¹ Cf. Copleston, Frederick, A History of Philosophy: Volume VIII, Garden City, 1985, p. 30.Google Scholar

² In On Liberty, Mill insists that the pleasures of malevolent intolerance, as well as those secured by benevolent paternalism, must never be allowed to outweigh the pleasures of autonomy, no matter how strong and numerous the former may be. While most of On Liberty is devoted to considerations of indirect utility, Mill also claims that the development of individuality is sufficiently valuable in itself that he ‘might here close the argument’ for liberty, and that he only invokes the indirect social benefits of liberty to convince those who do not recognize the instrinsic value of autonomy (‘On Liberty’, Essays on Politics and Society, ed. Robson, John M., 2 vols., Toronto, 1977Google Scholar, Collected Works of John Stuart Mill, xviii. 267Google Scholar (and thereafter as ‘On Liberty’, CW, xviii. 267)Google Scholar; cf. ‘The Subjection of Women’, Essays on Equality, Law, and Education, ed. Robson, John M., Toronto, 1984Google Scholar, CW xxi. 337.)Google Scholar Such remarks are reminiscent of Mill's claim in Utilitarianism that while earlier utilitarians had successfully argued for ‘the superiority of mental over bodily pleasures’ by pointing to ‘their circumstantial advantages rather than… their intrinsic nature… they might have taken the other, and, as it may be called, higher ground, with entire consistency’ (‘Utilitarianism’, Essays on Ethics, Religion, and Society, ed. Robson, John M., Toronto, 1969, CW, x. 211Google Scholar. Cf. Plato, , Polity II)Google Scholar. If Mill is suggesting that individuality and autonomy are among the higher pleasures, then his claim that there is ‘no parity’ (‘On Liberty’, CW, xviii. 283)Google Scholar between self-regarding and intolerant desires becomes more understandable.

³ Mill seems to attribute just such a view to his father (Autobiography and Literary Essays, ed. Robson, John M. and Stillinger, Jack, Toronto, 1981Google Scholar, CW, i. 51).Google Scholar

⁴ Raphael, D. D., ‘Fallacies In and About Mill's Utilitarianism,’ Philosophy, xxx (1955), 352.Google Scholar

⁵ Diary entry, 23 03 1854Google Scholar; in Journals and Debating Speeches, ed. Robson, John M., 2 vols., Toronto, 1988Google Scholar, CW, xvii. 663.Google Scholar

⁶ Of course, this might be one of Mill's not infrequent grammatical lapses. But we can hardly assume that it is.

⁷ ‘Utilitarianism’, CW, x. 211.Google Scholar

⁸ Similar considerations apply to the much more sophisticated interpretation of Mill in Riley, Jonathan, Liberal Utilitarianism: Social Choice Theory and J. S. Mill's Philosophy, Cambridge, 1988Google Scholar. Riley suggests (if I understand him properly) that, for Mill, pleasures may be divided into distinct classes, where quantitative differences within a given class will be finite, but quantitative differences between classes will be infinite (or indefinite). This would explain nicely why higher pleasures always trump lower ones. On Riley's interpretation, however, it remains the case, as with Raphael, that higher pleasures are themselves superior to the lower pleasures they trump; and this, once again, is just what Mill is at pains to deny. Making the quantitative difference infinite rather than finite doesn't seem to help; if anything, it exacerbates the problem.

⁹ Happiness is the aggregate of which pleasures are the component parts…. Let not the mind be led astray by any distinctions drawn between pleasures and happiness…. Happiness without pleasures is a chimera and a contradiction; it is a million without any units, a square yard in which there shall be no inches, a bag of guineas without an atom of gold. (Bentham, Jeremy, Deontology together with A Table of the Springs of Action, ed. Goldworth, A., Oxford, 1983Google Scholar (The Collected Works of Jeremy Bentham); quoted in Hazlitt, Henry, The Foundations of Morality, Los Angeles, 1964, p. 20).Google Scholar

¹⁰ Berger, Fred R., Happiness, Justice, and Freedom: The Moral and Political Philosophy of John Stuart Mill, Berkeley, 1984.Google Scholar

¹¹ This would allow Berger to escape Bentham's criticism in n. 9 above. It is not clear why Berger believes this, though, given his claim (pp. 12–16) that Mill thinks people can and should desire things other than pleasures.

