Published online by Cambridge University Press: 26 January 2009
The notion of self-interest has not received from philosophers of this century the attention it deserves. In this paper, I shall first elucidate the views on self-interest of a philosopher who nourished in the last century. It could be argued that Henry Sidgwick's views on this topic are the most considered in the history of philosophy. I shall then point to a number of misconceptions in his position, and suggest a more satisfactory account. I shall attempt also to solve a problem for this new account with the aid of a Sidgwickian distinction.
I am indebted to Robin Attfield, Neil Cooper, David Edmonds, James Griffin, Brad Hooker, Michael Lockwood, Derek Parfit, Wayne Sumner and Sir Geoffrey Warrnock for comments on earlier drafts of this paper.
1 See Hurka, T., ‘Critical notice of Weil-Being, by James Griffin’, Mind, xcvii (1988), 464Google Scholar. The notion of self-interest is roughly equivalent to ‘Ultimate Good for the individual’, ‘what it is prudentially or practically rational to do’, ‘what makes life valuable’, and indeed ‘utility’, in its now usual sense, as opposed to the sense in which a thing has utility in so far as it produces what is of intrinsic value.
2 Parfit, D., Reasons and Persons, Oxford, 1984, Appendix I, p. 493.Google Scholar
3 Sidgwick, H., The Methods of Ethics, 7th edn., London, 1907, Book III, ch. xiv, p. 398.Google Scholar
4 See Griffin, J., Weil-Being, Oxford, 1986, p. 9.Google Scholar
5 Parfit, , p. 500.Google Scholar
6 Sidgwick, (1907), Book I, ch. ix, pp. 111–12.Google Scholar
7 Ibid., p. 111. It might be thought that this definition is obviously lacking, since Sidgwick does not distinguish between desires a person has in relation to his own life and those he has in relation to other things. Presumably self-interest is concerned only with the former. In fact, however, Sidgwick does restrict his discussion to ‘what a man desires … for himself’ on p. 109.
8 Ibid., pp. 111–12. There is something of a puzzle in the transition from Definition (1) to Definition (2). Why should the fact that satisfying a given desire will cause one to become a ‘contented pig’ not lead to an informed desire for some other outcome, given that this fact will be ‘foreknown and adequately realized’? Why, in other words, is the possibility Sidgwick contemplates an objection to Definition (1)? I think that the puzzle is resolved if one takes account of the last three words of Definition (2*) below: ‘if actually experienced’. If Sidgwick were supposing that the only way desires can be formed by a consideration of consequences was by being formed in response to the (imagined) experience, then if becoming a ‘contented pig’ would lead to no desire to escape this condition, so likewise would an adequate imagining of this condition lead to no desire to avoid this condition. Sidgwick was almost certainly relying on this Hedonistic assumption when he rejected Definition (1).
9 Ibid., p. 112.
10 Sidgwick, , 3rd edn., London, 1884, p. 108Google Scholar. I am assuming here a Principle of Charity, viz. that we are not, without good reason, to interpret an author as changing his mind on an issue. Here there seems no reason to do so.
11 Sidgwick, (1907), p. 112.Google Scholar
12 Ibid.
13 The ‘however’ of the opening line of the second paragraph on p. 112 (that containing Definition (3)) is best read as opposing the last six words of the previous paragraph, not Definition (2) itself.
14 Sidgwick, , 3rd edn. (1884), p. 108.Google Scholar
15 Schneewind, J. B., Sidgwick's Ethics and Victorian Moral Philosophy, Oxford, 1977, p. 225Google Scholar. Schneewind's interpretation differs from Parfit's in that he stresses Sidgwick's insistence that the notion of good includes that of a dictate of reason, arguing that this is why he abandons Definition (2). One way of bringing out the difference would be to note that Schneewind's reading of Sidgwick is consistent with rejecting an Objective List Theory. It could be that good is the object of rational desire, and that the criterion of rational desire is neither simply informed desire, nor desire which responds to an independently grounded good.
16 Sidgwick, (1907), p. 381.Google Scholar
17 Ibid., pp. 113–14; pp. 128–9; pp. 398–402. It is interesting that there is no hint in the earlier ch. iv on ‘Pleasure and Desire’ that desires not directed at one's own pleasure are irrational, despite the fact that Sidgwick discusses at length and rejects Psychological Hedonism in that chapter. Again, in that chapter (pp. 44–5), Sidgwick follows Butler in arguing that there are occasions where a pleasure arises when one does not previously desire it but something else. His example is hunger. But this is all consistent with the Experience Requirement on what conduces to self-interest.
