1 See, e.g., Flanagan, Owen, “Quinean Ethics,” Ethics, 93 (1982): 56–74.CrossRefGoogle Scholar

2 Though I shall not have space to discuss Gibbard's treatment of reasons one may have to be moral, it should be noted that, although he does not use this particular vocabulary, he does not adopt an “internalist” conception of the moral. That is, he does not assume that if we are obligated to do something, we thereby have reason, or are motivated, to do it. This is important, as the denial of internalism is an important part of the challenge some moral realists make to expressivism. See Brink, David, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press, 1989), chap. 3CrossRefGoogle Scholar. Gibbard is, however, an internalist about rationality—that is, that something is rational, in his sense of the term, means that one has reason to do it (p. 49). So Brink's externalist case against moral expressivism cannot be used against Gibbard's general normative expressivism.

3 Warren Quinn wrote me that this would be a caricature of, say, the views of Stevenson.

4 As Gibbard notes, co-ordination could as well be facilitated by capacities to grasp and communicate normative facts—were there such facts (p. 108). So it may be that a nonexpressivist realist account of the normative is compatible with Gibbard's evolutionary story.

5 Though it should be noted that some such analyses are “externalist” and do not consider the element of endorsement or approval to be essential to moral judgments. (See note 2 for a brief mention of some of these issues.)

6 I am indebted here to Warren Quinn for a thoughtful comment, as well as to David Copp for correcting a misunderstanding of Gibbard's analysis.

7 Questions of interpretation and ascription of belief are discussed in some detail by Gibbard (pp. 101–2, 156–60, 200–201).

8 The disagreements between Gibbard and these realists are complicated because of the latter's externalism, and they will require considerable care to sort out. (Again, see note 2.)

9 Mackie, of course, recognized this. Gibbard does as well; he also rejects Humean accounts of rationality as accounts of meaning (pp. 10–11).

10 Warren Quinn wrote me that Stevenson stressed that making value judgments was more than merely expressing one's state of mind, and that at least some of the classical noncognitivists understood adopting norms as applying to oneself independently of one's acceptance of them.

11 It is not clear, however, that contractarian moralists must be committed to an intersubjective account of non-moral normative judgments. On Gibbard's account, while it is true that evolutionary origin of our normative capacities has an intersubjective element, it need not be true that the force of some individual's normative judgments to himself or herself must be intersubjective.

12 Almost a decade ago, in conversation, Gibbard persuaded me to abandon my full-information account of rationality simply by inquiring what culinary preferences I would have, were I fully and vividly informed of the voyage of my dinner down my digestive tract. (But do we not think that, prior to going on a long trip through the desert in a car with a leaky radiator, we should have it checked, even if we do not know of the problem? David Copp, in a conversation about these issues, made me pause in my acceptance of Gibbard's argument.)

13 This may not be true of norms that function as co-ordination devices for coping with individual planning, referred to by Gibbard but not discussed in this work (see p. 100, n. 8). For this type of argument, see Bratman, Michael, Intentions, Plans, and Practical Reason (Cambridge, MA: Harvard University Press, 1987)Google Scholar, and McClennen, Edward F., Rationality and Dynamic Choice (Cambridge: Cambridge University Press, 1990).CrossRefGoogle Scholar

14 See Gauthier, David, Morals by Agreement (Oxford: Clarendon Press, 1986)Google Scholar. Gauthier's reflections about socialization and preference formation, however, may make him more sympathetic to Gibbard's argument for the rationality of mutual fundamental influence than my reading of his 1986 presentation of morals by agreement suggests.

Also, I might note that the Humean instrumentalism of most contractarians (e.g., Gauthier, Narveson, Mackie, Harman, James Buchanan) notwithstanding, there seem to be no reason why such theories must assume Humean value subjectivism (or so I argue in an unpublished essay).