1 G.E. Moore suggested a similar account when he said that naturalistic ethics holds that “Ethics is an empirical or positive science: its conclusions could all be established by means of empirical observation and induction.” See Moore, G.E.Principia Ethica (Cambridge: Cambridge University Press, 1903), sec. 25Google Scholar. Naturalism is often explained differently, however, in terms of the sciences, or in terms of the entities postulated in the sciences, or in terms of the vocabulary of the sciences, or in terms of certain favored sciences. Moore says, in Principia, sec. 26, “By ‘nature,’ then, I do mean and have meant that which is the subjectmatter of the natural sciences and also of psychology.” I do not see why philosophers would give a privileged place to the sciences in explicating naturalism unless they thought that the scientific method was at least an especially reliable way of acquiring knowledge a posteriori. Notice that, on the proposal made in the text, a naturalistic theory need not be “reductive.” Notice also that a theory that is putatively naturalistic can be unsuccessful in a variety of ways. It might fail to make good on the claim that moral propositions are knowable, or that they are knowable a posteriori; it might propose an implausible analysis of moral propositions.

2 More would obviously need to be said in order to give an adequate account of the a posteriori. By a “synthetic proposition” I mean a proposition that is neither logically true nor “conceptually” true. If, for example, the concept of murder is the concept of a wrongful killing, a naturalist would not deny that we can know a priori that murder is wrong. But a naturalist denies that there is synthetic a priori moral knowledge. A naturalist would deny that we can know a priori that, say, killing the innocent is wrong. The notion of a proposition “about the instantiation” of a property is vague. Clearly, if a proposition implies that a property is instantiated, it counts as being” about the instantiation” of the property. A proposition is also in the relevant sense “about the instantiation” of a property if it implies a proposition about the circumstances in which the property would be instantiated. Consider, for example, the proposition that friendship is good. It is “about the instantiation” of goodness since it implies that if there is friendship, it is good. G.E. Moore therefore counts as a non-naturalist. He holds that we can know a priori that friendship is good. See Moore, Principia Ethica, sees. 112-3. He also holds that the proposition that friendship is good is synthetic. Moore, Principia Ethica, sec. 6.

3 The objection is briefly sketched in Scanlon, T.M.What We Owe To Each Other (Cambridge: Harvard University Press, 1998), 1Google Scholar. See also Audi, Robert “Intuitionism, Pluralism, and the Foundations of Ethics,” in Moral Knowledge? New Readings in Moral Epistemology, ed. Sinnott-Armstrong, Walter and Timmons, Mark (New York: Oxford University Press, 1996), 101–36Google Scholar, especially, 114-5. For brief discussion of a similar objection, see Railton, Peter “Moral Realism,” Philosophical Review 95 (1986): 166–8CrossRefGoogle Scholar.

4 An argument that assumptions about moral facts are irrelevant to explaining any observations is found in chapter one of Harman, GilbertMorality (New York: Oxford University Press, 1977)Google Scholar.

5 Harman, Morality, 4-8Google Scholar.

6 In this sentence I allude to three naturalistic proposals: a form of analytic consequentialism, the view I proposed in my recent book, and a close relative of the view T.M. Scanlon proposed in his recent book. See Copp, DavidMorality, Normativity, and Society (New York: Oxford University Press, 1995)Google Scholar, and Scanlon, T.M.What We Owe To Each Other, 4Google Scholar. Scanlon does not intend to propose a form of ethical naturalism.

7 Quine, W.V. “Epistemology Naturalized,” in Ontological Relativity and Other Essays (New York: Columbia University Press, 1969), 69–90CrossRefGoogle Scholar.

