Skip to main content Accessibility help

Mill, Sentimentalism and the Problem of Moral Authority



Mill's aim in chapter 3 of Utilitarianism is to show that his revisionary moral theory can preserve the kind of authority typically and traditionally associated with moral demands. One of his main targets is the idea that if people come to believe that morality is rooted in human sentiment then they will feel less bound by moral obligation. Chapter 3 emphasizes two claims: (1) The main motivation to ethical action comes from feelings and not from beliefs and (2) Ethical feelings are highly malleable. I provide a critical examination of Mill's use of these claims to support his argument that Utilitarianism can preserve morality's authority. I show how the two claims, intended to form a significant rebuttal to the worry about Utilitarianism, can in fact be combined to raise powerful skeptical concerns. I explain how Mill evades the skepticism, and why contemporary philosophers who lack Millian optimism about human nature find it harder to avoid the skeptical outcome.



Hide All

1 Mill, John Stuart, Utilitarianism, ed. Crisp, Roger (Oxford; New York: Oxford University Press, 1998), p. 75.

2 See Crisp, Roger, Mill on Utilitarianism (London and New York: Routledge, 1997), p. 90, and Korsgaard, Christine, The Sources of Normativity (Cambridge: Cambridge University Press, 1996), pp. 7886. My discussion of Mill on these matters is clearly indebted to Korsgaard's discussion. I speculate that one reason for chapter 3's comparative neglect is the influence of an idea that (ironically) one sees emerging in inchoate form in the chapter itself: the idea that metaphysical and epistemological considerations are irrelevant or neutral with regard to moral claims. This idea became conceptually (and, later, professionally) enshrined in the distinction between metaethics and normative ethics.

3 Mill, Utilitarianism, p. 73.

4 The two kinds of force need to be separated in order to discuss the relationship between how moral people are and how moral people should be. One could hold that people are not naturally suited to living morally and yet still think that moral concerns should, as a normative matter, be given the highest importance. Alternatively, one might think that humans are easily ‘moralized’ but consider them mistaken, from a normative point of view, in giving high priority to the demands of morality.

5 Rawls, John, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), p. 455.

6 See, for example, Smith, Michael, The Moral Problem (Oxford: Blackwell, 1994), p. 7. Note that we can think of the motivational power of morality as a given and then use this fact to test the explanatory appeal of rival moral theories. Alternatively, we can think of the practicality requirement as a normative demand on moral theories: i.e. they ought to sustain our ethical motivation.

7 Mill, Utilitarianism, p. 73.

8 As Simon Blackburn puts it in Ruling Passions (Oxford: Oxford University Press, 1998), many think that his form of sentimentalism ‘smells of sulphur’ (p. vi). The accusation against sentimentalism, as his evocative phrase suggests, has a long history.

9 Korsgaard, Sources, p. 14.

10 See Crisp's notes to chapter 3 in Mill, Utilitarianism, p. 128.

11 Mill, Utilitarianism, p. 74.

12 Mill, Utilitarianism, p. 75.

13 Gilbert Harman uses a similar kind of argument to Mill in his ‘Moral Relativism’, Moral Relativism and Moral Objectivity, ed. G. Harman and J. J. Thompson (Cambridge Mass.: Blackwell, 1996), although Harman's argument is on behalf of his moral conventionalism. Harman writes: ‘[I]t might be thought that morality loses its normative force if it does not have a nonconventional source. But that is just a mistake if moral conventionalism is true [for according to it] there is no such thing as objective absolute normative force. Morality does not lose that sort of normative force, because it never had it in the first place’ (pp. 27–8). You can see Harman's point: if conventionalism is true, it is true already. But note also the problem with the argument: if moral conventionalism comes to be believed, then arguably something will be lost that was present in moral practice in the first place, namely the belief that morality has a non-conventional source. And the skeptical worry concerns what might follow from this lost belief.

14 Mill, Utilitarianism, p. 76.

15 Or, to adapt a point of Anscombe's, whenever you want to unbind yourself, can't you always overturn the legislation by a vote of 1–0? Anscombe, of course, notoriously believed that any notion of self-legislation was incoherent because legislation requires a ‘superior power’ in the legislator. See Anscombe, G. E. M., ‘Modern Moral Philosophy’, Virtue Ethics, ed. Crisp, Roger and Slote, Michael (New York: Oxford University Press, 1997), p. 27. She names Kant as her target but does not consider, even to reject it, the Kantian idea that one aspect of the self can legislate over another.

16 See Crisp's notes to chapter 3 in Mill, Utilitarianism, pp. 129–30.

17 Terry Horgan and Mark Timmons provide an excellent overview of this experiential or phenomenological demand on moral theories in ‘Moral Phenomenology and Moral Theory’, Philosophical Issues 15 (2005), pp. 56–77.

18 Mackie, J. L., Ethics (Harmondsworth: Penguin, 1977), p. 32.

19 Wright, Crispin, ‘Truth in Ethics’, Truth in Ethics, ed. Hooker, Brad (Oxford: Blackwell, 1996), p. 17.

20 This, in a sense, is Kant's view, although of course the internal/external distinction is, in another sense, preserved by Kant, albeit now located inside the self. I discuss this distinction in relation to moral skepticism in section 5 of my ‘Internalism and the Self: Lessons from Korsgaard's Kantian Critique of Williams’, Southwest Philosophy Review 23.1 (Jan. 2007).