¹² Berger, , pp. 38–9Google Scholar. Berger is unclear as to whether the distinction is an absolute one, between pleasures that are included in happiness and pleasures that are not, or a difference of degree among pleasures that make more or less of a contribution to happiness.

¹³ Berger, , p. 38.Google Scholar

¹⁴ In particular, it rules out as non-utilitarian any form of rule-utilitarianism (such as Lyons's) that does not collapse into act-utilitarianism. But while Mill sometimes defines utilitarianism in terms of acts (‘Utilitarianism’, CW, x. 210)Google Scholar, he also defines it in terms of what it sets up as the ultimate standard of value (ibid., 234). And the discussion of Teleology in the System of Logic, ed. Robson, John M., 2 vols., Toronto, 1973Google Scholar, CW, viii., 950–2Google Scholar, suggests that the latter criterion is for Mill the more fundamental one. See also Lyons, David, ‘Mill's Theory of Morality’, Nous, x (1976), 101.CrossRefGoogle Scholar

¹⁵ ‘Utilitarianism’, CW, x. 234.Google Scholar

¹⁶ Berger, , p. 49.Google Scholar

¹⁷ ‘Utilitarianism’, CW, x. 237.Google Scholar

¹⁸ Or from habituation, possibly. But how would such a habit ever get started? Besides, as we will see, Millian habituation comes in two forms, and neither one seems to help us here. In one form, we are habituated to value certain things for themselves. But this, Mill tells us, is to value them as parts of happiness; and lower pleasures are excluded from Bergerian happiness ex hypothesi. In the other form, we are habituated to choose certain things mechanically, without anticipating any pleasure from them. But this seems implausible as an account of lower pleasures (or at least of all of them).

¹⁹ ‘Utilitarianism’, CW, x. 211–13.Google Scholar

²⁰ Ibid., 210.

²¹ Ibid., 212. I take Mill's distinction between happiness and contentment to be this: happiness is pleasure and the absence of pain, while contentment is merely the absence of pain. Since ‘the being whose capacities are low [that is, the being capable of less pleasure] has the greatest chance of having them fully satisfied’ (ibid., 212), the pig will have fewer unsatisfied desires—and so less pain. So if Socrates's life has 30 units of pleasure and 10 units of pain (giving it a happiness-total of 20 units), while the pig's life has 6 units of pleasure and 1 unit of pain (giving it a happiness-total of 5 units), Socrates is four times as happy as the pig, but the pig (having fewer desires to be frustrated) is ten times as contented, or satisfied, as Socrates.

²² Ibid., 211.

²³ Ibid., 214.

²⁴ Ibid., 211–13.

²⁵ Ibid., 211–12.

²⁶ For the purposes of this paper, I shall assume that Mill is a psychological hedonist. Berger, of course, denies the truth of this assumption; but I think he is mistaken. That however, is a subject for another article.

²⁷ ‘Utilitarianism’, CW, x. 211Google Scholar. The appeal to ‘experienced’ judges is not, as it is sometimes taken to be, elitism on Mill's part. Rather, it is a way of obtaining evidence whereby to determine what everyone is likely to enjoy. The expressed preferences of the uninformed and inexperienced cannot tell us what they would actually enjoy if they ever achieved it.

²⁸ Examination of Sir William Hamilton's Philosophy, ed. Robson, John M., Toronto, 1979Google Scholar, CW, ix. 468.Google Scholar

²⁹ Let me briefly distinguish my interpretation from two accounts it somewhat resembles—those of John Gray and George Harris. Gray's theory involves two levels of choice, but the distinction is not between the choice of pleasures and the choice of life or character; indeed, Gray seems to conflate the two (Gray, John, Mill on Liberty: A Defence, London, 1983, p. 72)CrossRefGoogle Scholar. Gray appears to claim that the higher pleasures are the ones that autonomous people living autonomous lives will choose (p. 70); thus far the interpre tation is rather like mine. But the connection with autonomy can only make the higher pleasures valuable if autonomy itself is valuable. Gray thinks the intrinsic pleasure of autonomy will not carry the entire burden of its overriding value; instead, he argues that, given the limitations of human knowledge, we can maximize utility better by weighting autonomy heavily than we could by pursuing the greatest happiness directly. The major problem I see for this broadly rule-utilitarian interpretation is that it seems to make the preferability of higher pleasures dependent on instrumental considerations; this does not seem to be Mill's view.