8 Quine, “Epistemology Naturalized,” 74-6, 82–3.Google Scholar

9 Here I agree with Barry Stroud and Jaegwon Kim. Stroud, Barry “The Significance of Naturalized Epistemology,” in Naturalizing Epistemology, ed. Hilary Kornblith (Cambridge: MIT Press, 1985), 71–89Google Scholar; Kim, Jaegwon “What is ‘Naturalized Epistemology'?” in Supervenience and Mind: Selected Philosophical Essays (Cambridge: Cambridge University Press, 1993), 216–36CrossRefGoogle Scholar. Perhaps it will be objected that normative epistemological issues, properly understood, are psychological. On this way of thinking, we would perhaps need to rethink the import of Quine's view that “Epistemology, or something like it, simply falls into place as a chapter of psychology and hence of natural science.” I take Quine to be claiming that we should give up normative epistemology as it has been practiced and instead confine ourselves to the use of scientific methodology in exploring the relation between evidence and theory. But the view that normative epistemological issues are psychological could be understood to imply instead that we should expand our conception of psychology, natural science, and the scientific method so that traditional philosophical explorations of normative issues are counted as “scientific” or “psychological.” This does not appear to be a substantive suggestion. Note: The thesis that all normative properties are natural properties does not imply that normative epistemology is” a chapter of psychology“; what it implies is that any normative epistemological knowledge is fundamentally empirical. In more recent work, Quine appears to accept that epistemology is a normative discipline. See Quine, W.V.O.In Pursuit of Truth (Cambridge: Harvard University Press, 1990)Google Scholar.

10 This does not mean that an epistemological naturalist would ignore or reject out of hand the traditional skeptical challenge to our knowledge of the external world. Quine appears to think that science can give us a kind of response to skepticism, for, as he pointed out, science can at least hope to explain why it is that our experience leads us to have largely correct beliefs. He says, “There is some encouragement in Darwin …. Creatures inveterately wrong in their inductions have a pathetic but praiseworthy tendency to die before reproducing their kind” (Quine, W.V. “Natural Kinds,” in Ontological Relativity and Other Essays, 126)Google Scholar. But an epistemological naturalist might not think that this point provides an adequate philosophical response to skepticism. A variety of views about skepticism are compatible with naturalized epistemology.

11 I am grateful to Bruce Hunter for help with this paragraph. Hilary Kornblith distinguishes two questions about our beliefs. The first is, “How ought we to arrive at our beliefs?” The second is, “How do we arrive at our beliefs?” He says that “the naturalistic approach to epistemology consists in [the thesis that] question 1 cannot be answered independently of question 2.” On my understanding, epistemological naturalism accepts the thesis of Kornblith's “naturalistic approach” but adds two additional doctrines. First is an explanation of why epistemological naturalism accepts this thesis. It does so because it is antiskeptical or at least non-skeptical. Second, naturalism does not merely hold that the psychology of belief formation is relevant to normative issues in epistemology. It holds that our normative epistemology is to be constrained by the psychology of belief formation. See Kornblith, Hilary “What is Naturalistic Epistemology?” in Naturalizing Epistemology, ed. Kornblith, Hilary (Cambridge: MIT Press, 1985), 1, 3Google Scholar.

12 Quine, “Epistemology Naturalized,” 75.Google Scholar

13 I shall leave open the question whether there is a priori knowledge, and if so, how it should be understood. Nothing in this paper turns on our having an answer to the question.

14 Quine, “Epistemology Naturalized,” 76.Google Scholar

15 The intuitionism proposed by Robert Audi does not postulate a special faculty. See Audi, “Intuitionism, Pluralism, and the Foundations of Ethics,” 121, 124Google Scholar. J.L. Mackie challenged any theory that postulates the existence of moral properties to provide a plausible epistemology of those properties. If a theory postulates the existence of a property of wrongness, for example, Mackie challenges it to provide a plausible account of how it is that we “discern” the wrongness of actions that we believe to be wrong, and of how it is that we “discern” the link between the actions’ feature of wrongness and the natural features, such as deliberate cruelty, that we believe the wrongness of the actions to be “consequential” to. He appears to think that no such account will be as comprehensible and as simple, and as plausible psychologically, as the idea that we are not perceiving a property of wrongness at all, but are rather simply responding subjectively and negatively to the natural features in question. See Mackie, J.L.Ethics: Inventing Right and Wrong (Harmondsworth: Penguin, 1977)Google Scholar, chap. 1, sec. 9.

