* I am indebted to Merrilee Salmon, Jeffrie Murphy, Keith Lehrer, and especially Ronald Milo for their helpful comments on an earlier draft of this paper.

1 See, for instance, Kant's examples of the first formulation of the categorical imperative.

2 Nagel, Thomas The Possibility of Altruism (Oxford: Clarendon Press, 1970).Google Scholar

3 Gauthier, David Practical Reasoning (Oxford: Clarendon Press, 1963).Google Scholar

4 Grice, Russell The Grounds of Moral Judgement (Cambridge: University Press, 1967).Google Scholar

5 This confusion seems pervasive enough that is somewhat arbitrary to single out any author as suffering from it. Grice surely confuses these two senses of 'reason’ in the previously cited work. D. A. J. Richard seems also to muddle the distinction because he treats rationality as if it is a substantive point of view and claims that moral reasons are superior to reasons derived from it.

6 This is true because someone might hold that if this person were to mistake a political reason for some other kind of reason, and hence to believe that he had some sort of non-political Justification for his action, his action may be (in some subjective sense) Justified.

7 I will here and in the future speak of whether prudential (economic, moral, etc.) reasons are reasons for every person. I shall understand this to be the question of whether such reasons are agent-perspective reasons for every person (that is, whether they are tied to what I shall call formal rationality). When people have asked whether moral reasons are reasons for every person they have meant to ask whether they are agent-perspective reasons for every person. It is trivially true that they are reasons in some sense of the word.

8 This sense of “self-regarding” is narrower than that employed by Broad, C. D. in his piece, “Egoism as a Theory of Human Motives” (The Hibbert Journal 48, (1949·50), pp. 105–114;Google Scholar Reprinted in Broad, C. D. Ethics and the History of Philosophy, London; Routledge ' Kegan Paul, Ltd., 1952Google Scholar). Nevertheless, it is a perfectly natural sense of the word.

9 A similar point must be noted regarding ‘imprudent'. Due to the negative suggestions of ‘imprudent’ we will not be inclined to call a person ‘imprudent’ if he sacrifices his own well-being for some laudable good.

10 It is this point that William Davie fails to appreciate in his article, “Being Prudent and Acting Prudently,” American Philosophical Quarterly. 10 (1973). Here he argues that there can be no conflict, as Kant supposed there could be, between morality and prudence. This implausible claim is defended by arguing that we would not call the perpetrator of a grossly immoral act ‘prudent'. If this is true, it is due to implicatures generated by our calling him ‘prudent', not to implications of ‘prudent'.

11 Sometimes, of course, we are interested in doing this. Psychiatrists, for example, are often concerned with assessing whether an action is Justified from the agent's perspective.

12 Butler, Joseph Five Sermons (London, 1726),Google Scholar Sermon IV. Paragraph 7.

13 Hobbes, Thomas Leviathan, ed. by Oakeshott, Michael (London: Collier-Macmillan Ltd., 1969), p. 105.Google Scholar

14 This point was, perhaps, first made by Darwin, Charles in The Descent of Manand Selection in Relation to Sex (London, 1871).Google Scholar See, in particular, chapter 5, “Of the Development of the Intellectual and Moral Faculties During Primeval and Civilised Times”.

15 Russel Grice. op. cit., p. 16.

16 Ibid. p. 16.

17 Hare, R. M.. Freedom and Reason (Oxford: Oxford University Press, 1972) Chapter 9.Google Scholar

18 Nagel, Thomas. The Possibility of Altruism. (Oxford: Clarendon Press, 1970). See, for example, page 79.Google Scholar

19 This view is suggested by Plato's protracted arguments in Gorgias and The Republic to show that the virtuous man is happier than the vicious man. Although in “Moral Beliefs” Phillipa Foot gives a correct account of agent-perspective reasons, some of what she says there suggests that she then thought that showing that morally right actions were also prudent would guarantee that everyone has a reason to be moral. She now clearly rejects both this view and the view that there always are prudential reasons for acting morally.