural Law, Consent, and Political Obligation, 18 SOC. PHIL. & POL'Y 70–92 (2001), at 83.
But Murphy does not pursue that thought, which in any case would raise the question that has widely been raised about hypothetical-consent theories: What does “reasons
r morally require X to consent to φ” generally add to “reasons
r morally require X to φ”—other than voluntaristic window-dressing? Instead, Murphy concedes that his consent-in-the-acceptance-sense account fails to underwrite a universally borne duty
151 and that natural lawyers may have to grant the philosophical anarchist's point—but with a rueful “so much the worse for us”
152—if at the end of the day it turns out that consent in the acceptance sense is insufficiently widespread. The extent of such consent has to be doubted. As the hoary maxim
ignorantia legis neminem excusat warns, the law purports to oblige us whether or not it has figured in our practical reasoning. Moreover, although the salient coordination equilibria the law represents may often figure in practical reasoning, they often figure not as determinations of a common-good principle but—from a Holmesian “bad man's perspective”—as predicative hypotheses about possible consequences. Murphy ingeniously explains how the common-good principle might require those who sincerely seek the law's guidance to follow it. This is not a trivial result but, as he acknowledges, this category of citizen was never thought to be the most troublesome.
C. Associative Accounts
An associative duty is one that is neither natural nor volitional. It is not natural insofar as it presupposes more than the common humanity of the obligee and obligor and therefore does not run to all other humans (or moral persons) as such. What an associative duty presupposes is that the obligee be related to some association of persons in a certain way and that the association itself have a certain character. Associative duties are like natural duties in that they do not presuppose any undertaking or particular preference structure on the part of the obligee. Ronald Dworkin and Michael Hardimon have argued that the possibility of reconstructing political obligations along associativist lines undermines any presumption in favor of holding them to a volitional standard. Leslie Green has plausibly countered that the “scope and supremacy” of the state's claimed authority is sufficiently unlike the favorite associativist paradigms that the argumentative burden must remain with apologists for the state.153
Particular associativist accounts include the following:
Fair Play
As indicated above, the principle of fair play is a natural duty whose particular applications presuppose at least an associative and perhaps a volitional predicate. If an associative predicate suffices to impose a fair-play duty, then the prospects of a fair-play account of the duty to obey are bright—or at least far brighter than can be the case if a volitional predicate must be satisfied. Unlike unmixed instrumentalist theories, fair-play accounts can respond to the “harmless disobedience” objection by insisting that disobedience is wrongfully unfair even if harmless—and this is a significant advantage.154
Cf. CHAIM GANS, PHILOSOPHICAL ANARCHISM AND POLITICAL DISOBEDIENCE (1992).
Even so, there are difficulties that make the duty reconstructed upon fair-play lines one that is short of comprehensive applicability, as discussed above.
Gratitude
The appeal to gratitude is one of the oldest arguments for the duty to obey. It figures prominently in the Crito but is not evident in the Enlightenment tradition, which emphasizes the agent's autonomy rather than the mere receptivity of the passive beneficiary. Its remaining influence seemed extinguished forever by Simmons's critique,155
Simmons, supra note 85, at 157–190.
but it has found modern defenders
156 A.D.M. Walker, Political Obligation and the Argument from Gratitude, 17 PHIL. & PUB. AFF. 191–211 (1988); Walker, Obligations of Gratitude and Political Obligation, 18 PHIL. & PUB. AFF. 359–364 (1989); TERRANCE MCCONNELL, GRATITUDE (1993), at 180–208.
who have in turn been subjected to criticism.
157 George Klosko, Political Obligation and Gratitude, 18 PHIL. & PUB. AFF. 352–338 (1989); Klosko, Four Arguments against Political Obligation from Gratitude, 5 PUB. AFF. Q. 33–48 (1991); Klosko, Fixed Content of Political Obligation, 46 POL. STUD. 53–67 (1998); Christopher H. Wellman, Gratitude and Political Obligation, 99 APA NEWSL. ON PHIL. & L. 71–77 (1999).
Duties of gratitude are associative in the sense that they may be owed despite there having been no interchange between the obligor and obligee and no voluntary assumption by the obligee. What is minimally required is something done by the obligor. I owe a debt of gratitude to, say, my rich uncle for paying for my private schooling and I owe this debt despite my having had no say in my being privately schooled and despite the fact that the benefits of my private education are ones I would have preferred not to receive (just as I owe a debt of gratitude to the fellow subway rider who hands me the bag of orange peels I have been trying to lose). I may owe a debt of gratitude willy-nilly, but I do not owe it merely in virtue of the humanity I share with my benefactor. My benefactor has done something which, whether I happen to like it or not, has imposed a duty upon me.
