¹ Scanlon, T. M., The Difficulty of Tolerance (New York: Cambridge University Press, 2003), 129, 133CrossRefGoogle Scholar.

² Scanlon, T. M., What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), 195, 218–20Google Scholar.

³ Ibid., 219. Scanlon writes here that personal reasons “have to do with the claims and status of individuals in certain positions” and seems to identify them with “the claims and interests of individuals.” But he does not make explicit here what makes a claim “a claim of an individual in a certain position” or how this sort of claim differs from a claim that an individual might make on behalf of something else of value, such as the Grand Canyon or an old-growth forest. To clarify the nature of Scanlon's distinction between personal and impersonal reasons, I believe we must therefore explicitly introduce the distinction between reasons that individuals have to prefer their own situations when a state of affairs obtains and the reasons that individuals have to prefer the world when a state of affairs obtains. On this view, an interest or claim just is a good reason for someone to prefer his own situation when a state of affairs occurs.

⁴ This is not to say that there would be no substantive difference between contractualism and utilitarianism. Utilitarianism is a particular view about how individual well-being should be taken into account in identifying the morally right action, a view that is arguably inconsistent with contractualism. Philosophical utilitarianism, in contrast, as Scanlon uses the term, is the name of a general kind of moral theory, of which (aggregating, maximizing) utilitarianism is just one example. The point here is that contractualism would be another example of this general view if all personal reasons were reasons of well-being.

⁵ Ibid., 229.

⁶ Ibid., 229.

⁷ For a similar distinction between two ideas of intrinsic goodness, see Korsgaard, Christine, “Two Distinctions in Goodness,” in Creating the Kingdom of Ends (New York: Cambridge University Press, 1996), 249–74CrossRefGoogle Scholar.

⁸ This example is from Parfit, Derek, Reasons and Persons (New York: Oxford University Press, 1984), 494Google Scholar. Parfit's stranger is not identified as being on a train. Griffin, James puts him there in Well-Being (New York: Oxford University Press, 1986), 171Google Scholar, where he has been ever since.

⁹ I believe that Sidgwick's formal analysis of well-being is a version of the restricted rational desire analysis. Sidgwick takes “ultimate good on the whole for me” to mean “what I should practically desire if my desires were in harmony with reason, assuming my own existence alone to be considered” (The Methods of Ethics [Macmillan, 1907; Indianapolis: Hackett Publishing Co., 1981], 112). This analysis addresses the stranger-on-the-train objection because it narrows the scope of the relevant objects of desire to what it would make sense to desire “assuming my own existence alone to be considered.” But what is it to assume this? If it is to assume that only my own well-being or good is to be considered, then the analysis is circular. If it is to assume that only I exist, this would entail that a number of claims by ethical theorists about well-being are conceptually incoherent. It would entail, for example, that being a good parent to our children is not intrinsically good for us, since it is not rational to want to be a good parent to someone on the assumption that she does not exist. Some would deny that being a good parent is intrinsically good for us, thinking that it is good for us only as a means to the pleasure it gives us. But we should reject any analysis that renders conceptually incoherent the claim that being a good parent contributes (directly) to our well-being, because any such analysis would fail to illuminate the concept that ethical theorists are using when they claim (perhaps mistakenly) that being a good parent does contribute (directly) to our well-being. (See, for example, Reasons and Persons, 495, where Parfit identifies being a successful parent as something that would contribute directly to our well-being if the “Success Theory” were true.) There is, however, another interpretation of Sidgwick that avoids these difficulties. On this interpretation “ultimate good for me” means “what I should practically desire for myself if my desires were in harmony with reason.” In other words, a person's well-being consists only in what he ought to desire, or would rationally desire, for himself. If this is the correct interpretation of Sidgwick's view, then his formal analysis of well-being is a version of the restricted rational desire analysis. His substantive theory of well-being, in contrast, is a form of hedonism, because he believes that the only things a person actually ought to desire for himself are “desirable states of consciousness” (The Methods of Ethics, 397).

¹⁰ Scanlon, The Difficulty of Tolerance, 31–32.

¹¹ Ibid., 33.

¹² Kagan, Shelly, “The Limits of Well-Being,” Social Philosophy and Policy 9, no. 2 (1992): 181CrossRefGoogle Scholar.

¹³ Ibid., 186.

¹⁴ Kagan, , Normative Ethics (Boulder, CO: Westview, 1998), 29–41Google Scholar.

¹⁵ Kagan, might in fact believe that desire-satisfaction and objective list views are not theories of well-being, but of something else. In “Me and My Life,” Proceedings of the Aristotelian Society New Series 94 (1994): 318CrossRefGoogle Scholar, he makes a distinction between a person being well off and a person's life going well. Perhaps Kagan believes that the intrinsic property analysis is the correct understanding of what is good for a person, but not of what makes a person's life go well. He might think, for example, that although having true friends would not in itself be better for a person than having false ones, it might still make this person's life go better. Perhaps Kagan believes that when philosophers use the term “well-being” they are confusing two different ideas. Desire-satisfaction theories and objective list theories are not, in fact, theories of well-being properly understood; they are theories of something else: how well a person's life goes (impersonally, and not for him). It remains the case that the intrinsic property analysis cannot possibly be an analysis of what other leading theorists mean when they use the term “well-being,” because they clearly understand it in a way that allows relational properties to contribute.

