Justification and legitimacy

Ethics

Volumn 109, Issue 4, 1999, Pages 739-771

  Simmons, A John ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0033162124     PISSN: 00141704     EISSN: None     Source Type: Journal    
DOI: 10.1086/233944     Document Type: Article

References (85)

1. Though it will become obvious later, I should issue here a preliminary warning that the accounts identified as Lockean and Kantian will depart from the actual positions of Locke and Kant in many ways. I seek only to identify the broad "spirit" of these accounts with those of the historical philosophies in question. There are, of course, many other prominent accounts of these matters that are not centrally discussed, or discussed at all, in this article. I take the Kantian account to be the most influential in contemporary political philosophy (though not in contemporary political science) and the Lockean account to be its clearest and most persuasive rival. For that reason I concentrate here on those two accounts. But it is not at all difficult to find in contemporary philosophical literature rival views of the ground of legitimacy claims (though it is harder to find rival accounts of the meaning of such claims). Ronald Dworkin has suggested to me that the natural competitor for the Lockean view is not the Kantian one but, rather, the (older) view that explains justification, legitimacy, and political obligation simply in terms of our having been born and raised in a particular (acceptably just) political community. I have tried to express my skepticism about that approach in "Associative Political Obligations," Ethics 106 (1996): 247-73. Another familiar rival view grounds legitimacy in a Hobbesian (rather than a Kantian) version of a hypothetical contract. And Harry Brighouse has identified a "widely shared . . . liberal" conception of legitimacy that requires both the satisfaction of hypothetical consent standards and the actual (free, authentic) consent of at least a majority of subjects ("Civil Education and Liberal Legitimacy," Ethics 108 [1998]: 719-45, pp. 720-21).
2. Though it will become obvious later, I should issue here a preliminary warning that the accounts identified as Lockean and Kantian will depart from the actual positions of Locke and Kant in many ways. I seek only to identify the broad "spirit" of these accounts with those of the historical philosophies in question. There are, of course, many other prominent accounts of these matters that are not centrally discussed, or discussed at all, in this article. I take the Kantian account to be the most influential in contemporary political philosophy (though not in contemporary political science) and the Lockean account to be its clearest and most persuasive rival. For that reason I concentrate here on those two accounts. But it is not at all difficult to find in contemporary philosophical literature rival views of the ground of legitimacy claims (though it is harder to find rival accounts of the meaning of such claims). Ronald Dworkin has suggested to me that the natural competitor for the Lockean view is not the Kantian one but, rather, the (older) view that explains justification, legitimacy, and political obligation simply in terms of our having been born and raised in a particular (acceptably just) political community. I have tried to express my skepticism about that approach in "Associative Political Obligations," Ethics 106 (1996): 247-73. Another familiar rival view grounds legitimacy in a Hobbesian (rather than a Kantian) version of a hypothetical contract. And Harry Brighouse has identified a "widely shared . . . liberal" conception of legitimacy that requires both the satisfaction of hypothetical consent standards and the actual (free, authentic) consent of at least a majority of subjects ("Civil Education and Liberal Legitimacy," Ethics 108 [1998]: 719-45, pp. 720-21).
3. James Sterba, Contemporary Social and Political Philosophy (Belmont, Calif.: Wadsworth, 1995), p. 1.
4. I am discussing here only practical (including political) justification, but I believe this point about the "defensive" nature of justification holds as well for, e.g., epistemic justification. Many other claims made here about justification, however, plainly do not apply (or apply in the same ways) to epistemic justification.
5. This terminology, along with some of the ideas in this section, are drawn from my "Original-Acquisition Justifications of Private Property," Social Philosophy & Policy 11 (1994): 63-84.
6. Both optimality and permissibility justifications, of course, may involve acknowledging that an act or arrangement can be morallyjustified even if it is suboptimal or impermissible in certain respects. Moral justifications are typically "all things considered" justifications: to show that an act or arrangement is justified (optimal or permissible) is to show that it is best or good enough on balance. Thus, I may maximize utility, performing the only act that classical maximizing utilitarianism regards as justified, even if in doing so I must cause some disutility (e.g., I may push the drowning swimmer away from the overloaded lifeboat). Or my act may be on balance permissible (on a satisficing version of utilitarianism, a rule utilitarianism, or on some deontological moral theory), even if it involves some elements that might otherwise be impermissible, provided that its other elements are sufficiently morally positive to overbalance these negative elements (e.g., I may be justified in breaking my promise to meet you for lunch in order to help a sick friend).