Insofar as the duty to obey can be cast as an associative duty, it can be particularized as owed to a limited set of obligors—whether one's nation or conationals—and it is universally borne in the sense that all of those who have relevantly benefited owe the duty, whether or not they have sought or desired the benefit. Therefore the duty to obey, qua a duty of gratitude, seems positioned to avoid both the particularity worry thought to taint natural-duty accounts and the acceptance problem that frustrates volitional theories. But Wellman158
summarizes a host of objections attending any gratitude account. The first four derive from Simmons:
159 (1) duties of gratitude are too vague to justify the relatively crisp duties the law imposes;
160 See also McConnell, supra note 156, at 206–208; Klosko, Fixed Content, supra note 157, at 54–57.
(2) duties of gratitude cannot flow from unwanted benefits, and so a duty to obey founded upon them cannot be universally borne and comprehensive; (3) duties of gratitude are not owed for routine, tax-financed services; and (4) duties of gratitude are oweable only to persons rather than institutions such as the state. Klosko adds that duties of gratitude lack the stringency of the duty to obey.
161 Klosko, Political Obligation, supra note 157.
Wellman piles on two further objections: that gratitude is never owed to a coercive agent such as the state, and that in any case gratitude is a matter not of duty at all but of virtue. To this impressive list might be added a final doubt about the immunity of a gratitude account to the particularity worry: It is believable, for example, that Canadians
ought to be grateful for U.S. protection even if it is not strictly speaking
unfair of Canadians to enjoy its benefits without its burdens.
Walker has rightly pointed out that gratitude may indeed be owed to an institution as distinct from an individual agent (as college development offices well know), and that gratitude may be owing even to those who have done a job (well) and been (well) paid for it (“Thank you, Paine Webber!”).162
Walker, Political Obligation, supra note 156; Walker, Obligations of Gratitude, supra note 156.
The stringency objection is perhaps manageable by the standard retreat to the pro tanto.
163 McConnell, supra note 156, at 203–208.
Wellman's claim that gratitude is never owed for coerced benefits is overdrawn: The drunk should be grateful, perhaps, to the friend who impounds his car keys, and in any event the received view that the state is inherently coercive is at least controversial.
164 Edmundson, supra note 24.
Finally, it is possible to accommodate the aretaic aspects of gratitude without altogether displacing its deontological ones. Gratitude is in this respect like beneficence, whose status as a virtue leaves room for duties of beneficence, such as those “easy rescues” featured in the literature on consequentialism.
What remains unclear is whether obedience is ever required as an expression of gratitude, absent some quasi-contractual relationship between benefactor and beneficiary, or other outré background (“Do as I say or I'll kill your benefactor”). If you, unbidden, squeegee my windshield, I may owe you a duty to say thanks, but I do not have a duty to do whatever you demand, even if what you demand is very important to you. Walker argues that gratitude requires more than expressing appreciation to one's benefactor but also “to avoid harming him or acting contrary to his interests.”165
Walker, Political Obligation, supra note 156, at 202.
On the further assumption that “noncompliance with the law is contrary to the state's interests,” it follows, he says, that “every citizen has an obligation of gratitude to comply with the law.”
166 So stated, the argument not only fails to acknowledge the commonplace occurrence of harmless lawbreaking but also—and more seriously—fails to face up to the fact that there is a radical difference between giving special weight to important interests of a benefactor and treating the directives of a benefactor as authoritative.
167 McConnell, supra note 156, at 206–207; Klosko, Four Arguments, supra note 157.
Liberal Associativism
Associative accounts other than fair-play accounts have been widely discussed in the last two decades. Ronald Dworkin dismissed fair-play accounts in Law's Empire as both vulnerable to Nozickian counterexamples and as failing to explain how fair treatment can impose a reciprocal duty. Dworkin went on to claim that “a state that accepts integrity as a political ideal has a better case for legitimacy than one that does not.”168
Dworkin, supra note 20, at 191–192.
The details of Dworkin's conception of integrity cannot be explored here, but for present purposes it will suffice to characterize integrity as a kind of community-building conception of equality. As a lemma in the argument for the superiority of law as integrity over its rivals, Dworkin hypothesized that “political obligation—including an obligation to obey the law—is a form of associative obligation.”
169 Dworkin's positive account of the duty to obey the law is confined to
Law's Empire170 and is (even there) less developed than his account of judicial obligation. It is, moreover, somewhat desultory and impressionistic—as well as confessedly tentative. Nonetheless, it has attracted extensive commentary. Green interprets Dworkin's theory as representing the duty to obey as underived or “parthenogenetic,”
171 Green, supra note 5, at 5.
in that it regards political association as “like family and friendship…pregnant of obligation.”
172 Dworkin, supra note 20, at 206.
Nonetheless, Dworkin's view could equally be seen as a derived account that proves similar to fair-play theory in its associative version.
Associative obligations are obligations that arise from certain but not all relationships we have to one another. They are not volitional because they do not presuppose any undertaking (tacit or otherwise) on part of the obligee nor any prior action (willing or otherwise) nor any prior encounters or interaction between the parties bound (although they may be reinforced by such undertakings or encounters). They are in this sense “natural” (but not in Rawls's sense), and Dworkin at one point refers to them as such.173
Hence they are not vulnerable to the “no agreement” objection directed against consent theories. At the same time, associative obligations seem to avoid the particularity problem, in that associative obligations arise from focused groupings and so, although not presupposing any voluntary act, are not general in the sense of being owed to all other humans as such. They are “special responsibilities social practice attaches to membership in some biological or social group.”
174 Dworkin cites the following examples: family, lovers, friends, neighbors, fellow union members, coworkers, commercial partners and joint enterprises (insofar as they become a “fraternal association” over and above a mere “long-standing contractual relationship”), academic faculties, and (some) armies.
Associative obligations thus do not arise from merely factitious groupings. At a minimum there must exist what he terms a “bare” community: “a community that meets the genetic or geographical or other historical conditions identified by social practice as capable of constituting a fraternal community.”175
He offers two examples in which “bare” community is lacking: in one, the citizens of Fiji become disposed to treat Dworkin as a fellow Fijian; in another, a stranger on an airplane decides to befriend Dworkin. Although Dworkin is not explicit on the point, the reason why he would not count himself as a Fijian nor as his fellow passenger's friend is not that he has not assumed and perhaps would not welcome the association but that there is no “social practice” of counting fellow passengers as automatic friends or of counting “honorary” citizens as citizens. But bare communities do not support associative obligations unless they meet four further conditions: the obligations they purport to impose must exhibit a concern among their members that is “special, personal, pervasive, and egalitarian.”
176 Dworkin explains:
First, [the members] must regard the group's obligations as…holding distinctly within the group, rather than as general duties…owe[d] equally to persons outside…. Second, they must accept that these responsibilities…run directly from each member to each other member, not just to the group as a whole…. Third, [they] must see these responsibilities as flowing from a more general responsibility each has of concern for the well-being of others in the group…. Fourth, [they] must suppose that the group's practices show…an equal concern for all members.177
If the train of psychological verbs—“regard,” “accept,” “see,” and “suppose”—pertained to a feeling or sense of common concern among group members, an obvious difficulty for Dworkin would be that the psychological conditions needed to elevate bare communities to the status of true, obligation-supporting communities are unlikely to be universally satisfied by citizens of modern nation-states. Dworkin is alert to the difficulty but denies that the four conditions are psychological:
These are not psychological conditions. Though a group will rarely meet or long sustain them unless its members by and large actually feel some emotional bond with one another, the conditions do not themselves demand this. The concern they require is an interpretive property of the group's practices of asserting and acknowledging responsibilities—these must be practices that people with the right level of concern would adopt—not a psychological property of some fixed number of the actual members.178
What is meant here can perhaps be approached by considering the case of an isolated community of dour individualists who—for purely prudential reasons—maintain a volunteer fire department. Suppose that they severally expect each other to assume the burdens of serving as firemen whenever the need may arise. Although none of the individualists satisfies the third of the four conditions, their practices are interpretable as ones that persons who did have the right level of concern—that is, who all satisfied all the four conditions as a matter of psychological fact—would have adopted. And thus their obligations are associative.
This might remain true after we alter the case by peopling the community not with individualists but with good samaritans who as a matter of psychological fact feel no special concern for other members of the community but an unbounded beneficence toward all humanity. Here, the members fail the first and perhaps the third conditions. Nonetheless, if their practices are interpretable as ones that persons satisfying all four conditions would adopt, the group's obligations are associative. Finally, if we imagine a mixed community of individualists and good samaritans, we reach the same result—that the obligations of service are associative—even though the motivating psychology varies drastically from person to person.
