¹ For recent philosophical attempts to understand hypocrisy, see: Szabados, Béla, ‘Hypocrisy’, Canadian Journal of Philosophy, ix (1979), 195–210CrossRefGoogle Scholar; Kittay, Eva Feder, ‘On Hypocrisy’, Metaphilosophy, xiii (1982), 277–89CrossRefGoogle Scholar; Shklar, Judith, ‘Let Us Not Be Hypocritical’, Ordinary Vices, Cambridge, Mass., 1984, pp. 45–86Google Scholar; Newman, Jay, Fanatics and Hypocrites, Buffalo, NY, 1986Google Scholar; Turner, Dan, ‘Hypocrisy’, Metaphilosophy, xxi (1990), 262–9CrossRefGoogle Scholar; McKinnon, Christine, ‘Hypocrisy, With a Note on Integrity’, American Philosophical Quarterly, xxviii (1991), 321–9Google Scholar; Crisp, Roger and Cowton, Christopher, ‘Hypocrisy and Moral Seriousness’, American Philosophical Quarterly, xxxi (1994), 343–9Google Scholar; Béla Szabados and Eldon Soifer, ‘Hypocrisy after Aristotle’, Dialogue, forth-coming.

² The family resemblance approach is notable for its fluidity and its openness to the rich variety of cases and contexts in which we speak of hypocrisy. We are offered a loose set of conceptual similarities and affinities, accompanied by a focus on different but related examples, to guide us in the recognition and organization of the rich variety of hypocrisies in everyday life. The definitional approach, on the other hand, boldly seeks to offer a precise formulation of a set of necessary and sufficient conditions for what it is to be a hypocrite. This approach is less interested in exploring or extending the garden variety of hypocrisies in life and literature but is more concerned with giving a sharply delineated conceptual landscape. Each approach has its pluses and minuses. We hope to bypass such partisan methodologicalquarrels by offering a middle of the road working account which focuses on salient central conceptual features of hypocrisy as well as exploring interesting and instructive particular cases.

³ A related kind of hypocrisy might involve cases where a person is known to have stated or otherwise endorsed moral, religious, aesthetic, or political values as applying to everyone including himself or herself, yet on some crucial occasion violates these valuesin his or her behaviour for self-interested or ideological reasons, and then conceals such deviations so as to appear better in the eyes of those to be deceived or in his or her own eyes. This kind of hypocrisy seems to allow for less self-consciousness about the project of deception, and allows for various degrees of self-deception.

⁴ For discussions touching on the relationship between hypocrisy and integrity, see Taylor, Gabriele, ‘Integrity’, Proceedings of the Aristotelian Society, suppl. vol., lv (1981), 143–59CrossRefGoogle Scholar; McKinnon; Crisp and Cowton; Szabados and Soifer.

⁵ The criticisms of consequentialism based on these sorts of concepts wereperhaps most clearly raised by Bernard Williams in connection with utilitarianism's difficulties with integrity. See Williams, Bernard, ‘A Critique of Utilitarianism’, Smart, J. J. C. and Williams, Bernard, Utilitarianism: For and Against, Cambridge, 1973Google Scholar.

⁶ James Griffin is one person who uses this phrase. See his Well-Being: Its Meaning, Measurement and Moral Importance, Oxford, 1986, p. 13Google Scholar. It is worth noting that the kind of desire-fulfilment referred to here is fulfilment of what Griffin would call ‘informed desires’, not ‘actual desires’.

⁷ Kittay, 280.

⁸ Ibid.

⁹ Ibid., 279.

¹⁰ McKinnon, 328.

¹¹ Ibid.

¹² Ibid.

¹³ There is one tradition which sees egoism as a type of consequentialism, aiming at good consequences for the individual rather than the wider good. Sidgwick might be seen as one who makes such a connection between the two approaches, describing egoism and utilitarianism, respectively, as ‘egoistic hedonism’ and ‘universalistic hedonism’ (Sidgwick, Henry, The Methods of Ethics, London, 1907Google Scholar). The distinction suggested in the text is clear and common enough however. If one insists on seeing egoism as a species of consequentialism, then the subject of this paper can simply be redescribed as ‘Hypocrisy and Non-Egoistic Consequentialism’, and nothing of the argument will be damaged.