7. I do not pretend to any serious Hobbes scholarship with this remark, and it may reasonably be contended that for Hobbesians "only a certain kind of state is justified" (Thomas Christiano, "The Incoherence of Hoboesian Justifications of the State," American Philosophical Quarterly 31 [1994]: 23-38, p. 26). But Hobbes at least appears to believe that (a) the presence of true sovereignty defines the state, (b) all true sovereignty is necessarily absolute, and (c) absolute sovereignty is always justified (i.e., preferable to all nonstate alternatives). It would seem to follow that all real states (of whatever form) are justified. If Hobbes means only that peace is preferable to (active) war, then he is surely correct. But it isobviously true neither that life in a bad state precludes active war between the sovereign and at least some of his subjects nor that life in the state of nature necessarily involves a constant or active war of all against all. If Hobbes means only that any state must be better than the (real or imagined) anarchy of the English Civil War, his claim both seems possibly false and to constitute no justification of the state (i.e., all states) at all. We cannot justify an institution by showing it to be at least preferable to the current situation or to the worst of the other imaginable possibilities. The best discussion of these questions, I think, remains Gregory Kavka, Hobbesian Moral and Political Theory (Princeton, N.J.: Princeton University Press, 1986), pt. 1.
8. I do not pretend to any serious Hobbes scholarship with this remark, and it may reasonably be contended that for Hobbesians "only a certain kind of state is justified" (Thomas Christiano, "The Incoherence of Hoboesian Justifications of the State," American Philosophical Quarterly 31 [1994]: 23-38, p. 26). But Hobbes at least appears to believe that (a) the presence of true sovereignty defines the state, (b) all true sovereignty is necessarily absolute, and (c) absolute sovereignty is always justified (i.e., preferable to all nonstate alternatives). It would seem to follow that all real states (of whatever form) are justified. If Hobbes means only that peace is preferable to (active) war, then he is surely correct. But it isobviously true neither that life in a bad state precludes active war between the sovereign and at least some of his subjects nor that life in the state of nature necessarily involves a constant or active war of all against all. If Hobbes means only that any state must be better than the (real or imagined) anarchy of the English Civil War, his claim both seems possibly false and to constitute no justification of the state (i.e., all states) at all. We cannot justify an institution by showing it to be at least preferable to the current situation or to the worst of the other imaginable possibilities. The best discussion of these questions, I think, remains Gregory Kavka, Hobbesian Moral and Political Theory (Princeton, N.J.: Princeton University Press, 1986), pt. 1.
9. John Locke, Second Treatise of Government, sec. 137.
10. Thus, Locke, say, would "justify the state" if he succeeded in his efforts to show that states ruled by limited governments (of the specified sort) are noncomparatively unobjectionable and preferable to even the best state of nature in which we could reasonably hope to live. Hobbes, presumably, would agree with Locke that states which in fact satisfied these standards were "justified" (though neither, of course, prominently uses this language of "justification").
11. "To justify an institution is, in general, to show that it is what it should be or does what it should do" (David Schmidtz, "Justifying the State," in For and Against the State, ed. John T. Sanders and Jan Narveson [Lanham, Md.: Rowman & Littlefield, 1996], p. 82).
12. Locke would probably not have identified his opponent as "anarchists," though the language of "anarchy" dates from the mid-sixteenth century and had become relatively common by die mid-to late seventeenth century when Locke was writing. Many radical tracts and familiar positions of Locke's own day - from the pre - Civil War period to the Glorious Revolution - were recognizably anarchist in tone or substance.
13. We should distinguish between versions of anarchism that in this way deny the state's moral standing on a priori grounds - no possible state can be justified - and those that only reject the state a posteriori, on account of the contingent character of actual states - no existing state is justified. Though most anarchists in one of these ways deny the justification of the state, I have argued elsewhere that the central unifying thesis of all forms of anarchism is in fact rather an overarching denial of state legitimacy ("Philosophical Anarchism," in Sanders and Narveson, eds., pp. 19-39).
14. Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), pp. 4-5.
15. See Bernard Williams, "The Minimal State," in Reading Nozick, ed. Jeffrey Paul (Totowa, N.J.: Rowman & Littlefield, 1981), p. 33.
16. Nozick does not ever clearly explain his position on this question. So Schmidtz, e.g., understandably maintains that Nozick's attempted hypothetical, invisible-hand, "emergent" justification of the minimal state couldn't really justify the state in any meaningful sense at all (Schmidtz, pp. 93-94). But I believe that Nozick's argument, if successful, would show at least that, according to his Kantian/Lockean right-centered conception of morality, the minimal state is morally justified. If it is logically and physically possible that a state arise and operate without violating anyone's rights, and if such a state would be rationally preferable to nonstate alternatives, then the anarchist objection is rebutted and (in that sense) the state (i.e., that particular kind of state) is justified. See Alan Nelson, "Explanation and Justification in Political Philosophy," Ethics 97 (1986): 154-76, p. 171.
17. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
18. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
19. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
20. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
21. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
22. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
23. Legitimate states have a special, unique right to be the state that operates in the territory (Nozick, p. 134). A state can acquire such legitimacy, according to Nozick, by having its operations consented to by enough residents of its (claimed) territory that it possesses the "greatest entitlement" (in that territory) to exact punishment for wrongdoing (ibid., pp. 108-14, 132, 139-40). So showing a state to be legitimate involves showing that it actually has (or has had) certain kinds of morally unobjectionable relations with those it controls; justifying the state only involves showing, against the anarchist, that it is possible for a state to have such relations and that having states at all is advantageous (so that we would expect them to arise naturally from within a state of nature). Notice that on this model a particular state could apparently be the sort of state that might be justified - it could (in Nozick's case) be a minimal state that provides protection to all within its territories and that performs no redistributive functions - but still not itself be a legitimate state. If, for instance, a minimal state were imposed on a people entirely by force, without any (or many) consenting "clients," it would have no greater right to enforce justice than would some group of its subjects; it could not then be legitimate, even if it did offer protection to all and operate without redistributing holdings. The state's special legitimacy arises from the fact that its consenting clients give it a greater share of the collectively held "right to punish" than is held by any of its competitors (e.g., individual nonclients, cooperative associations of allied nonclients, rival protective agencies [ibid., pp. 139-40]). A minimal state imposed from without would presumably lack such legitimacy and so have only a de facto monopoly on the use of force. Particular states are legitimate in virtue of the actual history of their relations with their subjects, relations that establish the state's right to rule and the subjects' obligation to comply. So we might use a different language and say that such accounts distinguish between the general justification for having political authority and the specific justification for a particular authority's being the authority. For related distinctions, see Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984), pp. 6-7; David A. J. Richards, "Conscience, Human Rights, and the Anarchist Challenge to the Obligation to Obey the Law," Georgia Law Review 18 (1984) 771-89, p. 781; and Harry Beran, "What Is the Basis of Political Authority?" Monist 66 (1983): 487-99, pp. 489-90, and The Consent Theory of Political Obligation (London: Croom Helm, 1987), pp. 12-13. Richard Taylor also distinguishes between general justifications of certain kinds of states, by reference to their good purposes or ends, and demonstrations of the legitimacy of a particular state. But he, unlike the others considered here, proposes a purely positivist criterion for legitimacy. See Freedom, Anarchy, and the Law (Englewood Cliffs, N.J.: Prentice-Hall, 1973), pp. 86-105.
24. Locke, sec. 95.
25. It is important to see that, on this Lockean model, showing that the limited state is justifiable may give us reasons to consent to a particular limited state's rule. But this implies neither that such consent can simply be assumed nor that actual consent is not necessary for a state's legitimacy and for its subjects' obligations. In Hanna Pitkin's well-known reading of Locke ("Obligation and Consent - I," American Political Science Review 59 [1965]: 990-99, pp. 995-97, 999), she (mistakenly, in my view) takes (what I here call) his general justification of the state to amount to a move by Locke to replace actual consent with hypothetical consent as the relevant standard of legitimacy.
26. Though Locke is far from clear on this point, it seems likely that the limited state's justification is intended by him to be a necessary condition for the legitimation (by actual consent) of any particular limited state's rule. Consent is necessary - but not sufficient - for legitimacy and political obligation, (in part) because the justification of a type of state is necessary for consent to a token of that type to be binding. We cannot bind ourselves by consent to immoral arrangements. And to this Locke adds that "no rational creature can be supposed to change his condition with an intention to be worse" (Locke, sec. 131). Locke appears to take this latter claim to imply that binding (rational) consent can only be given to states that are demonstrably superior on prudential grounds to (or at least as good as) the state of nature (i.e., that consent given to states whose kinds are not justified in this broad sense cannot bind us). A state must be on balance morally acceptable and a "good bargain" for our consent to succeed in legitimating it. For a defense of this reading of Locke, see my On the Edge of Anarchy: Locke, Consent, and the Limits of Society (Princeton, N.J.: Princeton University Press, 1993), chap. 5.
