¹ Here and elsewhere the qualifier ‘non-harmful’ is often added, because it is usually assumed that for reasons unrelated to anything discussed in this article, certain courses of action are impermissible. However, it should be noted that the qualifier ‘non-harmful’ must be taken in a normative sense, because under certain neutral understandings of ‘harm’, many permissible courses of action are harmful. For example, when one produces a better product than a competitor and drives him out of business, one harms him in some way. And when one drives home after work and lets one's car emit exhaust, one harms others, if only in a very slight way. Deciding how to separate impermissible and permissible harms is obviously quite difficult. One option would be to think of ‘non-harmful’ as ‘not violating an agent-centered restriction’.

It should also be noted that most people assume that the existence of these agent-centered restrictions is unrelated to the truth of the Integrity Explanation, discussed below. For example, in The Rejection of Consequentialism, Samuel Scheffler argues that the Integrity Explanation justifies agent-centered permissions, but can't justify agent-centered constraints. The latter are to be justified, if at all, for some other reason. However, some people believe that the Integrity Explanation could justify constraints. For a discussion of this position, see Hurley's, Paul ‘Scheffler's Argument for Deontology’, Pacific Philosophical Quarterly 74 (1993)CrossRefGoogle Scholar.

² For Scheffler, see The Rejection of Consequentialism, rev. edn. (New York, 1982), esp. ch. 2; for Williams, ‘Integrity’, Utilitarianism: For and Against, ed. J. J. C. Smart and Bernard Williams (New York: Cambridge, 1973). See Harris's, John ‘Williams on Negative Responsibility and Integrity’, Philosophical Quarterly 24 (1974)CrossRefGoogle Scholar, for an argument that Williams is advocating something like the Integrity Explanation, as defined below. See also Michael Slote, Common-Sense Morality and Consequentialism (London, 1985), ch. 2. Slote advocates an explanation for the falsity of consequentialism that he believes is distinct from the Integrity Explanation. In fact he specifically says he is not appealing to ‘integrity’ at all. However, see Dan Brock's ‘Defending Moral Options’, Philosophy and Phenomenological Research 51 (1991), and especially Seana Shiffrin's ‘Moral Autonomy and Agent-Centered Options’, Analysis, 51.4 (1991), for reasons to think that Slote's explanation is either quite similar to the Integrity Explanation or at least similar enough that the arguments of this article would apply to it.

³ The most important explanation of the concept has been given by Scheffler. Instead of merely asserting that morality must not take away our integrity and leaving that notion undefined, he defends and elaborates the claim using an argument that may be briefly represented as follows: (i) each of us is concerned with our projects and interests out of proportion to their weight from an impersonal point of view, (ii) the having of this personal point of view is part of the nature of a person, and (iii) ‘the correct regulative principle for a thing depends on the nature of that thing’ (The Rejection of Consequentialism, p. 57). Step (iii) is affirmed by Scheffler, but credited by him to Rawls (see Theory of Justice (Cambridge, MA, 1971), p. 29). My preferred explanation is similar to one given by Shiffrin in ‘Moral Autonomy and Agent-Centered Options’. When one may blamelessly and rationally choose between two different life-paths, one has made use of one's capacity for choice and determined a future that was not independently required by the principles of rationality or morality. It is plausible that exercising one's capacity for choice in this way makes one's life better, and thus that for such exercise to be made impossible is rightly described as a loss of an important moral value. That value doesn't have a clear name in ordinary English, but it is plausibly described as ‘integrity’.

⁴ See The Limits of Morality, chs. 7, 8, and 9, but esp. pp. 372ff.

⁵ This requirement is fraught with difficulty, since as is often noted, many people agree that they must make such sacrifices when the good is present to mind, but resist the principle when applied to goods that are ‘out of sight’. For example, many people agree that they must ruin their $300 shoes to save a baby drowning in a puddle, but resist the idea that they must give $300 to charity to save a starving baby in the developing world. This article offers no special insights into how to resolve this paradox. For one solution that might help justify this puzzling aspect of ordinary morality, see Hershenov's ‘A Puzzle about the Demands of Ordinary Morality’, Philosophical Studies 102 (2002).

⁶ Not everyone agrees that we have this imperfect duty, and those who agree often disagree about what exactly it is a duty to do, and about how extensive it is. Those complications are left aside here because the arguments of this article would work no matter how those issues are resolved.

⁷ For example, Kagan has alleged that ordinary morality, properly understood, requires that we must promote either our own interests or the overall good. See Shelly Kagan, The Limits of Morality (Oxford, 1989), esp. ch. 1 and p. 240.

⁸ For more on satisificing consequentialism, see Slote's Common-Sense Morality and Consequentialism, ch. 3.

⁹ John Stuart Mill, Utilitarianism, in Utilitarianism and Other Essays, ed. Alan Ryan (London, 1987), p. 295. Mill gives the following guidance about how to pick our rules: ‘Again, defenders of utility often find themselves called upon to reply to such objections as this – that there is not time, previous to action, for calculating and weighing the effects of any line of conduct on the general happiness. . .. The answer to the objection is, that there has been ample time, namely, the whole past duration of the human species. During all that time, mankind have been learning by experience the tendencies of actions’ (p. 295). Interpreting this passage is difficult. But here and in surrounding remarks, Mill seems to adopt the view that we should adopt the moral standards that have been used by mankind throughout history unless we see clear, definitive reason to overturn them. If we followed Mill's advice, we might end up with a rule-consequentialist system that was more demanding than ordinary morality (because some of its precepts would be overturned), but still quite similar to it.

¹⁰ Slote highlights other disconnects between the Integrity Theory and Ordinary Morality – see Common-Sense Morality and Consequentialism, pp. 23ff.

¹¹ The Limits of Morality, pp. 235–6.

¹² Of course, if we loaf our lives away, then we will have failed to fulfill our imperfect duty of beneficence, and hence our loafing (at some specific moment) will be part of a pattern which constitutes a moral failure.

¹³ The reply here presupposes that critics do not believe that doing anything you choose to do ipso facto counts as pursuing your happiness. I do not find that position plausible at all, but I should admit that I have no definitive objection to it, and that the proof would fail if it were correct.

¹⁴ Slote discusses this way of seeing morality. See Common-Sense Morality and Consequentialism, ch. 2.

¹⁵ A related problem is discussed in Samuel Scheffler's The Rejection of Consequentialism, pp. 17–18.

¹⁶ I have Paul Hurley to thank for this point.

¹⁷ Cf. Kagan's The Limits of Morality, pp. 239ff.

¹⁸ For more on alternative theories of happiness, see Griffin's, JamesWell-Being (Oxford, 1986)Google Scholar.

¹⁹ Another reason why defenders of the Integrity Theory might try to vindicate ordinary morality is that they are hoping that morality turns out to be lax – for any reason whatsoever – because they don't want to be compelled to make larger sacrifices than they do. But while I'm sympathetic to the idea that this pragmatic reason is (unfortunately) behind much moral inquiry, it's obvious that this is not a good theoretical reason for ignoring the discord between the Integrity Theory and ordinary morality.