¹⁴ This possibility depends upon the now well-known distinction within consequentialism between a criterion of rightness and a decision procedure. (See e.g. Railton, Peter, ‘Alienation, Consequentialism, and the Demands of Morality’, Philosophy and Public Affairs, xiii (1984)Google Scholar; Parfit, Derek, Reasons and Persons, Oxford, 1984Google Scholar; and Crisp, Roger, ‘Utilitarianism and the Life of Virtue’, Philosophical Quarterly, xlii (1992)Google Scholar.) It could be that the best way to maximize good consequences is to do something other than aim at them. If this is a possibility, then one cannot conclude simply from the fact that hypocrites do not aim directly at maximizing the good that their actions do not in fact maximize the good. As will be argued below, there is indeed good reason to think that the actions in which hypocrisy consists may tend to have good consequences. Consequentialists in this case might have to offer two different evaluations of hypocrisy, at least one of which would be favourable. The possibility of such a ‘double evaluation’, and the scope it might offer consequentialists for resolving the problem posed by hypocrisy, is discussed more fully below.

¹⁵ There is an argument to the effect that egoists can never admit that they favour that position. This would be true if it would never be conducive to one's well-being to have others know one to be an egoist, for example because then they would not be willing to engage in social cooperation with one. Even if this is true, it is merely empirically so, and it remains at least conceptually possible that there could be such an acknowledged egoist, which is all the argument in the text requires.

¹⁶ Of course, it is likely that there have simply been more hypocrites than acknowl-edged egoists, and this might account for some of the discrepancy in terms of amountsof condemnation each has received, but we think there is more to it than that.

¹⁷ Smilansky, Saul has provided an interesting recent contribution (‘On Practicing What We Preach’, American Philosophical Quarterly, xxxi (1994), 73–9)Google Scholar which argues that people who say they will practice what they preach only if others do so as well are essentially hypocritical. His work, too, suffers however from a failure to explain whythe person who preaches one thing and practices another is worse than someone who practices the same thing but does not preach the contrary.

¹⁸ Kittay, 277.

¹⁹ See ‘The Hypocrites’ in Dante Alighieri's Inferno, trans. J. Ciardi, repr. in Vice and Virtue in Everyday Life: Introductory Readings in Ethics, ed. C. H. Sommers and F. Sommers, San Diego, Calif., 1985, pp. 255–9. For Joseph Butler's views, see his sermon, ‘Upon Self-Deceit’, in Sommers and Sommers, pp. 264–70Google Scholar.

²⁰ Walzer, Michael, Just and Unjust Wars, New York, 1977, p. xvGoogle Scholar; see also pp. 19f.

²¹ See Shklar, pp. 45–86.

²² This distinction might also be referred to in the well-known terminology of ‘expected’ and ‘actual’ utility, but the terminology in the text emphasizes the fact that the primary interest here is in one way of evaluating the agent as opposed to the agent's action.

²³ See Smart, J. J. C., ‘Extreme and Restricted Utilitarianism’, Philosophical Quarterly, vi (1956), 344–54CrossRefGoogle Scholar; revised version in Theories of Ethics, ed. Foot, P., Oxford, 1967, pp. 171–83Google Scholar.

²⁴ Conversely, consequentialists must disapprove to some degree of actionswhich do not aim at the good, but somehow accidentally achieve it.

²⁵ Recent interesting discussions of this possibility have included Adams, Robert Merrihew, ‘Motive Utilitarianism’, Journalof Philosophy, lxxiii (1976)Google Scholar; Railton, ‘Alienation’; Parfit, Reasons and Persons; Railton, Peter, ‘How Thinking about Character and Utilitarianism Might Lead to Rethinking the Character of Utilitarianism’, Ethical Theory: Characterand Virtue, ed. French, P., Uehling, T. and Wettstein, H., Midwest Studies in Philosophy, xiii (1988), pp. 398–416CrossRefGoogle Scholar; Johnson, Conrad, ‘Character Traits and Objectively Right Action’, Social Theory and Practice, xv (1989)Google Scholar; and Crisp, ‘Utilitarianism’.