¹⁶ Darwall, Stephen, Welfare and Rational Care (Princeton: Princeton University Press, 2002), 4, 7–9CrossRefGoogle Scholar. Note to nonspecialist readers: “=df.” means that the expression to the left of the symbol is being defined by the expression to the right of the symbol.

¹⁷ Ibid., 8. See, also, 38.

¹⁸ Darwall, Stephen, “Reply to Feldman, Hurka, and Rosati,” Philosophical Studies 130, no. 3 (2006): 642CrossRefGoogle Scholar.

¹⁹ If one wonders what it means to say that p is good in some way only because S is worthy of care, here is one interpretation: there is a reason to want p that is a good reason only because S is worthy of care. So: p is good for S =df. there is a good reason to want p that is a good reason only because S is worthy of care.

²⁰ Darwall, Welfare and Rational Care, 6.

²¹ For a different circularity objection, see Rosati, Connie, “Darwall on Welfare and Rational Care,” Philosophical Studies 130, no. 3 (2006): 619–635CrossRefGoogle Scholar.

²² I imagine someone like this in “An Objection to Attitudinal Hedonism,” Philosophical Studies 115, no. 2 (August 2003): 197–200CrossRefGoogle Scholar.

²³ Hooker, Brad, “Does Moral Virtue Constitute a Benefit to the Agent?” in Crisp, Roger, ed., How Should One Live? (New York: Oxford University Press, 1996), 149, 155Google Scholar.

²⁴ But for a nontraditional hedonist view that allows one to say that Jones's life goes better for him than Smith's life goes for him, see Feldman, Fred, Pleasure and the Good Life (New York: Oxford University Press, 2004), 112CrossRefGoogle Scholar.

²⁵ Note, too, that this objection to the rational sympathy view does not seem to provide a decisive reason to favor the rational care view. The rational care view says that p is bad for S if and only if p is bad in a way that it would not be bad if S were unworthy of care. So if S is worthy of care, then p can be bad for S even if we should not feel sorry for S because he is so evil. But if S is worthy of care, then it seems we should feel sorry for S even if he is evil. And if he is not worthy of care, then the rational care analysis does not support the conclusion that his punishment is bad for him, since now there is no way in which it is bad that it would not be bad were S unworthy of care. S's suffering might still be impersonally bad, but it would be bad in this way even if S were unworthy of care.

²⁶ See, for example, Rabinowicz, Wlodek and Ronnow-Rasmussen, Toni, “The Strike of the Demon: On Fitting Pro-attitudes and Value,” Ethics 114, no. 3 (2004): 391–423CrossRefGoogle Scholar.

²⁷ One might argue that the rational care analysis has its own wrong kind of reasons problem. What is it, after all, for x to be good in a way that it would not be good if S were unworthy of care? One might answer that it is for there to be a reason to want x that is a good reason only because S is worthy of care. If so, then the rational care view seems to face a wrong kind of reasons problem of its own because it makes implicit use of the notion of “reasons to want.” The way in which it makes use of this notion, however, blocks the objection. The fact that it is rational to want S to suffer if one will be tortured unless one wants S to suffer does not render it true, on this analysis, that suffering is good for S, because it is not the case that this is a good reason because S is worthy of care. One might now puzzle over what it is for a reason to be a good reason because S is worthy of care, but this is another issue. So the rational care analysis is not vulnerable to the wrong kind of reasons problem in the way that the rational sympathy analysis appears to be.

²⁸ See Gibbard, Allan, Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press, 1990), 37Google Scholar, and Parfit, Derek, On What Matters, vol. 1 (New York: Oxford University Press, 2011), 50–51, 420–32Google Scholar.

²⁹ One might offer a similar defense of the restricted rational desire analysis of well-being. Some might argue that if I will be tortured unless I want someone else to suffer, then I ought to want this person to suffer. Because this does not make it true that this person's suffering is good for him, one might conclude that the restricted rational desire analysis is wrong. A reasonable response is that although the fact that I will be tortured unless I want someone else to suffer might be a good reason for me to want this person to be tortured, it is not a good reason for me to want this for him, and so does not present a counterexample to the restricted rational desire analysis.

³⁰ Scanlon, What We Owe to Each Other, 112–13.

³¹ Nozick, Robert, Anarchy, State, and Utopia (New York: Basic Books: 1974), 42–45Google Scholar.

³² See Feldman, Pleasure and the Good Life, 9–20.

³³ Thanks to Doug Portmore for suggesting this in discussion.

³⁴ See Scanlon, What We Owe to Each Other, 126–33.