Does this way of understanding the conditions make it plausible to think that associative obligations “scale up” from small, closely knit groups to the vast and anonymous plane of modern political life? At first glance, Dworkin's invocation of hypothetical concern parallels the mechanism by which consent theorists have tried to avoid the objection that insufficiently many within the state's territorial grasp have in fact consented to obey its every command. A hypothetical consent theory ties legitimacy to what properly rational and fully informed persons would consent to. But the parallel would be misleading, for Dworkin is not rendering as obligatory those practices that “people with the right level of concern would adopt” but, rather, he renders as associative obligations those obligation-assigning practices that are interpretable as reflecting a concern satisfying the four conditions.
This, I suggest, is the best way to read Dworkin's view. But even if it were granted that associative obligations exist as a distinct genus and that they are capable of being scaled up to the size of modern states, there would remain the further question of whether political obligation can be defended as an obligation grounded in fraternal concern.179
Leslie Green, Associative Obligations and the State, in LAW AND COMMUNITY: THE END OF INDIVIDUALISM? 93–118 (Allan C. Hutchinson and eds., 1989); Wellman 1997; Simmons 1996b.
Obligations of gratitude exist, to take a comparison case, and we may for the moment suppose that such obligations exist not only on a person-to-person scale but on a political scale. We may assume, for example, that Americans owe a debt or obligation of gratitude to the Founding Fathers or to the “Greatest Generation” or the New York Fire Department. It is a further and very difficult question as to what this debt comes to; as noted in the section above, there is a gap between gratitude and obedience. The gap remains if we substitute “concern” for “gratitude”: What reason is there to suppose that obedience is the only, the best, or even a candidate way to express my concern or my gratitude?
180 Green, supra note 179, at 100–102.
If the answer on Dworkin's behalf is that
equal concern so requires, his account would appear to be a somewhat tortuous restatement of fair-play theory (which he officially dismisses as requiring—as per Rawls and Simmons—voluntary acceptance of benefits).
Dworkin says we “have no difficulty in describing the main obligations associated with political communities. The central obligation is that of general fidelity to law, the obligation that political philosophy has found so problematic.”181
Dworkin supra note 20, at 208.
But “general” here is ambiguous. Does it mean a duty to obey the law qua law? Or is it a duty to obey the law to the extent that we are seriously expected to obey the law? Or is it a duty not to interfere with the law's administration? Dworkin's assurance that “justice will play its normal interpretive role in deciding for any person what his associative responsibilities, properly understood, really are”
182 could be taken by a philosophical anarchist as an invitation to read the social practice in a voluntaristic way (as though, given the voluntaristic rhetoric of American legal doctrine, any further invitation were needed in this country). Dworkin's characterization of the citizen's engagement with law as a protestant “conversation with one self”
183 makes it doubtful whether a content independence is among the characteristics Dworkin can consistently attribute to the duty to obey the law.
184 William A. Edmundson, Stephen Guest, Ronald Dworkin, 104 ETHICS 394–396 (1994).
D. Mixed Accounts
The repair of many theorists to mixed views does not necessarily signify a mood of desperation on the part of apologists for the duty to obey; David Ross's account of the duty to obey the law was a mixed one, combining duties of gratitude, fidelity, and beneficence. Mixed views may, however, have greater difficulty capturing the authoritativeness of law. If, for example, one takes the position that law is legitimately authoritative in virtue of its doing a better job of getting its subjects to act on reasons that apply to them,185
Raz, supra note 30, at 53–57.
one must be prepared to explain law's superior competence with respect to those reasons. It seems doubtful that the law has a generally superior competence with respect to acting upon reasons or upon moral reasons,
186 Raz, supra note 112; Hurd, supra note 62.
and so, understandably, various theorists have emphasized law's importance in overcoming coordination problems.
187 Honoré, supra note 114; Finnis, supra note 109.
It is widely accepted that law can indeed solve social coordination problems, and a schematic mixed view would combine an account of a set of goods requiring coordination with a fair-play principle to bind all of those who receive goods of that set. But law claims a more wide-ranging authority than this.
188 Accordingly, a further admixture will be needed to legitimate the law's claim to general-purpose authority—and its claim to impose a duty that is comprehensively applicable—if such legitimation seems desirable.
Samaritanism
Christopher Wellman's approach is an example of the strengths and weaknesses of a mixed theory.189
Christopher Wellman, Toward a Liberal Theory of Political Obligation, 111 ETHICS 735–759 (2001a).
Wellman begins with a standard Hobbesian account of the benefits a legal order secures, but his emphasis is on its benefits to others rather than to the obligor and is in that sense a samaritan account. Just as one may justifiably commandeer a vehicle to take an accident victim to the emergency room, the state may justifiably coerce citizens in order to “rescue all within the state's borders from peril”—that is, the peril of the Hobbesian war of all against all.
190 A samaritan account of the moral permissibility of the coercive state
lays claim to a number of advantages. Samaritanism is arguably a more
general and plausible candidate than a freestanding, underived, natural
duty to obey.191
Id.; George
Klosko, Samaritanism and Political Obligation: A Response to
Christopher Wellman's “Liberal Theory of Political
Obligation.” 113 ETHICS 835–840
(2003).
Like an appeal to benefit or necessity, it “does
not depend upon any prior action, agreement, or contract” on the
part of citizens and subjects.
192 Wellman, supra note 189, at 747.
A
samaritan account thus avoids the “descriptive” failing of
consent or social-contract accounts (and, one might add, it avoids also
the strained disingenuousness of hypothetical-consent theories, which
endow “should agree” with the same normative power as
“did agree”). Moreover, Wellman says, a samaritan account
avoids the taint of paternalism that an appeal to what is for the
citizen's or subject's
own good would have, in
liberal eyes. (Although we may doubt that the taint disappears
entirely—coercing me for the good of my soul is as objectionably
paternalistic to a scrupulously individualistic liberal as any other
coercion, and perhaps more so. Making me do my duty surely benefits me
to the extent that it makes it the case that my duty is done and I am
relieved of moral debt I might otherwise owe for not doing it. This
point is of a piece with the standard libertarian and virtue-theoretic
complaints about “forced charity.”)
Granting that there is something to the analogy between government and samaritanism, at least in terms of a moral permission to employ coercion to avoid grave harm, what of the further step to a moral duty to obey? This step looks easy at first, because most of us will readily admit that samaritan duties exist, as well as samaritan permissions, at least under the usual proviso that the duty is not unreasonably burdensome; and the analogy between samaritanism and government seems sturdy enough to support the passage from samaritan duty to provide an easy rescue to a citizen's duty to obey the law. But on second look, complications arise. Disobedience is often inconsequential, and particularly so with respect to the state's mission to rescue others from the perils and inconveniences of a state of nature. Wellman's treatment is to complement samaritanism with a fair-play duty along the line initiated by Hart and discussed above. Disobedience violates the duty of fair play, even when it does not impair the state's rescue of us all from Hobbesian natural warfare: “each person has an obligation to obey the law as her fair share of this samaritan task.”193
Wellman's theory is thus a mixed account, and he argues that the combination is an improvement upon the two elements in isolation. Fairness theory alone is objectionable on at least two scores, in his view. Fairness theory is objectionably paternalistic, in that it is applicable only where the duty-bearer has been the net beneficiary of the cooperative scheme to which she is expected to conform. Fairness theory, second, reintroduces the consent-like element of acceptance (as opposed to mere receipt; see discussion above) of benefits and is therefore open to the by-now-familiar objection that too few have validly accepted the state's benefits. By folding fairness into a fundamentally samaritan theory, Wellman claims to have cleansed fairness theory of these two objectionable features. They are effaced because the fundamental theory—samaritanism—is not paternalistic and is not voluntaristic. “Combining fairness with samaritanism…enables us to reap the benefits of fairness without being saddled with either of the difficulties.”194
What is unclear, though, is why the difficulties vanish rather than communicate themselves to the combined view? Wellman does not explain. No one, as far as I am aware, has objected to Hart's fair play account on the ground that it is paternalistic—rather, objectors such as Nozick have argued that the fair-play account is implausible unless it is amended to assure that the duty-bearer is a net beneficiary from the duty-bearer's
own point of view. As far as the “receipt versus acceptance” point goes, it would be premature to conclude that Simmons's intuition pump is more powerful than Klosko's. Thus it may well be that the two alleged vices of fairness theory are less serious than Wellman supposes, but in that case it becomes difficult to decide whether Wellman is justified in offering his account as samaritanism with a fairness supplement rather than as a samaritan apology for fairness theory.