²⁶ See Adams.

²⁷ More detail would be needed to complete this account. For example, it must be decided whether the test would be ‘have the best consequences if everyone had thisdisposition’, or ‘if most people did’, or ‘given the existing dispositions of others’, etc. The debate about this might parallel the discussion of how rule utilitarians should formulate their theory.

²⁸ There is a variety of ways to formulate ‘general acceptance’ – universal compliance, a simple majority, some number in between (e.g. 90% of people), and either counting or not counting the costs of education and conversion, etc. It is not necessary for present purposes to settle on one of these possibilities.

²⁹ See Williams, esp. pp. 108–18.

³⁰ McKinnon, 327.

³¹ Ibid., 321.

³² Ibid., 328.

³³ For a sustained argument about this, see Lyons, David, The Forms and Limits of Utilitarianism, Oxford, 1965CrossRefGoogle Scholar.

³⁴ J. J. C. Smart has famously levelled this accusation. See Smart, revised version, pp. 176–7.

³⁵ Indeed, Peter Railton (‘How Thinking About Character…’), offers an account of ‘character utilitarianism’ whereby it is ‘a theory of the lightness of individual acts … [which] no more than act utilitarianism makes the moral evaluation of an act depend upon the motive from which the act was actually performed’ (p. 400). On this account, the actions of the hypocrite are right so long as theywould also be done by someone with the disposition which would be best in consequentialist terms, even if those actions are not done out of those dispositions.

³⁶ In addition to the response in the next paragraph, we might also want to question the implicit comparison here. The ‘reliability’ point invites comparisonof the hypocrite with a good moral agent (one who is always motivated by the right sort of moral reasons). But in offering her example, Kittay clearly invites a different comparison, as we did above. This other comparison is with the person who admittedly acts against the norms – in this case, an unregenerate sexist. In that comparison, the hypocrite comes off well– if anything, he or she will perform good actions more reliably. And again, consequentialists should be able to do more than simply say which possibility is best – they should be able to rank the others as well.

³⁷ Bok, Sissela, Lying: Moral Choice in Public and Private Life, New York, 1978, pp. 26–7Google Scholar.

³⁸ Ibid., p. 31.

³⁹ Ibid., esp. pp. 24–7. It should also be noted that this tendency to undermine trust may be in a sense self-defeating, because hypocrisy (like lying) is much more difficult to practice once others' trust has been undermined.

⁴⁰ In a paper entitled ‘Taking on Trust’, delivered at the Canadian Philosophical Conference in St Catherine's, Ontario, June 1996. Much of the argumentof the next few paragraphs is indebted to Gombay's insightful paper.

⁴¹ Ibid., 4.

⁴² One might bear in mind here, however, Bok's observation (p. 25) that ‘it is inevitable that more frequent lies do increase the chance that some willbe discovered’.

⁴³ Gombay, 7.

⁴⁴ Bok, too, in effect acknowledges the role of merely suspected deceptionwhen she refers to ‘the impact of discovered and suspected lies on trust and social co-operation’ (p. 30, emphasis added). She does not appear to appreciate the importanceof this for her general argument, however.

⁴⁵ Kittay, 285.

⁴⁶ Acknowledging this fact about human nature does not commit us to the view, all too often taken, that people who are taken in by others are somehow to blame for being too naive. It is important to trust others, at least sometimes, even if one is aware that trusts can be betrayed, and it is no part of our intention to exonerate those who betray others' trust. One can acknowledge that the capacity to deceive is part of human nature without advocating a ‘buyer beware’ approach to human interactions in general.

⁴⁷ One persuasive version of such an account is offered by Griffin. The term ‘fulfilment’ is here being used in a technical sense, in which it stands opposedto ‘satisfaction’. (This distinction has been drawn by Ross, W. D., in his Foundations of Ethics, Oxford, 1939, p. 300Google Scholar, and utilized by Feinberg, Joel in his Harm To Others, Oxford, 1984, p. 84)Google Scholar. If one has a desire for x to be the case, the desire is fulfilled if and only if x does come to be the case; the desire is satisfied if and only if one believes that the desire has been fulfilled. Again, it should also be understood that the desires in question may need to be qualified through being ‘informed’ in some ways. The desire accounts are not the only alternative to mental state accounts here, and as Brad Hooker has pointed out to us objective list accounts (to borrow a term from Parfit, pp. 466,493, and 499, and Griffin, pp. 33–4) may have the same advantages as desire accounts when it comes to dealing with the problem of hypocrisy. We believe desire accounts to be preferable to objective list accounts on other grounds, but that debate is not crucial to the current discussion.