16 Benacerraf argued that “the concept of mathematical truth … must fit into an over-all account of knowledge in a way that makes it intelligible how we have the mathematical knowledge that we have. An acceptable semantics for mathematics must fit an acceptable epistemology.” See Benacerraf, Paul “Mathematical Truth,” Journal of Philosophy 70 (1973): 667CrossRefGoogle Scholar. Benacerraf then went on to argue that the standard “platonistic” account of mathematics, according to which numbers are abstract objects, “makes it difficult to see how mathematical knowledge is possible,” given a familiar causal account of knowledge (Benacerraf, “Mathematical Truth,” 673, 671-3Google Scholar). On any acceptable epistemology, he suggests, there must be some “link between our cognitive faculties and the objects known.” For “We accept as knowledge only those beliefs which we can appropriately relate to our cognitive faculties.” To fill this gap, one might postulate the existence of a special cognitive faculty, which we could call “mathematical intuition.” Benacerraf notes that Kurt Godel postulated the existence of just such a faculty to account for mathematical knowledge, but he says, “the absence of a coherent account of how our mathematical intuition is connected with the truth of mathematical propositions renders the over-all account unsatisfactory” (Benacerraf, “Mathematical Truth,” 674–5)Google Scholar. Benacerraf does not explicitly say that a satisfactory epistemology of mathematics would have to mesh with an empirically plausible psychology of mathematical belief. But the idea is implicit in his response to Godel’ s intuitionism. He appears to rule out the soundness of an argument from metaphysics and epistemology to the existence of a psychological faculty. He therefore appears to accept tenets that would be characteristic of a naturalized mathematical epistemology. For similar views, Benacerraf cites Steiner, Mark “Platonism and the Causal Theory of Knowledge,” Journal of Philosophy 70 (1973): 57–66CrossRefGoogle Scholar.

17 Richmond Campbell objected, in personal correspondence, that epistemology and psychology are interdependent, since psychological methodology reflects certain assumptions about epistemology, and since epistemological theory depends on certain assumptions about psychology. But I don't see this independence as a decisive objection to scientific prioritism. It may still be true that epistemology must be constrained by what we take to be the settled results of empirical science.

18 Naturalized epistemology is not committed to ethical naturalism for it is not committed to the idea that moral knowledge is empirical. It is compatible with naturalized epistemology to hold that we do not have moral knowledge, or to hold that our moral knowledge is not empirical. Similarly, it is compatible with naturalized epistemology to deny that we have mathematical or theological knowledge, or to hold that such knowledge is not empirical.

19 Copp, Morality, Normativity, and Society. For a brief introduction to the view, see Copp, David “Does Moral Theory Need the Concept of Society?” Analyse & Kritik 19 (1997): 189–212CrossRefGoogle Scholar. For a reply to some objections, see Copp, David “Morality and Society - The True and the Nasty: Reply to Leist,” Analyse & Kritik 20 (1998): 30–45CrossRefGoogle Scholar.

20 A basic moral proposition is such that, for some moral property M, it entails that something instantiates M. An example is the proposition that capital punishment is wrong. Among non-basic moral propositions are propositions such as that nothing is morally wrong and that either abortion is wrong or 2 + 2 = 4. In Copp, Morality, Normativity, and Society, I called basic moral propositions “paradigmatic.“

21 In a fuller discussion of society-centered theory, I would qualify this claim. It is correct that a basic moral claim is true only if a corresponding standard is relevantly justified, but there are other conditions necessary for the truth of some moral claims. Some details are set out in Copp, Morality, Normativity, and Society at, for example, 24-6, 28–30Google Scholar.

22 Copp, Morality, Normativity, and Society, chaps. 6 through 11. Copp, “Does Moral Theory Need the Concept of Society?“

23 The issue is what human beings could know, given their nature. Perhaps a god could know a priori things that humans could only know through experience. If some humans can know certain things a priori, then these things are knowable a priori regardless of whether some other humans would need to rely on experience in order to know them.

24 The objection raises issues about the individuation of propositions that I cannot address here.

25 I take it that this objection is an episternic variation on the famous “open question argument.” See Moore, Principia Ethica, sec. 13.Google Scholar

26 See the discussion of moral issues, including abortion and cruelty to animals, in Copp, Morality, Normativity, and Society, 201-9, 213–6Google Scholar.

27 See Copp, Morality, Normativity, and Society, 237-40Google Scholar. See also Copp, David “Moral Knowledge in Society Centered Moral Theory,” in Moral Knowledge? New Readings in Moral Epistemology, ed. Sinnott-Armstrong, Walter and Timmons, Mark (New York: Oxford University Press, 1996), 243–66Google Scholar.

28 This was pointed out to me by Elijah Millgram. If one dollar bills were counterfeited very commonly then matters would be different. In that case, the one dollar bill would not serve very well its intended function in the economy.

29 To give a second example, it is clear that we can know various propositions about water without knowing, believing, or having any idea of, the chemical truth conditions of the propositions. And the grounds of our “water beliefs” need not include any grounds to believe corresponding propositions about H₂O that constitute the truth conditions of our water beliefs.