Harris clearly draws the distinction between ‘first order preferences, those that are made within a mode of existence (e.g. coffee or tea, music or sport), and second order-preferences, those that are made for a mode of existence (e.g., Socrates or the fool)… the packaging of one's preferences is itelf a matter of preference’ (Harris, George W., ‘Mill's Qualitative Hedonism,’ Southern Journal of Philosophy, xxi (Winter 1983), 507)Google Scholar. But Harris's interpretation, unlike mine, is not even indirectly reductive; he appeals to ‘an irreducibly qualitative dimension to human preference satisfaction’ (505). He considers two different versions of a two-level theory, but neither is like mine. In the first version, we choose a mode of life solely on account of the (irreducibly qualitatively superior) higher pleasures it involves; in the second version, we choose a mode of life because we get an (irreducibly qualitatively superior) higher pleasure out of choosing it, regardless of what pleasures it involves (507–8). Harris does not consider the possibility that the qualitative superiority of first-order pleasures might be analysable in terms of or parasitic on the (merely) quantitative superiority of second-order pleasures. It is this last view that I am defending.

³⁰ ‘On Liberty’, CW, xviii. 266.Google Scholar

³¹ ‘On Genius’, CW, i. 338.Google Scholar

³² Bain, Alexander, John Stuart MillGoogle Scholar; quoted in Sparshott, F. E., Introduction to Essays on Philosophy and Classics, ed. Robson, John M., Toronto, 1978Google Scholar, CW, xi. p. xvii.Google Scholar

³³ ‘Utilitarianism’, CW, x. 205 and 209Google Scholar; ‘On Liberty’, CW, xviii. 235; ‘The Protagoras’, CW, xi. 61.Google Scholar

³⁴ ‘Grote's Plato’, CW, xi. 419.Google Scholar

³⁵ See Gibbs, Benjamin, ‘Higher and Lower Pleasures’, Philosophy, lxi (1986), 33–40Google Scholar, for a discussion of some parallels between Mill and the Ancients.

³⁶ Autobiography, CW, i. 49Google Scholar. Appropriately, Xenophon's fable concerns Herakles's choice between two modes of life. For an exploration of the choice-of-Herakles theme in Mill's thought, see Semmel, Bernard, John Stuart Mill and the Pursuit of Virtue, New Haven, 1984.CrossRefGoogle Scholar

³⁷ Borchardt, Ruth, ‘John Stuart Mill and the Ancients’, in John Stuart Mill, Four Dialogues of Plato; Including the ‘Apology of Socrates’, ed. Borchardt, Ruth, London, 1946) p. 11.Google Scholar

³⁸ Autobiography, CW, i. 147.Google Scholar

³⁹ ‘Grote's Plato’, CW, xi. 416.Google Scholar

⁴⁰ ‘On Genius’, CW, i. 329.Google Scholar

⁴¹ Letter to Sterling, John, 24 05 1832Google Scholar, in Earlier Letters, 1812–48, ed. Mineka, Francis E., 2 vols., Toronto, 1963Google Scholar, CW, xii. 101Google Scholar; ‘Civilization’, CW, xviii. 139Google Scholar; ‘Bentham’, CW, x. 99Google Scholar;, ‘On Liberty’, CW, xviii. 263, 265 and 267Google Scholar; ‘Considerations on Representative Government’, Public and Parliamentary Speeches, ed. Robson, John M. and Kinzer, Bruce L., 2 vols., Toronto, 1988Google Scholar, CW, xix. 400Google Scholar; Examination, CW, xix. 466Google Scholar; and ‘Utility of Religion’, CW, x. 406–7.Google Scholar

⁴² ‘Remarks on Bentham's Philosophy’, CW, x. 7–8.Google Scholar

⁴³ ‘Utilitarianism’, CW, x. 213.Google Scholar

⁴⁴ ‘The philosopher who is a half-hearted sensualist cannot estimate the attractions of a debauched existence, any more than the sensualist flicking through the pages of Hume can estimate the pleasures of philosophy’ (Ryan, Alan, J. S. Mill, London, 1974, p. 111).Google Scholar

⁴⁵ Polity, 582 a–d.Google Scholar

⁴⁶ Nikomakhean Ethics, 1179b, 15–16 (hereafter NE).Google Scholar

⁴⁷ NE, 1174a, 1–3Google Scholar; cf. Plato, , Philebos, 21a–d.Google Scholar

⁴⁸ ‘Utilitarianism’, CW, xviii. 211.Google Scholar

⁴⁹ Laws, 658e–659aGoogle Scholar; cf. Theaitetos, 178d.Google Scholar

⁵⁰ NE, 1176a, 5–18Google Scholar; cf. Rhetoric, 1364b, 21.Google Scholar

⁵¹ I thank David Schmidtz and Jeremy Shearmur for pointing out to me the importance of this last fact in Mill's theory.

