² I take the natural reading of this to say more than that one should do only what is good, but rather that one should do what is most good, or something like that, as will be explained below.

³ See Scanlon, , ‘Utilitarianism and Contractualism’ Utilitarianism: for and Against, Williams, B. and Smart, J. J. C. (eds.) (Cambridge: Cambridge University Press, 1973)Google Scholar. One of its most determined and influential opponents, Elizabeth Anscombe, appears never to have felt this appeal. She rejected anything like the promotion of value thesis silently, not feeling the need to analyse its underlying mistakes, or to dispel its magic power. The nearest she came to an argued rejection is in the brilliant and enigmatic page and a half comment on a paper by Mrs Foot, P., published in the Oxford Review 1967Google Scholar.

⁴ In the hands of Rawls, both (A Theory of Justice (Cambridge: Harvard University Press, 1971Google Scholar)) and Scanlon, (What We Owe to Each Other (Cambridge: Harvard University Press, 1998))Google Scholar.

⁵ On some views these ‘indirect reasons’ will not be additional reasons; rather they are the only reasons. I think that this view is mistaken. It may be motivated by the thought (mentioned in the text below) that it is a necessary condition for anything being a reason that one can, in principle, be guided by it. But this thought does not require us to accept that ‘the indirect reasons’ are the only reasons. We can be motivated to perform the action by the original reasons, and the case for recognizing indirect reasons requires nothing more than that one would be more likely to conform to those reasons if one is not guided by them, but by some alternative considerations. In such cases it is natural to take the alternative consideration to be a special kind of instrumental reason enabling one to conform to the source reason, rather than replacing it. My analysis of the way rules issued by legitimate authority function (see The Morality of Freedom (Oxford: Oxford University Press, 1984Google Scholar) Ch. 3) is an example. If I am right then this is an additional objection to the view which says (stated very roughly) that valid reasons are those facts which, if people take to be valid reasons, and try to follow them, would lead to the most desirable actions being performed. However, the complications of the indirect interpretations of the promotion of value thesis are too many to encompass here.

⁶ Indeed, sometimes it may be better if one conforms to the reason without being guided by it (as when it is better to avoid raping people without ever contemplating whether to do so).

⁷ Or to its modulation either through the introduction of some reasons which do not derive from the value of the actions for which they are reasons, or by relativizing values or reasons to agents in the way introduced by Nagel, T. in The Possibility of Altruism (Oxford: Oxford University Press, 1969)Google Scholar, given these names by Parfit, D. (Reasons and Persons (Oxford: Oxford University Press, 1984) p. 143)Google Scholar, and developed and applied to values as well as to reasons by Sen, A. in ‘Rights and Agency’ (Philosophy and Public Affairs, 1982)Google Scholar.

⁸ See in particular The Morality of Freedom Ch. 13, and Engaging Reason (Oxford: Oxford University Press, 2000)Google Scholar, Chapters 3 and 4.

⁹ And the trains will depart shortly so no further information can become available in time to affect the case.

¹⁰ I ignore the impact of the possibility, where it exists, that due to change in circumstances the conflict may disappear.

¹¹ It is possible, of course, to give a relatively formal characterization of ‘dependent reasons’, but thus understood the first premise, and therefore maximization, is false, unless subjected to additional exceptions, themselves based on substantive evaluative claims.

¹² See my account of duties of respect in Chapter four of Value, Respect and Attachment.

¹³ This is the assumption, i.e. that the law generally is morally acceptable.

¹⁴ The examples rely on the fact that we have duties not to rape and not to murder, independently of the law and of the promises.

¹⁵ Some people think that what is special about murder and rape is that the reasons to refrain from them are maximally strong, and therefore cannot be enhanced by the law or by promises. This is a supposition that cannot be accepted without other theoretical claims. Otherwise, it will turn out to assume not only that the reasons against rape and murder are of equal stringency, but also that that the reason to avoid killing one person is as strong as the reasons not to kill a million. These conclusions or the additional assumptions needed to avoid them may well be true, but are unlikely to give any support to the promotion of value thesis.

¹⁶ Some maximizers analyse practical reasons in terms of expected utility, and, of course, the expected utility of the combined occurrence of two events is not the sum of the expected utility of each of them occurring.

¹⁷ When acting in (or out of) friendship the friendship is a (part of the) reason for the action, but the reason is not necessarily for the sake of the friendship, understood as aiming to protect it from erosion, or other dangers.

¹⁸ See, e.g. Nozick's, famous experience machine in Anarchy, State and Utopia (NY: Basic Books, 1974)Google Scholar.

¹⁹ See my discussion of things which are valuable in themselves in Value, Respect and Attachment (Cambridge University Press, 2001)Google Scholar Chapter 4.

²⁰ Notice that having reinterpreted maximization as well as the promotion of value as principles justifying actions, without being principles about what reason we have, I am not claiming that they are right only if we can act in order to conform with them. The objection is based on the weaker assumption, namely that the principles which justify actions must be co-ordinated with the doctrine of reasons for action, i.e. that acting for sufficient reasons will not turn out to be unjustified, and that whether or not it is justified will not be a matter of chance.

²¹ The Morality of Freedom, Ch. 12, Engaging Reason, Chs. 12 and 13, Scanlon, , What We Owe to Each OtherGoogle Scholar, Ch. 3.