27. See, e.g., R. P. Wolff, In Defense of Anarchism (New York: Harper & Row, 1970), pp. 3-4.
28. State legitimacy, according to this conception of it, includes an exclusive power over subjects to impose duties and enforce them coercively, which correlates with obligations on others to refrain from these tasks. It also includes a right, held against subjects, to be obeyed (i.e., to have any imposed duties discharged). This latter right is the logical correlate of subjects' political obligations. The correlativity of political legitimacy and political obligation has been denied in a number of recent works, most notably in Kent Greenawalt, Conflicts of Law and Morality (New York: Oxford University Press, 1989), chap. 4. But while Greenawalt offers convincing arguments for a noncorrelativity thesis, I think his arguments in fact support only the noncorrelativity of (what I am here calling) the state's justification with political obligation.
29. For a defense of this reading of Locke, see my On the Edge of Anarchy, chap. 3.
30. See, e.g., Robert Ladenson, "In Defense of a Hobbesian Conception of Law," in Authority, ed. J. Raz (New York: New York University Press, 1990), pp. 36-37: "The right to rule is . . . a justification right . . . [which] by itself implies nothing about either the subject's duty of allegiance to the state or of compliance with the law"; and Christopher Wellman, "Liberalism, Samaritanism, and Political Legitimacy," Philosophy & Public Affairs 25 (1996): 211-37, pp. 211-12): vAn account of political legitimacy explains why this coercion [i.e., punishment of those within the state's borders] is permissible. . . . It is crucial to notice that political legitimacy is distinct from political obligation."
31. See, e.g., Robert Ladenson, "In Defense of a Hobbesian Conception of Law," in Authority, ed. J. Raz (New York: New York University Press, 1990), pp. 36-37: "The right to rule is . . . a justification right . . . [which] by itself implies nothing about either the subject's duty of allegiance to the state or of compliance with the law"; and Christopher Wellman, "Liberalism, Samaritanism, and Political Legitimacy," Philosophy & Public Affairs 25 (1996): 211-37, pp. 211-12): vAn account of political legitimacy explains why this coercion [i.e., punishment of those within the state's borders] is permissible. . . . It is crucial to notice that political legitimacy is distinct from political obligation."
32. Recent international law has tended to treat the persecution of racial, ethnic, or religious groups within their territories as delegitimating states. See, e.g., Allen Buchanan, "Theories of Secession," Philosophy & Public Affairs 26 (1997): 31-61, pp. 50-51.
33. Charles Taylor, "Alternative Futures: Legitimacy, Identity, and Alienation in Late Twentieth Century Canada," in Communitarianism: A New Public Ethics, ed. M. Daly (Belmont, Calif.: Wadsworth, 1994), p. 58.
34. See C. Taylor, "Legitimation Crisis?" in Philosophy and the Human Sciences: Philosophical Papers 2 (Cambridge: Cambridge University Press, 1985), pp. 248-88. Although Habermas is perhaps the best known of those who have discussed the possibility of a contemporary "legitimation crisis" (in terms of the "contradictions" that weaken advanced capitalism) (e.g., in his Legitimation Crisis [Boston: Beacon Press, 1975]), he does not seem to accept the "attitudinal" sense of legitimacy that is being discussed here. If I understand him correctly, Habermas's notion of legitimacy corresponds to neither of Taylor's senses.
35. See C. Taylor, "Legitimation Crisis?" in Philosophy and the Human Sciences: Philosophical Papers 2 (Cambridge: Cambridge University Press, 1985), pp. 248-88. Although Habermas is perhaps the best known of those who have discussed the possibility of a contemporary "legitimation crisis" (in terms of the "contradictions" that weaken advanced capitalism) (e.g., in his Legitimation Crisis [Boston: Beacon Press, 1975]), he does not seem to accept the "attitudinal" sense of legitimacy that is being discussed here. If I understand him correctly, Habermas's notion of legitimacy corresponds to neither of Taylor's senses.
36. C. Taylor, "Alternative Futures: Legitimacy, Identity, and Alienation in Late Twentieth Century Canada," p. 58.
37. M. Weber, The Theory of Social and Economic Organization (London: William Hodge, 1947), p. 114, and Wirtschaft und Gesellschaft, 4th ed. (Tübingen: J. C. B. Mohr, 1956), pp. 23, 157.
38. M. Weber, The Theory of Social and Economic Organization (London: William Hodge, 1947), p. 114, and Wirtschaft und Gesellschaft, 4th ed. (Tübingen: J. C. B. Mohr, 1956), pp. 23, 157.