195 If, as Wellman claims, samaritanism rather than fairness is the driving wheel of his combined view, then (even supposing that the flaws of fairness theory vanish in the process of being combined) one must ask whether the combined view similarly sheds the notorious difficulties of samaritanism or—as it is perhaps more often known—a principle of beneficence. A duty to provide an easy rescue is widely admitted, but with the proviso that the rescue be easy and that the rescue be indeed a rescue from an extreme peril not otherwise avoidable. Far from all of our legal duties can plausibly be cast as counterparts of recognized samaritan duties. If the state were to collapse tomorrow, a plausible case could be made that it was everyone's samaritan duty to restore something like the legal system that had somehow suddenly vanished. But it would be highly implausible to argue that all would in those circumstances have a samaritan duty to restore and observe every jot and tittle of the laws that had previously been in force. In other words, rather few of our present legal duties are independently compelled by samaritan considerations. Wellman confesses that he does not know whether the “power of samaritanism,” applied to the problem of political obligation, “can justify more than a minimal, ‘night-watcher’ state.”196
Samaritan theory, by Wellman's own account, thus falls short of establishing a comprehensively applicable duty. Moreover, samaritanism standing alone furnishes no basis for preferring compatriots over noncompatriots. Thus even Wellman's qualified conclusion, that his samaritan account can “explain our obligation to obey the laws of minimal states”
197 is liable to the (perhaps valetudinarian) particularity worry associated with natural-duty accounts. Wellman notices this and concludes that particularity is “at most a
desideratum rather than a requirement.”
198 Wellman, supra note 78, at 115.
Other Mixed Accounts
George Klosko has proposed a mixed account, similar to Wellman's, which combines associative fairness with a principle of beneficence and an appeal to the common good.199
Similarly, David Lefkowitz sketches a combination of fairness and moral necessity.
200 Lefkowitz, supra note 27.
David Miller's mixed account combines notions of citizenship and of nationality.
201 Space does not permit discussing these proposals in detail.
V. CONCLUSION
Where does political obligation lie? Rawls's mature theory classifies political obligation among natural duties. The Lockean tradition, in contrast, insists that political obligation must be understood voluntaristically. Associative obligations have been invoked as a middle way that avoids both the “no agreement” problem besetting volitional accounts and the particularity worry that many have recently thought exposes a major and even incurable defect of any natural-duty theory. But the associativist trial balloons have attracted withering fire, and it is unclear whether they are worth the trouble of reinflating—with the possible exception of the associative strain of fair-play theory. An associative fair-play principle seems best positioned to manage both the “no agreement” problem and the particularity worry, especially if a cosmopolitan treatment of the latter—along the lines Waldron suggests—is workable.
Such seems to be the state of discussion with respect to the duty qua universally borne. It is noteworthy that as the case for a universally borne duty is strengthened, the case for one that is comprehensively applicable and content-independent need not be advanced and in fact may be rendered more problematic. Mixed theories, by their mere eclecticism, may have an easier time reconstructing the comprehensive applicability of the duty. But such eclecticism seems to fudge the gap between law's moral permissibility and its obligatory force. Moreover, eclecticism seems to shirk the need to reckon with the content independence that is distinctive of legal authority.
Authority, conceived in terms of content-independent duty, is (no joke) powerful stuff. As is evident from the literature stemming from Joseph Raz's seminal exposition of the second-order “exclusionary” nature of norms, a defense of authoritative norms of any kind—legal or otherwise—will be subject to abrasive forces of the sort first loosed by William Godwin and lately applied by neo-Godwinian consequentialists, Dancy-style moral particularists, and rule-skeptics. These forces are powerful because the defense of authority is the defense of requiring what is often, in the circumstances of the particular case, contrary to the balance of reasons. Putting reason in its place is no small feat. The position that a just state of modern scope possesses a moral power to impose a content-independent duty to obey that is comprehensively applicable to whatever is law in the jurisdiction may be simply too ambitious to defend. To return to the set of propositions set out at the beginning of Section II, legitimate authority has, I think, to be decoupled from the full-blown, traditionally conceived duty of obedience. That is to say, the proposition that “Political authority is legitimate only if it imposes a general moral duty of obedience on those subject to it” must be given up. Instead, political authority must be reconceived as narrower moral power to impose duties to submit to legal processes. It should also be better appreciated that this moral power is one capable of vesting in trans- and supraterritorial agencies (such as, for example, the International Court of Justice), and is not confined—conceptually or practically—to territorial governments. The motive for this proposal is not merely that of avoiding difficulties that have not dissolved despite two and a half millennia of treatment, but more important, that of making sure that what is legitimated by the theory of political authority corresponds to the still-evolving conditions in which we find ourselves.
