Published online by Cambridge University Press: 01 January 2020
There has recently been a revival of interest in ‘naturalizing’ ethics. A naturalization seeks to vindicate ethical realism — the idea that ethical judgments can be true reflections of a moral reality — without violating the naturalist constraint that science sets the limits of ontology. The recent revival has been prompted by examples of successful scientific reduction (e.g. temperature, water), and by the emergence of new, nonreductive naturalist strategies (e.g. for biological and mental properties). In this paper, I argue against such naturalist approaches to ethics. My argument builds on the traditional one offered by G.E. Moore, namely that a naturalization would fail to respect an existing difference between the meanings of moral and naturalistic terms. I defend this line of argument against the common claim that it cannot block ‘synthetic’ property identities, ones grounded not in meaning equivalences but in empirical discoveries (as in the cases of temperature and water). I then go on to show that the Moorean argument can make trouble even for recent revisionist and nonreductive naturalist approaches.
1 I would like to thank Steve Darwall, Ted Hinchman, Brooke Moore, Peter Railton, Stephen Sullivan, Mark Timmons, and Steve Yablo for helpful comments on earlier drafts. I am especially indebted to Allan Gibbard and Andy Hyrcyna, for originally inspiring me to think about the disanalogies between ethics and science which led me to develop the arguments of this paper.
2 I follow the literature in restricting ‘naturalization’ to cognitive, realist accounts. This leaves out expressivism and emotivism, though they are usually also grounded in a naturalist world view.
4 See Moore, G. E. Principia Ethica (Cambridge: Cambridge University Press 1903) 15ffGoogle Scholar.
5 Moore advanced different versions of the argument, for instance sometimes asking of the natural property (rather than the thing) whether it was good. The version on which I focus above reflects Moore's main line of thought.
6 Velleman, David ‘Brandt's Definition of “Good”,’ Philosophical Review 97 (1988) 359CrossRefGoogle Scholar. For a similar recent criticism of Brandt, see Gibbard, Allan Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press 1990) 18–22Google Scholar. Gibbard puts the point by saying Brandt's definitions lose the ‘recommending force’ of the original terms.
7 Copp, David ‘Moral Realism: Facts and Norms,’ Ethics 101 (1991), 623CrossRefGoogle Scholar. See also Sayre-McCord, Geoffrey ‘Moral Theory and Explanatory Impotence,’ in French, Peter Uehling, Theodore and Wettstein, Howard eds., Midwest Studies in Philosophy 12 (Minnesota: University of Minnesota Press 1988) 433-57Google Scholar.
9 Versions of this objection are pressed by Harman, Gilbert in The Nature of Morality (New York: Oxford University Press 1977)Google Scholar, and by Putnam, Hilary in Reason, Truth, and History (Cambridge: Cambridge University Press 1981), 207CrossRefGoogle Scholar. There are other long-standing objections to the Moorean argument; for a recent attempt to reply to some of them, see Ball, Stephen ‘Linguistic Intuitions and Varieties of Ethical Naturalism,’ Philosophy and Phenomenological Research 51 (1991) 1–38CrossRefGoogle Scholar.
10 Kripke, Saul Naming and Necessity (Cambridge, MA: Cambridge University Press 1980) 118–119Google Scholar
11 For a recent statement of naturalism, see Papineau, David Philosophical Naturalism (Cambridge, MA: Blackwell 1993)Google Scholar. See also Brink, David Moral Realism and the Foundations of Ethics (New York: Cambridge University Press 1989) 182-97CrossRefGoogle Scholar. Brink defends his version of ethical naturalism by appealing to the role of ethical properties in causal explanations.
12 Stephen Ball has also appealed to a difference between natural and ethical kinds in order to resurrect Moore's anti-naturalist argument (Ball, 8-15). The difference he discusses concerns whether it is possible to explain a divergence between ordinary and theoretical conceptions of the kinds. For natural kinds, the explanation is that we ordinarily lack microscopic vision, while ‘there is no obvious analogy to the “molecular theory” in ethics’ (Ball, 13). But it seems to me that a naturalist could easily offer an analogy. For most utilitarians distinguish between two levels of moral thinking: the commonsense level of rules and particular intuitions, and the theoretical level provided by utilitarianism. Add a reductive definition of utility, and the difference between the approaches of commonsense and theory is explained. To block such a proposal requires showing the reduction would have to be backed by a referential intention that is not to be found, as I am arguing here.
13 Sklar, Lawrence Philosophy and Spacetime Physics (Berkeley: University of California Press 1985) 310-15Google Scholar
15 See Gibbard, Wise Choices; Stevenson, C.L. Facts and Values (New Haven: Yale University Press 1963)Google Scholar; and Velleman, ‘Brandt's Definition of “Good”.’
16 Copp, ‘Moral Realism,’ and Sayre-McCord, ‘Moral Theory and Explanatory Impotence’
18 There are definitional moves to be made here, such as that the torturer must torture primarily in order to feel pleasure (but what if the pleasure is moral satisfaction?). My aim is simply to show one can grant causal features are relevant to moral judgment, without thereby endorsing a reduction.
20 See Brink, 132, and Boyd, Richard ‘How to Be a Moral Realist,’ in McCord, Geoffrey Sayre ed., Essays on Moral Realism (Ithaca, NY: Cornell University Press 1989) 181–228, esp. 201Google Scholar.
21 Boyd and Brink may only be taking the causal/explanatory role of consequentialist properties to be one piece of evidence in favor of consequentialism. Yet it is not clear why causal/explanatory role should count as evidence at all, given the lack of Causal Specification in our referential intentions. Perhaps the thought is that a causal/explanatory role is something we'd like all properties to have, even if preexisting referential intentions do not incorporate that goal. But to take causal/explanatory role as a piece of evidence, potentially contrary to other evidence, is to pit causal/explanatory criteria against ordinary ethical ones; and as I argued above, in this contest the causal/explanatory criteria lose out to the ordinary ones (as they do in the case of up-down). So the causal/explanatory role of consequentialist properties has very little if any evidential bearing, given the wealth of data from ordinary moral thinking. (I am grateful to an anonymous reviewer for pressing me on this point.)