39. Tom R. Tyler, Why People Obey the Law (New Haven, Conn.: Yale University Press, 1990), p. 26.
40. A second familiar criticism of such attitudinal accounts of legitimacy is that they preclude the possibility of evaluating regimes according to objective standards - that is, they preclude judging regimes in terms of Taylor's second sense of legitimacy. As Hanna F. Pitkin has argued, "Weber in effect made it incomprehensible that anyone might judge legitimacy and illegitimacy according to rational objective standards" (Pitkin, Wittgenstein and Justice [Berkeley: University of California Press, 1972], p. 283). But this objection is hardly likely to impress theorists (like those who typically advance Weberian accounts of legitimacy) who believe that there are no "objective" standards according to which states or societies or regimes can be evaluated.
41. Seymour M. Lipset, "Social Conflict, Legitimacy, and Democracy," in Legitimacy and the State, ed. W. Connolly (New York: New York University Press, 1984), p. 88.
42. To be perfectly clear, I should emphasize at this point that nothing turns on whether we use the language of "justification" and "legitimacy" to identify the distinction with which I am concerned (or instead, say, reserve the term 'legitimacy' to identify one of the social scientific properties mentioned above). My interest is only in claiming that there is a distinction of some philosophical importance at issue here that can, I think, be happily captured by these terms and that traditional usage encourages us to express in this language.
43. Some readers of Locke (e.g., Pitkin) mistakenly conclude from his remarks on justification and legitimacy that Locke in fact embraced a hypothetical consent standard of justification and legitimacy. If we focus our attention on Locke's justification of the limited state, arguing that in his view this is a state we ought to give our consent to, we may think Locke's account can do without any reliance on actual consent. Legitimate states are just those that are good (i.e., justified - morally acceptable and a good bargain). But such a reading is forced to ignore all of Locke's explicit references to actual consent. This could only be a good interpretive move if there were no philosophical point to distinguishing one's arguments for state justification from those for state legitimacy. I try to show here that this distinction is important. So we hardly do Locke a favor by reading him (with Pitkin) as really wanting to collapse an important distinction that the text indicates he was in fact unwilling to collapse. See my On the Edge of Anarchy, sec. 7.2. Another very recent misreading of Locke, similar to Pitkin's, can be found in Jonathan Waskan, "De Facto Legitimacy and Popular Will," Social Theory and Practice 24 (1998): 25-56, p. 29, where it is claimed of Locke (as well as Hobbes and Rousseau) that for him "'legitimacy' is roughly synonymous with 'justified' or 'acceptable'."
44. Some readers of Locke (e.g., Pitkin) mistakenly conclude from his remarks on justification and legitimacy that Locke in fact embraced a hypothetical consent standard of justification and legitimacy. If we focus our attention on Locke's justification of the limited state, arguing that in his view this is a state we ought to give our consent to, we may think Locke's account can do without any reliance on actual consent. Legitimate states are just those that are good (i.e., justified - morally acceptable and a good bargain). But such a reading is forced to ignore all of Locke's explicit references to actual consent. This could only be a good interpretive move if there were no philosophical point to distinguishing one's arguments for state justification from those for state legitimacy. I try to show here that this distinction is important. So we hardly do Locke a favor by reading him (with Pitkin) as really wanting to collapse an important distinction that the text indicates he was in fact unwilling to collapse. See my On the Edge of Anarchy, sec. 7.2. Another very recent misreading of Locke, similar to Pitkin's, can be found in Jonathan Waskan, "De Facto Legitimacy and Popular Will," Social Theory and Practice 24 (1998): 25-56, p. 29, where it is claimed of Locke (as well as Hobbes and Rousseau) that for him "'legitimacy' is roughly synonymous with 'justified' or 'acceptable'."
45. See my Moral Principles and Political Obligations (Princeton, N.J.: Princeton University Press, 1979), pp. 32-34, 143-56.
46. For an account of the ways in which I think Locke should have dealt with the issue of tacit consent, see my "'Denisons' and 'Aliens': Locke's Problem of Political Consent," Social Theory and Practice 24 (1998): 161-82.
47. Immanuel Kant, The Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University Press, 1991), pp. 63, 85-87, 123-24. "Division" is the title of an unnumbered section of the work.
48. Kant, Perpetual Peace, in Kant: Political Writings, ed. H. Reiss (Cambridge: Cambridge University Press, 1991), p. 98n.
49. An obligation to enter (by consenting to) some civil society is presumably identical to neither an obligation to do what such obligatory consent would be consent to nor an obligation to do whatever one's particular society requires of its citizens. Thus, the "gaps" in Kant's argument may be sufficiently large to lead one to conclude that he was in fact simply uninterested in questions about legitimacy (in the Lockean sense), focusing entirely on questions of justification.
