No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 23 June 2016
Many scholars have argued that well-being is the satisfaction of preferences, or of fully informed preferences, or of fully informed preferences about one's own life. But none of those theories can be true if it is possible to prefer, with full information, to decrease one's own well-being. And because it is possible to have such a preference, those theories cannot be true.
1 Sumner, L. W., Welfare, Happiness, and Ethics (Oxford, 1996), p. 122 Google Scholar; Haybron, D., The Pursuit of Unhappiness (Oxford, 2008), p. 34 Google Scholar.
2 Kraut, R., ‘Desire and the Human Good’, Proceedings and Addresses of the American Philosophical Association 68 (1994), pp. 39–54, at 40-1CrossRefGoogle Scholar.
3 Murphy, M., ‘The Simple Desire-Fulfillment Theory’, Noûs 33 (1999), pp. 247–72CrossRefGoogle Scholar; Heathwood, C., ‘The Problem of Defective Desires’, Australasian Journal of Philosophy 83 (2005), pp. 487–504 CrossRefGoogle Scholar.
4 Rawls, J., A Theory of Justice (Cambridge, MA, 1971), p. 408 Google Scholar; Brandt, R., A Theory of the Good and the Right (Amherst, NY, 1979), ch. 6Google Scholar; Harsanyi, J., ‘Morality and the Theory of Rational Behavior’, Utilitarianism and Beyond, ed. Sen, A. and Williams, B. (Cambridge, 1982), pp. 39–62 Google Scholar; Railton, P., ‘Facts and Values’, Philosophical Topics 14 (1986), pp. 5–31 CrossRefGoogle Scholar.
5 Parfit, D., Reasons and Persons (Oxford, 1984), p. 494 Google Scholar.
6 Keller, S., ‘Welfare as Success’, Noûs 43 (2009), pp. 656–83, at 658CrossRefGoogle Scholar.
7 Darwall, S., ‘Self-Interest and Self-Concern’, Social Philosophy and Policy 14 (1997), pp. 158–78, at 164-5CrossRefGoogle Scholar.
8 Adler, M., Well-Being and Fair Distribution (New York, 2012), p. 180 Google Scholar.
9 Feldman, F., What Is This Thing Called Happiness? (Oxford, 2010), pp. 67–9CrossRefGoogle Scholar.
10 Sumner, Welfare, p. 135.
11 Arneson, R., ‘Human Flourishing Versus Desire Satisfaction’, Human Flourishing, ed. Paul, E. Frankel et al. (Cambridge, 1999), pp. 113–42, at 124CrossRefGoogle Scholar.
12 Griffin, J., Well-Being: Its Meaning, Measurement, and Moral Importance (Oxford, 1986), p. 22 Google Scholar.
13 Kraut, ‘Desire’, pp. 40-1.
14 See Feldman, F., Pleasure and the Good Life (Oxford, 2004), p. 17 CrossRefGoogle Scholar; Adams, R., Finite and Infinite Goods (New York, 1999), pp. 89–91 Google Scholar; Carson, T., Value and the Good Life (Notre Dame, IN, 2000), p. 92 Google Scholar. Feldman says only that the example seems to force a paradox upon the theory that well-being is the satisfaction of a person's actual preferences. Adams says that a preference theory cannot avoid the example by excluding from the well-being calculus people's preferences against their own well-being, because such a theory would be circular and would not sufficiently honour people's actual preferences. Carson argues that the example is not problematic for the preference-satisfaction theory of value, as opposed to the preference-satisfaction theory of well-being, because ‘a bad person who punishes herself’ may increase the value of her life even as she decreases her well-being.
15 Bradley, B., ‘A Paradox for Some Theories of Welfare’, Philosophical Studies 133 (2007), pp. 45–53 CrossRefGoogle Scholar.
16 In addition, Bradley supplies a response to Chris Heathwood, who argues that the paradox of wanting not to get what one wants exists regardless of whether well-being is preference satisfaction, and that therefore the example is not specifically a problem for preference theories of well-being but instead a general problem like the liar paradox. See Heathwood, ‘The Problem’, pp. 501-3. Bradley rebuts that argument by pointing out that only if preference theories are true does wanting to decrease one's own well-being equate with wanting not to get what one wants. Therefore, only preference theories must claim that the apparently non-paradoxical preference to decrease one's own well-being is in fact paradoxical. See Bradley, ‘A Paradox’, pp. 51-2.
17 I will discuss below the issue that arises if the satisfaction of one such preference causes other such preferences to go unsatisfied.
18 Such a preference is incompatible with preference theories whether or not the preference is satisfied. According to preference theories, Sarah's level of well-being would be lower if any one of her fully informed preferences about her own life were unsatisfied than if that preference were instead satisfied. But if Sarah has a fully informed preference to decrease her own well-being, then her level of well-being would necessarily be higher if her preference were unsatisfied than if it were satisfied.
19 Parfit, Reasons, p. 493.
20 Adler, Well-Being, p. 180.
21 Indeed, a claim of psychological impossibility would run counter to the way that well-being theories are typically assessed. For example, Nozick's experience-machine counterexample to hedonism does not depend on whether human beings are physically or psychologically capable of tolerating an experience machine. See R. Nozick, Anarchy, State, and Utopia (New York, 1974), pp. 42-5.
22 Sumner, Welfare, p. 131.
23 Whatever other wants Ellie may have, we may hypothesize that she prefers to decrease her own well-being above all of those other wants combined. However unusual that might be, it is surely possible.
24 For example, only preference theories, and not hedonic or objective theories, must deny that it is possible to prefer only to decrease one's own well-being. And a claim of conceptual impossibility seems insupportable, whereas a claim of psychological impossibility seems insufficient.
25 Indeed, Peter Railton privileges second-order preferences because doing so insulates preference theories from various criticisms. See Railton, ‘Facts’.
26 See Bradley, ‘A Paradox’, p. 50.
27 Skow, B., ‘Preferentism and the Paradox of Desire’, Journal of Ethics and Social Philosophy 3 (2009), pp. 1–16 Google Scholar.
28 As I do throughout, I am referring only to fully informed preferences about one's own life. And in my discussion of Skow, I am also following Skow's useful simplifying assumption that the preference for one's life to go badly is one's only preference. The argument would be unaffected if one had other preferences as well, as long as one preferred for her life to go badly more strongly than she preferred those other things collectively.
29 As discussed above, an advantage of hedonic and objective theories is that they need not insist that a preference for one's life to go badly differs radically from other seemingly similar preferences such as a preference for one's life to go well or a preference for others’ lives to go badly or to go well. That advantage is undiminished if preference theories shift from claiming that it is impossible to prefer one's life to go badly to claiming that it is impossible to satisfy such a preference to any degree other than 50 per cent. According to hedonic and objective theories, it is possible to satisfy a preference for or against one's own well-being, or a preference for or against others’ well-being, to any degree.
Moreover, it seems highly unusual for any preference to be satisfiable only to a degree of 50 per cent. To be sure, some preferences may be satisfiable only to a particular degree due to physical laws or other restrictive circumstances. For example, if Martin prefers to live on the sun, then his preference can never be satisfied to any degree other than zero. But examples like Martin's preference lend no support to the claim that a preference for one's own life to go badly can only be precisely half-satisfied, because the reason supporting that claim (unlike the reason – physical laws – in examples like Martin's preference) is merely the asserted truth of preference theories themselves. And the assertion that preference theories are true cannot, without circularity, be used to substantiate the truth of preference theories.
30 Bricker, P., ‘Prudence’, The Journal of Philosophy 77 (1980), pp. 381–401 CrossRefGoogle Scholar.
31 Bricker's calculus also includes alternative versions of oneself that would have existed if one had made different choices. Thus, satisfying a current preference could diminish one's lifetime well-being if such satisfaction leads to a future with less preference satisfaction than one would otherwise have had.
32 This makes the case simpler but is not required for the argument to work. There is no reason to believe that if Ryan lives far beyond the age of 40, satisfying his preference to decrease his own well-being will necessarily reduce his level of preference satisfaction at future ages or in alternative versions of himself.
33 The example of a preference to decrease one's own well-being actually poses a more intractable problem for preference theories than does the experience machine for hedonic theories. Hedonists can and often do bite the bullet by claiming that a life on the experience machine would be high in well-being. But when a preferentist faces the counterexample discussed here, he cannot bite the bullet because satisfying a preference to decrease one's own well-being necessarily does not increase one's well-being.
34 Even Parfit's Success Theory must do so to explain why it excludes preferences that are not about one's own life. And although it might be easier to find a principled reason for ruling out such preferences than for ruling out a preference to decrease one's own well-being, this burden still weighs heavily on the Success Theory and may be a reason that it has not been embraced nearly as widely as the full-information preference theory. The full-information preference theory is not similarly undercut by excluding uninformed preferences because, as explained above, those can plausibly be claimed not to be actual preferences.
35 I thank an anonymous reviewer for this point.
36 Overvold, M., ‘Self-Interest and the Concept of Self-Sacrifice’, Canadian Journal of Philosophy 10 (1980), pp. 105–18CrossRefGoogle Scholar; Darwall, ‘Self-Interest’.
37 Heathwood, C., ‘Preferentism and Self-Sacrifice’, Pacific Philosophical Quarterly 92 (2011), pp. 18–38, at 32-5CrossRefGoogle Scholar.
38 I am grateful to Benjamin Callard, Richard Kraut, Douglas MacKay, Sonja Starr and the reviewer for their helpful comments on earlier drafts. In particular, I thank Josh Sheptow, who commented indefatigably on many drafts and without whose help this article could not have been written.