¹ The recurrent “as such” is my device for sliding past the various ways in which the promotion of an agent's agent-relative value can have vicarious value for another agent.

² Although I have focused on agent-relative reasons that are reflective of agent-relative value, I have not excluded the possibility of an agent's having agent-relative reasons for or against performing certain actions without those reasons being reflective of the agent-relative value or disvalue of those actions for that agent.

³ This inference is no better than the inference from its not being an agent-relative reason to its being an agent-neutral reason.

⁴ A similar denial of the agent-relative character of deontic restraints can be found in the later sections of Kamm, Frances's “Non-Consequentialism, the Person as an End-in-Itself, and the Significance of Status,” Philosophy and Public Affairs, vol. 21, no. 4 (Fall 1992)Google Scholar. In the present essay, I provide a much more extended account of the genesis of this mistaken characterization of deontic restraints. I also suggest how an appeal to agent-relative value may undergird and motivate the invocation of persons' “status” as ends-in-themselves through which Kamm seeks to account for deontic restraints (see ibid., pp. 386–89). The thesis that deontic restrictions are “victim- rather than agent-focused” (ibid., p. 385) is anticipated in my “Moral Individualism: Agent-Relativity and Deontic Restraints,” Social Philosophy and Policy, vol. 7, no. 1 (Autumn 1989), p. 105.Google Scholar

⁵ Kamm is probably correct to deny that even these “duty-based” restrictions are agent-relative. She says, “I believe that duty-based theories are agent-focused but not agent-relative, in that the constraint supposedly arises from the content of what the agent is doing, not from his (versus others') doing it” (“Non-Consequentialism,” p. 382).Google Scholar

⁶ Thus, Scheffler's search for a rationale for deontic restrictions takes place explicitly within the context of assessing Robert Nozick's ascription to persons of rights that impose side-constraints on others' behavior. See Scheffler, Samuel, The Rejection of Consequentialism (Oxford: Clarendon Press, 1982), p. 82Google Scholar; and Nozick, Robert, Anarchy, State, and Utopia (New York: Basic Books, 1974), pp. 26–33Google Scholar. And, when he introduces the topic of deontic reasons, Nagel speaks of them as stemming from “the claims of other persons not to be maltreated in certain ways” and as a man's relative reasons “not to maltreat others himself, in his dealings with them (for example by violating their rights, breaking his promises to them, etc.).” Nagel, Thomas, The View from Nowhere (Oxford: Oxford University Press, 1986), p. 165.Google Scholar

⁷ Williams, Bernard, “A Critique of Utilitarianism,” in Utilitarianism: For and Against, Smart, J. J. C. and Williams, Bernard (Cambridge: Cambridge University Press, 1973), p. 115.Google Scholar

⁸ Ibid., p. 116.

⁹ Williams announces at the outset that his purpose is to offer a critique of utilitarianism that, instead of dealing with justice, will be “concerned with something rather different, integrity” (ibid., p. 82).

¹⁰ See, e.g., Williams, Bernard, “Utilitarianism and Moral Self-Indulgence,” in Contemporary British Philosophy, 4th series, ed. Lewis, H. D. (London: George Allen and Unwin Ltd., 1976), pp. 306–21.Google Scholar

¹¹ I have set aside a number of complicating features of the case. Among these are the fact that the one who Jim would kill would be among the twenty Pedro would otherwise kill and the fact that all twenty of the villagers are begging Jim to kill one of them. My guess is that Williams includes the latter feature in order to derail the thought that the key anti-utilitarian consideration is the injustice of Jim's killing one of the villagers. The problem for Williams is that if we really attend to this stipulated fact, then it becomes pretty clear that Jim ought to kill one of the villagers. His refusal to do so would merely be a matter of self-indulgence. For, given this authorization, the otherwise crucial deontic restriction against killing innocents will no longer apply.

¹² Scheffler, , The Rejection of Consequentialism, p. 102.Google Scholar

¹³ Ibid., p. 103.

¹⁴ Ibid., p. 102.

¹⁵ Ibid., p. 100.

¹⁶ Ibid., p. 102.

¹⁷ Ibid., pp. 102, 104, and 106.

¹⁸ Nagel's counterpart to Scheffler's prerogative is more than a mere permission. For Nagel, it would be unreasonable for an agent to sacrifice agent-relative value in the single-minded pursuit of the agent-neutrally best. In contrast, Scheffler seems to want to say that it would not be unreasonable to forgo exercising one's prerogative.