30 One of the first to propose and defend “reliabilism” was Goldman, Alvin I. “What is Justified Belief?”, Justification and Knowledge, ed. Pappas, George S. (Dordrecht: Reidel, 1979), 1–23Google Scholar.

31 Harman, Morality, 4-8Google Scholar.

32 Walter Sinnott-Armstrong pointed out, in correspondence, that evolutionary explanations of processes of belief formation can explain why these processes are sometimes not reliable, as in the case of explanations of visual and cognitive illusions.

33 The S-ideal code obviously would not consist solely of a master standard calling on people to act on the standard or standards, whatever they are, that would be part of the moral code whose currency in the society would best enable the society to meet its needs. H it consisted solely of a master standard, it would be rather more familiar, substantive, and concrete standard, such as, perhaps, a utilitarian standard.

34 See Rawls, JohnA Theory of Justice (Cambridge: Harvard University Press, 1971)Google Scholar. See also Daniels, Norman “Wide Reflective Equilibrium and Theory Acceptance in Ethics,” Journal of Philosophy 76 (1979): 256–82CrossRefGoogle Scholar.

35 Richmond Campbell urged me to discuss this objection.

36 This objection was suggested by comments made by Justin d’ Arms and Walter Sinnott-Armstrong, in personal communications.

37 Mullen, Peter “What's Wrong Can Never Be Right,” Manchester Guardian Weekly (December 4, 1983): 4Google Scholar.

38 Aberle, D.F.Cohen, A.K.Davis, A.K.Levy, M.J. Jr and Sutton, F.X. “The Functional Prerequisities of a Society,” Ethics 60 (1950): 100–11CrossRefGoogle Scholar. They discuss sexual abstinence in relation to the needs of society (103-4).

39 See the discussion of moral issues in Copp, Morality, Normativity, and Society, 201-9, 213–6Google Scholar. See also Copp, David “The Idea of a Legitimate State,” Philosophy and Public Affairs 28 (1999): 3–45CrossRefGoogle Scholar.

40 Mackie, Ethics: Inventing Right and Wrong.

41 Blackburn, Simon “How to Be an Ethical Antirealist,” Midwest Studies in Philosophy, vol. 12, Realism and Antirealism, ed. French, P.A.Uehling, T.E. and Wettstein, H.K. (Minneapolis: University of Minnesota Press, 1988), 361–75Google Scholar; Blackburn, SimonEssays in Quasi-Realism (New York: Oxford University Press, 1993)Google Scholar; Gibbard, AllanWise Choices, Apt Feelings (Cambridge: Harvard University Press, 1990)Google Scholar.

42 For an argument that society-centered theory supports a prohibition on cruelty to animals see Copp, Morality, Normativity, and Society, 204-7Google Scholar.

43 For a more thorough discussion of a similar proposed feedback mechanism, see Railton, “Moral Realism,” 192–7, 204-7Google Scholar.

44 For discussion of such cases, see Copp, “Moral Knowledge in Society Centered Moral Theory,” 262–4.Google Scholar

45 Robert Audi describes a “modified” form of ethical intuitionism in Audi, “Intuitionism, Pluralism, and the Foundations of Ethics.” Given the restrictions he imposes on the idea of an “intuition,” I do not know whether he would describe my view as intuitionistic.

46 Churchland, Paul “Neural Representation and the Social World,” Mind and Morals: Essays on Cognitive Science and Ethics, ed. May, LarryFriedman, Marilyn and Clark, Andy (Cambridge: MIT Press, 1996), 91–108Google Scholar. The quotations are from pp. 101, 102, 106. For particularism, see, for example, Dancy, JonathanMoral Reasons (Oxford: Blackwell, 1993)Google Scholar.

47 Goldman, Alvin I. “Ethics and Cognitive Science,” Ethics 103 (1993): 340–1CrossRefGoogle Scholar. Goldman cites Stitch, Stephen “Moral Philosophy and Mental Representation,” in The Origin of Values, ed. Hechter, MichaelNadel, Lynn and Michod, Richard E. (Hawthorne, NY: Aldine de Gruyter, 1993), 215–28Google Scholar; Medin, Douglas and Schaffer, M.M. ”A Context Theory of Classification Learning,” Psychological Review 85 (1978): 207–38Google Scholar; Estes, William ”Array Models for Category Learning,” Cognitive Psychology 18 (1986): 500–49CrossRefGoogle ScholarPubMed.