⁵² ‘Utilitarianism’, CW, xviii. 213.Google Scholar

⁵³ System of Logic, CW, viii. 952.Google Scholar

⁵⁴ This fact makes Berger's summary of Mill's point rather weak: ‘In this paragraph, Mill endorsed the position… that character development must not be ignored in moral reasoning’ (Berger, , p. 99).Google Scholar

⁵⁵ ‘Bentham’, CW, x. 95.Google Scholar

⁵⁶ Ibid., 98.

⁵⁷ ‘The Gorgias’, CW, x. 149.Google Scholar

⁵⁸ Ibid., 150. Cf. Aristotle, , NE, 1179b, 23–5.Google Scholar

⁵⁹ ‘Whewell on Moral Philosophy’, CW, x. 184.Google Scholar

⁶⁰ ‘Bentham’, CW, x. 95.Google Scholar

⁶¹ Cf. Plato, , Laws, 653b–cGoogle Scholar; Aristotle, , NE, 1180a, 14.Google Scholar

⁶² Autobiography, CW, i. 176.Google Scholar

⁶³ ‘On Genius’, CW, i. 338Google Scholar; Examination, CW, xi. 466.Google Scholar

⁶⁴ System of Logic, CW, viii. 842–3Google Scholar; ‘Utilitarianism’, CW, xviii. 235–9Google Scholar; Examination, CW, xi. 468Google Scholar. Mill appears to confuse these two types in his early ‘Remarks on Bentham's Philosophy’, CW, x. 12–13)Google Scholar, but they are carefully distinguished in later works.

⁶⁵ System of Logic, CW, viii. 842–3.Google Scholar

⁶⁶ ‘Utilitarianism’, CW, xviii. 217.Google Scholar

⁶⁷ Ibid., 217–18.

⁶⁸ Ibid., 211.

⁶⁹ Ibid., 239.

⁷⁰ Examination, CW, xi. 468.Google Scholar

⁷¹ ‘Utilitarianism’, CW, xviii. 236–7.Google Scholar

⁷² ‘The Gorgias’, CW, x. 150Google Scholar. Presumably this only happens when the choice is not made ‘knowingly and calmly’ (‘Utilitarianism’, CW, xviii. 213)Google Scholar; otherwise the anticipation of ‘psychic disease’ would have prevented such a choice in the first place.

⁷³ Or plus 10 disutils of self-image pain. For the purpose of simplification I am assuming that gaining a unit of pain is the same thing as losing a unit of pleasure.

⁷⁴ ‘Remarks on Bentham's Philosophy’, CW, x. 12–13.Google Scholar

⁷⁵ ‘Utilitarianism’, CW, xviii. 212.Google Scholar

⁷⁶ ‘Bentham’, CW, x. 95 and 98.Google Scholar

⁷⁷ It might be objected that, on my interpretation, the preference for the higher pleasures rests ultimately on the agent's own pleasure. This may be plausible enough if the choice is between poetry and pushpin; but Mill includes the ‘moral sentiments’—the unwillingness to be ‘selfish’ or a ‘rascal’—among the higher pleasures. And is not morality based on the happiness of others, not of oneself? Well, given Mill's theoretical commitment to utilitarianism, the justification of other-regarding virtues is the general happiness, not one's own. But Mill, refuses to ‘confound the rule of action with the motive of it’ (‘Utilitarianism’, CW, xviii. 219)Google Scholar. Under Mill's psychological hedonism, nobody can value anything except his own happiness—though of course the happiness of others can be a ‘part’ of his own happiness.

⁷⁸ ‘Utilitarianism’, usually taken as the authoritative statement of Mill's views, was only a popularizing magazine serial. It is regrettable that we have no System of Ethics from Mill, 's hand to match his System of Logic.Google Scholar