⁴⁸ See Griffin, p. 19.

⁴⁹ Some important contemporary works in this area include Brandt, Richard, A Theory of the Good and the Right, Oxford, 1979Google Scholar, esp. ch. 13; Parfit, esp. Appendix I; Griffin, esp. chs. 1–2; Sprigge, T. L. S., The Rational Foundations of Ethics, London, 1988Google Scholar, chs. 5–6; and Sumner, L. W., ‘Welfare, Preference, and Rationality’, Value, Welfare, and Morality, ed. Frey, R. G. and Morris, W., Cambridge, 1993Google Scholar.

⁵⁰ In fact, neither is the good Kantian agent, who must ensure that inclinations such as compassion are ruled by reason, and that charity is given out of respect to the moral law, not out of mere feeling or inclination. Virtue ethicists might here have a point against both major parties to the traditional debates about normative theories. See Williams, Bernard, Ethics and the Limits of Philosophy, Cambridge, Mass., 1985Google Scholar, for a sustained argument about the limitations of focusing our ethical discussion on what he considers to be a minor family squabble.

⁵¹ Pettit, Philip, ‘Consequentialism’, A Companion to Ethics, ed. Singer, P., Oxford, 1991, p. 231Google Scholar.

⁵² Kant, , Metaphysics of Morals, trans. Gregor, M., pt. I, p. 141 (standard pagination, p. 331), New York, 1991Google Scholar.

⁵³ This claim is reminiscent of the Williams-type claim that utilitarianism treats people merely as containers of utility.

⁵⁴ R. M. Hare has made such a suggestion. See his ‘Could Kant Have Been a Utilitarian?’, Utilitas, v (1993), 1–16Google Scholar.

⁵⁵ This name for it comes from Kittay, who offers a short but interesting discussion of the phenomenon, 287–9.

⁵⁶ Ibid., p. 287.

⁵⁷ That great literary critic of hypocrisy, Alceste, Molière's (in The Misanthrope, in The Plays ofMolière, trans. Waller, A. R., Edinburgh, 1926, v.)Google Scholar, seems to consider social niceties hypocritical and condemn them in the strongest terms. It iseasy to read the work as a criticism of Alceste's excessive anti-hypocrisy (as indeed Shklar does, see pp. 52–3), but doing so is tantamount to allowing that some hypocrisy may bemorally acceptable.

⁵⁸ One such individual might be Peter Railton's ‘sophisticated hedonist’ (or a consequentialist cousin) who is trying to cultivate non-instrumental friendships. Railton says of this individual: ‘He could then attempt to focus more on his friends as such, doing this somewhat deliberately, perhaps, until it comes more naturally.’ (‘Alienation’, 144.)

⁵⁹ On the moral significance of emotions, see Neblett, William, The Role of Feelings in Morals, Washington, DC, 1981Google Scholar, and Oakley, Justin, Morality and the Emotions, London, 1993, ch. 2Google Scholar.

⁶⁰ It is informative in this regard to consider Saul Smilansky's attempts to give a more or less Kantian account of the wrongness of failing to live up to one's pronouncements. There are frequent references throughout the article to ‘certain limited exceptions’ (p. 73), which are dismissed simply as being less frequent than might be supposed. The existence of any exceptions is problematic for the Kantian, however, and indeed an examination of the cases discussed seems to reveal that when exceptions do need to be made, it is on consequentialist grounds.

⁶¹ We are thankful for helpful comments on earlier versions of this paper from William Shaw, Brad Hooker, David Elliott, and those who participated in the discussion when this paper was presented at the University of Saskatchewan and at the International Society for Utilitarian Studies conference in New Orleans.