¹⁹ Nagel, , The View from Nowhere, p. 165, emphasis added to the pronouns.Google Scholar

²⁰ Ibid., pp. 175 and 178.

²¹ Ibid., p. 180, emphasis added.

²² In either case, Nagel's account seems to violate his own requirement that deontic restraints “not themselves be understood as the expression of neutral values of any kind” (ibid., p. 177).

²³ Nagel does not address the issue of the magnification of the value of intended good ends. But he does deny that intention would magnify the value of an intended means (ibid., p. 181).

²⁴ Ibid., p. 182.

²⁵ Ibid., p. 183.

²⁶ Ibid., p. 165.

²⁷ The argument over the next couple of paragraphs assumes that some level of injurious force is justified to suppress an aggressor's violation of rights. Nagel himself endorses this view—on the very principle of double effect that is central to his account of deontic restrictions. See Nagel, Thomas, “War and Massacre,” Philosophy and Public Affairs, vol. 1, no. 1 (Winter 1972), pp. 123–44.Google Scholar

²⁸ If it did, then any choice to allow certain benefits to occur that also involved allowing certain evils to occur would involve a magnification of those evils.

²⁹ Nagel, , The View from Nowhere, p. 183.Google Scholar

³⁰ Ibid., p. 184.

³¹ I do not mean to suggest that Nagel is unaware of his focus on the agent rather than the victim. He says, “I have concentrated on the point of view of the agent, as seems suitable in the investigation of an agent-relative constraint” (ibid., p. 183). But I do think he fails to connect how this focus—which is dictated by the program of providing an account of restrictions in terms of the agent's agent-relative reasons—precludes accommodating the intuitions about others' claims and inviolabilities that largely motivate the deontic disposition.

³² Ibid., p. 184.

³³ Ibid.

³⁴ Darwall, Stephen, “Agent-Centered Restrictions from the Inside Out,” Philosophical Studies, vol. 50 (1986), pp. 291–319.CrossRefGoogle Scholar

³⁵ Ibid., p. 311.

³⁶ Spector, Horacio, Autonomy and Rights (Oxford: Clarendon Press, 1992), p. 168Google Scholar, first emphasis added. Cf. also Kagan, Shelly, The Limits of Morality (Oxford: Oxford University Press, 1989), p. 31:Google Scholar

It is clear that no argument [against killing one to prevent the killing of two or more] based on the loss of moral integrity can succeed … unless killing the one does actually involve a sacrifice of the agent's integrity. But if killing the one to save the two is justified—as the [opponent of constraints] believes—then it is no sacrifice of moral integrity for the agent to do so.

³⁷ Darwall, , “Agent-Centered Restrictions,” p. 314.Google Scholar

³⁸ Nozick, , Anarchy, State, and Utopia, pp. 198–204Google Scholar. For the Rawlsian approach, see Rawls, John, A Theory of Justice (Cambridge: Harvard University Press, 1971).Google Scholar

³⁹ Kamm, , “Non-Consequentialism,” p. 382Google Scholar. As previously noted, Kamm rejects this agent-relativist understanding of deontic restrictions. Note that I exploited this “last misleading factor” in my introductory explication of the program under examination.

⁴⁰ Kamm, , “Non-Consequentialism,” pp. 384 and 385.Google Scholar

⁴¹ Spector, , Autonomy and Rights, p. 178.Google Scholar

⁴² Any standard appeal to the agent-neutral disvalue of the violation points to the minimization of violations. For nonstandard appeals to agent-neutral value, see Specter, , Autonomy and Rights, pp. 152–78Google Scholar; and Kamm, , “Non-Consequentialism,” pp. 382–89.Google Scholar

⁴³ Darwall, , “Agent-Centered Restrictions,” p. 304Google Scholar. Darwall ultimately rejects this argument. His reason is that the agent-relative disvalue for the victim does not provide the agent with reason for noninterference. This is correct but irrelevant to the argument at hand. For this argument focuses on the purported incompleteness of an ethical theory that affirms strong prerogatives but no restrictions that protect the exercise of any of those prerogatives. For a more extended statement of this sort of argument (and the next one to be discussed in the text), see my essay “Personal Integrity, Practical Recognition, and Rights,” The Monist, vol. 76, no. 1 (1993).Google Scholar