18 Griffin, , note 7, p. 313.Google Scholar
19 Sidgwick, (1907), Book III, ch. xiv, pp. 398–402.Google Scholar
20 Ibid., p. 399.
21 Ibid., p. 401; see p. 113.
22 As do those of, for example, Rashdall, H., The Theory of Good and Evil, Oxford, 1907, vol. I, p. 68Google Scholar; Nagel, T., ‘Death’, in Mortal Questions, Cambridge, 1979, p. 5Google Scholar; Darwall, S., ‘Pleasure as ultimate good in Sidgwick's ethics’, The Monist, lviii (1974), 478Google Scholar; Feinberg, J., ‘Harm and self-interest’, in Hacker, P. and Raz, J. eds., Law, Morality, and Society, Oxford, 1977, p. 303Google Scholar; Gombay, A., ‘What you don't know doesn't hurt you’, Proceedings of the Aristotelian Society, lxxix (1978–1979), 245Google Scholar; Sen, A., ‘Pluralutility’, Proceedings of the Aristotelian Society, lxxxi (1980–1981), 203Google Scholar; Levenbook, B., ‘Harming someone after his death’, Ethics, xciv (1984), 407CrossRefGoogle Scholar; Griffin, , pp. 9–10.Google Scholar
23 One philosopher who disagrees with Sidgwick and me here is G. E. Moore. See Moore, G., Principia Ethica, Cambridge, 1903, §50, pp. 83–5Google Scholar. It is important, however, to note the difference between the following claims: (a) the existence of a plant has value, whether or not it is ever perceived or desired; and (b) the existence of a plant is valuable for that plant, whether or not it is ever perceived or desired. Moore makes claim (a). But, as I argue below in the text, the notion of good for may be parasitic on that of good simpliciter.
24 A contemporary of Sidgwick's says: ‘It does not follow that, because nothing is good … out of relation to consciousness … therefore its goodness … lies in the mere state of consciousness’. Seth, A., ‘Is pleasure the summum bonum?’, International Journal of Ethics, vi (1986), 422Google Scholar. Quoted in Darwall, , p. 483.Google Scholar
25 See Nagel, , p. 5.Google Scholar
26 Sidgwick, (1907), p. 401.Google Scholar
27 Ibid., pp. 406–7.
28 Griffin makes a similar objection to Sen's notion of ‘plural utility’. This objection fails for the same reason as Sidgwick's. See Griffin, , note 27, pp. 316–17.Google Scholar
29 See Griffin, , p. 31; p. 103.Google Scholar
30 On the value of accomplishment, see Griffin, , p. 19; passim.Google Scholar
31 I include this sentence in the example merely to insure it against the claim by the Hedonist that, by allowing in different qualities of pleasure, the difference in value of the two lives can be accounted for.
32 Michael Lockwood mentioned to me in this connection the case of Arthur Conan Doyle. Conan Doyle wanted to write historical novels. He did so, and they were mediocre. To finance the writing of these novels, he wrote the Sherlock Holmes stories. He did not want to write the latter for their own sake, and thought that they were of no merit. But I would argue that writing these stories constitutes a great accomplishment.
33 I owe this example to Brad Hooker. Likewise that of The Lottery below.
34 See Moore, , pp. 98–9.Google Scholar
35 Sidgwick, (1907), pp. 111–12.Google Scholar
36 Parfit, , Appendix I, pp. 499–500Google Scholar; Rawls, J., A Theory of Justice, Oxford, 1972, p. 432.Google Scholar
37 Sidgwick, (1907), pp. 41–2; p. 51Google Scholar; pp. 58–9; p. 110.
38 This stipulation might seem ad hoc. For sheer desire fulfilment does not seem of great value. If you were to place my life in danger constantly, always rescuing me at the last minute, you would indeed fulfil many strong desires of mine. But you would not make me better off. (The example is from Rosalind Godlovitch. See Griffin, , p. 15Google Scholar.) There is at least one good reason for the stipulation, however. It rests on the distinction between ‘desired’ and ‘desirable’ mentioned in the final section of the paper. Goods are desirable. They ought to be desired. And finding goods desirable, desiring them, is itself a further good. For if I desire what is in fact good, I am in touch with the world as it is. I have understanding, and it is likely that this will feature as a separate substantive value on a plausible Objective List.
39 See Moore; Rashdall; also Sen; and the Organic Whole conception tentatively suggested in Parfit, , p. 500.Google Scholar
40 This problem was suggested to me by L. W. Sumner. See also Sumner's suggestion in Griffin, , note 5, p. 318.Google Scholar
41 Sidgwick, (1907), pp. 110–11.Google Scholar
42 Robin Attfield has suggested that the ‘goods’ of plants may be desirable in the sense that we, or some other conative beings, may desire those goods. But I cannot see why we should find the ‘goods’ of plants desirable other than instrumentally with regard to our own interests.