21 Shakespeare, William, Hamlet, ed. Hibbard, G. R. (Oxford: Clarendon Press, 1987), p. 216.

22 See my introduction to Reading Bernard Williams, ed. Daniel Callcut (London: Routledge, 2009).

23 Note, however, that one could argue that the value given to self-interest is indeed underpinned by a mistaken metaphysics of the self. Derek Parfit argues this case in Reasons and Persons (New York: Oxford University Press, 1984). See also Susan Wolf's rejoinder to Parfit: ‘Self-Interest and Interest in Selves’, Ethics 96 (July 1986), pp. 704–20. More generally, a number of political theorists have argued that modernity is characterized by, in C. B. Macpherson's phrase, a ‘possessive individualism’ that has metaphysical components.

24 Wright, ‘Truth in Ethics’, p. 5.

25 Charles Taylor has defended this view. See, for example, his ‘Ethics and Ontology’, The Journal of Philosophy, C.6 (June 2003), pp. 305–20. George I. Mavrodes argues for the importance of the ‘deep grounding’ (p. 220) of morality in ‘Religion and the Queerness of Morality’, Rationality, Religious Belief, and Moral Commitment, ed. Robert Audi and William J. Wainwright (Ithaca: Cornell University Press, 1986). He defends the following claim (one that he associates with Platonism, Christianity, and, in this quotation, Kantianism): ‘there cannot be, in any “reasonable way”, a moral demand upon me, unless reality itself is committed to morality in some deep way. It makes sense only if there is a moral demand on the world too and only if reality will in the end satisfy that demand’ (p. 220). Mavrodes argues, in effect, for the exact thesis that Mill aims to rebut, viz. that taking morality seriously is incompatible with the metaphysics of modern scientific naturalism.

26 One might object, as Kant did, that the desire to avoid punishment (or receive reward) is not really a moral reason: nonetheless, as Kant also acknowledged, it is a reason to be moral (that is, to conform one's actions with moral norms). The presence of such metaphysical reasons within various models of morality – and what the consequences might be of removing such metaphysical guarantees (and thus exposing reasons to be moral to contingency and luck) – has been subjected to a deep exploration in the work of Bernard Williams. See, in particular, his ‘Persons, Character and Morality’ and ‘Moral Luck’, Moral Luck (Cambridge: Cambridge University Press, 1981) and Ethics and the Limits of Philosophy (Cambridge, Mass.: Harvard University Press, 1985).

27 Mill clearly has Christianity primarily in mind, although he does cite other religions in Utilitarianism.

28 Mill, Utilitarianism, p. 74, emphasis added. One can see how remarks such as this point to the development (alluded to in n. 2 above) of a separation of the internal substantive content of morality from such matters as its metaphysical status. See also Mill, Utilitarianism, p. 68.

29 There is a whiff of what Bernard Williams called ‘Government House Utilitarianism’ (which, in a nutshell, means: utilitarianism for the learned; whatever works best for the vulgar). See Williams' Ethics and the Limits of Philosophy, pp. 108–10.

30 Mill, Utilitarianism, p. 73.

31 Mill, Utilitarianism, p. 77.

32 Mill, Utilitarianism, p. 77.

33 Notice how anxiety over this question plays an important backdrop role in debates over genetic engineering and (so-called) enhancement technologies more generally since new technologies dramatically increase the power to change human nature (including human emotions).

34 John Stuart Mill, ‘On Liberty’, p. 58; The Basic Writings of John Stuart Mill (New York: Modern Library, 2002).

35 Mill, Utilitarianism, p. 77.

36 Mill writes in Utilitarianism: ‘[A]s men's sentiments, both of favour and of aversion, are greatly influenced by what they suppose to be the effects of things upon their happiness, the principle of utility . . . has had a large share in forming the moral doctrines even of those who most scornfully reject its authority’ (p. 51). Mill argues that application of the utility principle admits of ‘indefinite improvement’ (p. 70), but it is clear that he expects the pace of social improvement to increase once the ‘tacit influence’ (p. 51) of the utility standard on conventional morality (and on rival moral philosophies) is recognized and consciously affirmed.

37 Williams, Bernard, ‘The Primacy of Dispositions’, Philosophy as a Humanistic Discipline (Princeton: Princeton University Press, 2006), p. 73. Notice that Williams is writing from the perspective of someone for whom moral naturalism is a revisionary position: this is the person for whom moral obligation might not have the resonance it once possessed. What about the person who was raised as a moral naturalist, we might wonder?

38 John Skorupski has rightly stressed the importance of the progressive conception of human nature to Mill's thought: see his John Stuart Mill (London and New York: Routledge, 1989), pp. 22–3, and his recent and engaging Why Read Mill Today? (London and New York: Routledge, 2006), ch. 2. For more on Williams, see my doctoral thesis, ‘Bernard Williams and the End of Morality’ (Baltimore: Johns Hopkins University, 2003).

39 T. M. Scanlon, ‘Metaphysics and Morals’, Presidential Address to the Eastern Division of The American Philosophical Association, Proceedings and Addresses of The American Philosophical Association 77.2 (Nov. 2003), p. 19.

40 Mill, Utilitarianism, p. 79. I have presented material from this article to audiences at Lewis and Clark College, UCL, Oxford University, Florida International University, Florida State University, and the University of North Florida. I am grateful for the criticisms and suggestions I received at these institutions. I am also grateful for the financial support and intellectual stimulation I received in the summer of 2005 (at Yale Law School) and 2006 (at the Berlin-Brandenburgische Akademie der Wissenschaften) as a SIAS Summer Institutes Fellow. I owe special thanks to Jennifer Fisher, Josh Gert, Chris Grau, Brenda Heideman, Bert Kögler, Jerry Schneewind and Susan Wolf for conversations on these issues.

Mill, Sentimentalism and the Problem of Moral Authority



Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed