1 Rawls, John ‘The Independence of Moral Theory,’ Proceedings and Addresses of the American Philosophical Association 68 (1974-75) 5–22CrossRefGoogle Scholar, and A Theory of Justice (Cambridge, MA: Harvard University Press 1971)

2 Cf. Daniels, Norman ‘Wide Reflective Equilibrium and Theory Acceptance in Ethics,’ Journal of Philosophy 76 (1979) 256–82CrossRefGoogle Scholar.

3 For example, Brandt, Richard A Theory of the Good and the Right (Oxford: Clarendon Press 1979)Google Scholar, Hare, R.M. ‘Rawls’ Theory of Justice-I,’ Philosophical Quarterly 23 (1973) 144–55CrossRefGoogle Scholar, and Singer, Peter ‘Sidgwick and Reflective Equilibrium,’ The Monist 58 (1974) 490–517CrossRefGoogle Scholar.

4 Chisholm, R.M. The Foundations of Knowing (Minneapolis: University of Minnesota Press 1982)Google Scholar, Ch. 5; The Theory of Knowledge, 2nd ed. (Englewood Cliffs: Prentice Hall 1977), Ch. 7; and The Problem of the Criterion (Milwaukee: Marquette University Press 1973)

5 Chisholm, The Problem of the Criterion, 12

6 Ibid., 14

7 Perhaps the most significant point Chisholm makes about methodism, particularism and scepticism is that there is no way of proving which of these positions is correct. When a proponent of any attempts to do so he or she will eventually be forced to beg the question against the others. Although I shall try to add to Chisholm's taxonomy and even to argue for one of the positions I describe, I do not claim to avoid this point.

8 Prichard, H.A. ‘Does Moral Philosophy Rest on a Mistake?’ reprinted in Sellars, Wilfrid and Hospers, John eds., Readings in Ethical Theory (Englewood Cliffs: Prentice-Hall 1970) 86–96Google Scholar

9 Ross, David The Right and the Good (Oxford: Clarendon Press 1930)Google Scholar

10 Some might think that my argument cannot be very strong, since the non-sceptical methods I consider do not exhaust the field. In particular, one might think that the approaches of authors, e.g., Hare, R.M. Moral Thinking (Oxford: Clarendon Press 1981)CrossRefGoogle Scholar, and Richard Brandt, A Theory of the Good and the Right, who attempt to deduce moral principles, or establish them in some other way, on the basis of non-moral propositions do not fit into my taxonomy. I maintain that such positins can be fit into my taxonomy, and hence, that my defence of coherentist methods is quite strong. However, because of space limitations, I shall not be able to argue here for this claim, so my ‘official’ position must be that I offer a partial defence of coherentism.

11 The statement that all killings of Jews at Auschwitz in 1944 were wrong selects a group of actions that are wrong by appeal to a property that is accidental to the wrongness of the action, i.e., the time and place of the killing. For if the world were very different it might have been that only one Jew was killed at that place during that year, and that this person was killed in self defense.

12 This is no benign assumption. For one thing, by understanding the answer to B in this way I side with the second of the two questions that make up B, the question that asks for a criterion of right action. In fact it is not even clear that I side with this member of the pair, since my specification of B* is somewhat weaker than Chisholm's traditional understanding of a criterion would require. Chisholm adopts Cardinal Mercier's requirement that an answer to B be internal, objective and immediate. Chisholm, The Problem of the Criterion, 6-8 and 36-7. The first question of the pair seems to ask for a procedure that one might follow in order to determine which actions are right. It is, therefore, more general than the second question in the pair, since a criterion of right action would provide only one sort of procedure for determining which actions are right, i.e., find the actions that have the criterial property. Other procedures are possible, as illustrated by the many moral methodologies that philosophers have proposed. For example, one might be inclined to think that Rawls’ method of reflective equilibrium provides an answer to B. We are to decide whether an action is morally right by bringing our moral beliefs into a state of wide reflective equilibrium, and then seeing whether our moral beliefs in wide reflective equilibrium entail that the action is right. The simplest way I can explain why we cannot understand question B in this very broad and general way is to point out that it collapses the distinction between A and B, and thus, between particularism and methodism. For surely even the particularist proposes a procedure in the broad sense. He tells us to take our particular moral beliefs, and if one of these doesn't concern the case we are interested in, generalize from these cases and apply the principle this generalization yields to the case in question.

13 For simplicity I here ignore problems that might arise if real moral dilemmas are possible. I assume any such problems could be resolved simply by amending the definition of consistency. Although I do not think the definition of consistency she proposes is adequate, on this issue see Marcus, Ruth ‘Moral Dilemmas and Ethical Consistency,’ Journal of Philosophy 77 (1980) 121–36CrossRefGoogle Scholar.

14 I intend by ‘determine’ a logical relation that will be deductive in the case of B* and inductive for A*. For a discussion of a weaker way in which one might wish moral principles to determine the moral status of actions see Rawls, John ‘Outline of a Decision Procedure for Ethics,’ Philosophical Review 60 (1951) 184–6CrossRefGoogle Scholar, who requires that moral principles ‘explicate’ particular moral judgments.

15 Dan Brock has pointed out to me that there is a serious question here about whether completeness with respect to any determinate class of actions should be required of the beliefs which are to base moral inquiry. The point is not that there might be some ‘smaller’ class of actions that a moral theory must be able to deal with if it is to be adequate. Rather, it is that there might be indefinitely many different and possibly disjoint classes of actions such that, a moral theory should be considered adequate if it could deal with the actions in one of these classes. I can respond to this objection only by pointing out that, even if we might be willing to settle for less in the end, ideally we would want a moral theory to be complete in the sense I've identified. I do not think I will go too far wrong if I begin by considering methods that assume this ideal to be readily attainable. I hope it will be clear below which methods are wedded to this naive assumption, and which are compatible with the more sober estimation that we certainly do not begin with complete moral beliefs, and perhaps never will be able to attain them.

16 One exception is provided by Ernest Sosa who has commented favorably on a classical intuitionist position regarding epistemic justification (’Foundations of Foundationalism,’ Nous 14 (1980] 554-6 and 563-4).

17 The same point is made in Sosa, 554-5.

18 A position that is perhaps slightly less implausible might be obtained by weakening the epistemic status claimed for our moral beliefs and claiming that they are justified rather than known.

19 ‘Outline of a Decision Procedure for Ethics’

20 Since Rawls limits the class of starting points to particular beliefs, the method he describes is in fact a form of particularism. In order to extend the spirit of his restrictions to coherentism and methodism I shall omit or alter this and a few others of his restrictions.

21 Rawls, ‘Outline of a Decision Procedure for Ethics,’ 180

22 Ibid., 183

23 The appearance may be deceptive. Stefan Sencerz has recently argued that the ‘filtering’ involved in obtaining considered moral judgments does not insure that these judgments will have a significant positive epistemic status (Philosophical Studies 50 (1986] 77-95).

24 The method endorsed by Rawls is just such a version of particularism.

25 Others obviously have a different opinion about the chance for intersubjective agreement, e.g., Norman Daniels, ‘Wide Reflective Equilibrium and Theory Acceptance in Ethics,’ but I do not think it would be helpful to argue the point. For one thing, it is an empirical question that should not be decided on the basis of philosophical speculation. More importantly, I do not think that proving my claim about the likelihood of intersubjective agreement is essential to my project. My aim in this paper is to categorize certain approaches to the construction of moral theories and to provide some reasons for thinking that one of these is more promising than the others. My interest, then, is in part strategic, and adopting a method which can be successful only if a controversial empirical assumption turns out to be true when one can avoid doing so is a poor strategy indeed.

26 I should also note that bringing philosophical beliefs that are not ‘moral’ to bear upon moral theory construction in the way here suggested is characteristic of wide reflective equilibrium, the version of coherentism I shall end up endorsing. I shall therefore have occasion to return to this point when we consider that version of coherentism below.

27 Ignoring, of course, the off chance that he is equally committed to both.

28 I am indebted to a number of persons for discussions of these issues and comments on earlier versions of this paper, most especially Philip Quinn, Richard Foley, Dan Brock, David Solomon and the referees for the Canadian Journal of Philosophy.