50. Thomas Nagel, Equality and Partiality (New York: Oxford University Press, 1991), p. 330. My emphasis throughout.
51. John Rawls, Political Liberalism (New York: Columbia University Press, 1993), p. 224; my emphasis.
52. Ronald Dworkin, Law's Empire (Cambridge, Mass.: Harvard University Press, 1986), pp. 190-91; my emphasis.
53. For instance: "The justification of authority . . . depends on one main argument. . . . The main argument for the legitimacy of any authority is that . . . a person is more likely to act successfully for the reasons which apply to him" (Joseph Raz, The Morality of Freedom [Oxford: Oxford University Press, 1986], pp. 70-71). "A state is legitimate only if, all things considered, its rule is morally justified" (Leslie Green, The Authority of the State [Oxford: Oxford University Press, 1990], p. 5). "A system of political authority or law can be legitimate, can be morally justified" (Jeffrey Reiman, In Defense of Political Philosophy [New York: Harper & Row, 1972], pp. 41-42). "To justify . . . coercive institutions, we need to show that the authorities within these institutions have a right to be obeyed and that their members have a corresponding duty to obey them. In other words, we need to show that these institutions have legitimate authority" (Sterba, p. 1). My emphases throughout.
54. For instance: "The justification of authority . . . depends on one main argument. . . . The main argument for the legitimacy of any authority is that . . . a person is more likely to act successfully for the reasons which apply to him" (Joseph Raz, The Morality of Freedom [Oxford: Oxford University Press, 1986], pp. 70-71). "A state is legitimate only if, all things considered, its rule is morally justified" (Leslie Green, The Authority of the State [Oxford: Oxford University Press, 1990], p. 5). "A system of political authority or law can be legitimate, can be morally justified" (Jeffrey Reiman, In Defense of Political Philosophy [New York: Harper & Row, 1972], pp. 41-42). "To justify . . . coercive institutions, we need to show that the authorities within these institutions have a right to be obeyed and that their members have a corresponding duty to obey them. In other words, we need to show that these institutions have legitimate authority" (Sterba, p. 1). My emphases throughout.
55. For instance: "The justification of authority . . . depends on one main argument. . . . The main argument for the legitimacy of any authority is that . . . a person is more likely to act successfully for the reasons which apply to him" (Joseph Raz, The Morality of Freedom [Oxford: Oxford University Press, 1986], pp. 70-71). "A state is legitimate only if, all things considered, its rule is morally justified" (Leslie Green, The Authority of the State [Oxford: Oxford University Press, 1990], p. 5). "A system of political authority or law can be legitimate, can be morally justified" (Jeffrey Reiman, In Defense of Political Philosophy [New York: Harper & Row, 1972], pp. 41-42). "To justify . . . coercive institutions, we need to show that the authorities within these institutions have a right to be obeyed and that their members have a corresponding duty to obey them. In other words, we need to show that these institutions have legitimate authority" (Sterba, p. 1). My emphases throughout.
56. Rawls, Political Liberalism, pp. 135-36, 137.
57. John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), p. 7. Rawls repeatedly claims that the basic structure of society, for the regulation of which the principles chosen by the original position contractors are designed, has two parts, the first of which is a political constitution (ibid., pp. 7, 61).
58. John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), p. 7. Rawls repeatedly claims that the basic structure of society, for the regulation of which the principles chosen by the original position contractors are designed, has two parts, the first of which is a political constitution (ibid., pp. 7, 61).
59. John Rawls, "Reply to Habermas," Journal of Philosophy 92 (1995): 132-80, pp. 144-45.
60. Ibid., p. 143.
61. John Rawls, "The Law of Peoples," in On Human Rights, ed. S. Shute and S. Hurley (New York: Basic Books, 1993), p. 64.
62. Rawls, "Reply to Habermas," p. 148. Nagel, similarly, suggests that the unanimous acceptance of a political arrangement that its legitimacy (and justification) requires "is neither actual unanimity among persons with the motives they happen to have, nor the kind of ideal unanimity that simply follows from there being a single right answer which everyone ought to accept because it is independently right, but rather something in between: a unanimity which could be achieved among persons in many respects as they are, provided they were also reasonable and committed within reason to modifying their claims, requirements, and motives in a direction which makes a common framework of justification possible" (Nagel, pp. 33-34).
63. Rawls's only explicit discussion of the idea of legitimacy (in his "Reply to Habermas") focuses principally on the legitimacy of governments (regimes) and laws, not on state legitimacy, and on the grounds of legitimacy, not the meaning of legitimacy. Accordingly, he there associates legitimacy with "lawfulness," claiming that the legitimacy of specific rulers and laws is a function of "their pedigree," of how they came to power or came to be in force ("Reply to Habermas," p. 175). But the legitimacy (lawfulness) of the pedigree depends in turn on whether or not the constitution that specifies the relevant procedures for determining adequate pedigree is just. The constitution must "be sufficiently just, even though not perfectly just, as no human institution can be that" (ibid.). It seems reasonable to conclude that state legitimacy turns, for Rawls, on the justice of the basic structure, as the passage cited in the text above suggests. Legitimate states are those that use their power according to the provisions of a just constitution. It thus seems to me that Rawls's accounts of justice (justification) and legitimacy are much more closely tied together than is suggested by Brighouse (p. 721). What exactly Rawls means by legitimacy - that is, what a state's legitimacy consists in - is less clear. But given his skepticism about citizens' political obligations (A Theory of Justice, pp. 113-14), obligations that correlate with (part of) the strong Lockean conception of legitimacy rights, it seems likely that Rawlsian legitimacy is only a liberty (or justification) right. This, of course, might explain why Rawls never makes explicit the form of his argument from justification to legitimacy. It may simply seem obvious to him that a justified state has a "justification right" to rule and use political coercion (though even this limited claim seems to me not at all obviously true, as my arguments in the text suggest).
64. Ibid., p. 13. See Rawls's similar remarks in Political Liberalism, pp. 135-37, 222.
65. Ibid., p. 13. See Rawls's similar remarks in Political Liberalism, pp. 135-37, 222.
66. Nagel, p. 36.
67. Indeed, Rawls's gestures toward voluntarism just seem inconsistent with the spirit of his project. For societies whose structures have been legitimated (in the Lockean sense) by the free, unanimous consent of their members may have quite illiberal shapes without thereby losing their legitimacy. Highly restrictive religious orders or extremely conservative agricultural communes, empowered by the free, informed consent of all members, could count as perfectly legitimate "societies" on the Lockean model.
68. Paradoxically, perhaps, such "individualist" changes might well also bring an increase in feelings of communal solidarity. Where we feel that we have genuinely chosen our place, we may be less likely to feel oppressed and alienated by aspects of the social world that are "given" and that seem immune to change. Choice is not the enemy of community (contrary to the suggestions of many communitarian thinkers). Indeed, such choice may be essential to both a community's vitality and its virtue (since if virtues must be voluntary, communitarian and republican emphases on artificial means of character formation may be self-defeating).
69. Rawls, Political Liberalism, pp. 137, 217.
70. Nagel, pp. 4, 17-18.
71. See, e.g., Christopher Bertram, "Political Justification, Theoretical Complexity, and Democratic Community," Ethics 107 (1997): 563-83, p. 568.
72. "Justification is argument addressed to those who disagree with us. . . . Being designed to reconcile by reason, justification proceeds from what all parties to the discussion hold in common" (Rawls, A Theory of Justice, p. 580).
73. Bertram, p. 574.
74. For Rawls, the "principle of legitimacy" has "the same basis as the substantive principles of justice" (Political Liberalism, p. 225).
75. I use the term 'transactional' here with the intention of ignoring one of its senses: that which conveys "negotiation" or "multilateral participation." In the sense I intend here, a "transaction" has occurred even where only one of the parties involved is active (e.g., where one party benefits another without the other's knowledge or participation).
76. It may be initially unclear how my distinction between generic and transactional evaluations in political philosophy relates to others, such as the distinction drawn by Schmidtz between (what he calls) "teleological" and "emergent" justifications. According to Schmidtz, these latter are the "two kinds of justification in political theory" (p. 81). A teleological approach "seeks to justify institutions in terms of what they accomplish," while "the emergent approach takes justification to be an emergent property of the process by which institutions arise" (ibid., p. 82). I believe Schmidtz's account of the two types of justification, as stated at least, is not an exhaustive classification. His conception of a teleological justification is insufficiently broad, since, unlike what I here call generic justification, it fails to include justifications that appeal to the moral virtues of states in consequence of which those states deserve support. The state (or some kind of state) may be justified by appeal to its virtues, or by appeal to what it might reasonably be expected to accomplish, even if, through bad luck (e.g., natural disaster or war) or lack of public support, say, it actually fails to accomplish much of value. There can be good reasons to support a state that may not be translatable into "accomplishments" by the state (as when people simply fail to act on those good reasons). Similarly, Schmidtz's notion of an "emergent" justification seems to me too narrow, since it includes only justifications that concern the state's origin. Appeals to actual consent, for instance - Schmidtz's paradigm of an emergent justification - may in fact be appeals to consent given over time, not just once (and for all) in "the process by which institutions arise"; and concerns about rights-violations later in the game, rather than at the state's origin, seem to be left out of Schmidtz's classification. My alternative suggestion is that we distinguish institutional evaluations in political philosophy according to whether they appeal to the state's general moral relations with its subjects conceived as a body, or instead to its particular relations with individual subjects.
77. Rawls, Political Liberalism, p. 49.
78. While I agree with little else in Michael Sandel's critique of liberal political philosophy, I think he is correct to raise the question (against Rawlsian - or what Sandel calls "minimalist" - liberalism) of "why the practical interest in securing social cooperation . . . is always so compelling as to defeat any competing moral interest." Sandel (rightly, in my view) argues that "it is not always reasonable to set aside competing values that may arise from substantive moral and religious doctrines" (Democracy's Discontent [Cambridge, Mass.: Harvard University Press, 1996], p. 19). Rawls at one point briefly considers the position of a Quaker pacifist ("Reply to Habermas," pp. 148-49) and argues that because Quakers support a constitutional regime and majority rule (as the best form of political association for those concerned with the rights and interests af all), their view is reasonable, and the decisions of their less pacifistic compatriots can be justified to them. But one who is an antistate pacifist, believing (not implausibly, I think) that modern states are by their very natures fundamentally opposed to pacifism, holds what Rawls seems to count as "unreasonable" views, so that no justification of state policy is owed to him.
79. Jean Hampton, e.g., justifying the state in a Hobbesian fashion, argues that "there are moral grounds for generating such a remedy [i.e., the state] because these problems [in the state of nature] have a severe negative impact on the well-being of other people. Moreover, in order to work, such a remedy must be collective in the sense that all or most people in a territory must . . . participate in it so that the warfare will end" (Political Philosophy [Boulder, Colo.: Westview, 1997], p. 73; second emphasis mine). But if most participating in the collective solution can solve the moral problem, then some opting out (on, say, individualist or pacifist grounds) is not necessarily wrong or "unreasonable."
80. Simmons, Moral Principles and Political Obligations.
81. See, e.g., Rawls, A Theory of Justice, secs. 18, 52; and Dworkin, pp. 192-94. Nagel does not deal with these arguments in Equality and Partiality.
82. See my Moral Principles and Political Obligations, chap. 6; and Dworkin, p. 193.
83. Hegel and contemporary Hegelians, e.g., argue, of course, for a sharp distinction between the rules governing "civil society" and those governing the State and so will be unimpressed by the fact that the rule I mention governs our relations with all nonpolitical institutions and arrangements. The state (with the family) is said to be "special" and governed by different rules, rather than being (as the Lockean claims) on one end of a uniformly governed continuum, along with other useful social arrangements. But Hegelian explanations of what is "special" about political arrangements seem to me either to rely on very obscure and dubious metaphysics or to simply beg the questions at issue (e.g., by without argument taking the well-conditioned beliefs of many in the state's natural authority as true and unsuspicious).
84. If all states are illegitimate, how important can questions about legitimacy be? The proper answer, I think, is that state legitimacy remains an important dimension of institutional evaluation because where states are legitimate with respect to persons, those states can justify acting (in the sense detailed below) in more restrictive fashions, and those person can justify less in the way of noncompliance and resistance than where states are illegitimate with respect to persons. A state's justification functions similarly to increase state options and decrease subject options, so that questions of justification would also remain important even if the anarchist were right that no existing states were justified.
85. Schmidtz has argued that my "claim that legitimizing the state requires a deliberate act but justifying it does not" in fact amounts to a retreat on my part from my proclaimed voluntarist ("emergent") justifications for political obligation and legitimacy to an unacknowledged "teleological" justification (p. 95). But what I discuss in the passages to which Schmidtz refers is not justifying the state, or even legitimating it, but only "a justification of government action": even if "a certain government does not have the right to command, its actions may nonetheless be morally justifiable" (Moral Principles and Political Obligations, p. 199; my emphasis). One can, I think, coherently support (as I do) voluntarist standards for legitimacy (and political obligation) - for demonstrating a general right to rule - while at the same time denying that the justification for particular actions or policies, of either states or individuals, is uniquely determined by the presence or absence of general state legitimacy (or political obligation).