Abstract
Moral discourse allows for speakers to disagree in many ways: about right and wrong acts, about moral theory, about the rational and conative significance of moral failings. Yet speakers’ eccentricities do not prevent them from engaging in moral conversation or from having (genuine, not equivocal) moral disagreement. Thus differences between speakers are compatible with possession of moral concepts. This paper examines various kinds of moral disagreements and argues that they provide evidence against conceptual-role and informational atomist approaches to understanding our moral concepts. Conceptual role approaches fail because they cannot account for shared concepts among speakers with different commitments to the practical and conative ramifications of moral judgments. Informational atomist views fail because speakers need not be locked on to the same moral properties to share moral concepts.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Of course, there are limits here as well. Tying concept-possession too closely to particular evidential standards or other theoretical commitments will make it hard to account for conceptual continuity across theory change, so some degree of flexibility is necessary. Nonetheless, the point remains that moral interpretation seems to take a permissive attitude toward eccentricity.
See also (Merli 2008).
I thank an anonymous reviewer for emphasizing the importance of the difference between variation in beliefs about our concepts and in the role they play in speakers’ reasoning.
The notion of a plan is much like an intention, though somewhat broader, since plans can include things one cannot intend, e.g., I can plan but not intend that my friend be in New York tomorrow. Or, more formally,
Your “plan about what to do at t,” as I am understanding it, is just a proposition—roughly, a proposition that represents a way in which you might behave at t, and a way things might be if you did behave in that way. To ‘adopt’ the proposition p as your plan about what to do at t is to have a set of intentions about what to do at t such that, if the conjunction of the contents of those intentions is q, you believe ‘If it were the case that q, it would be the case that p.’ (Wedgwood 2007 p. 96)
There is a complication here regarding just what disposition is constitutive of concept-possession. For example Wedgwood writes that
The version of the claim that the intentional is normative that I have proposed here seems to imply that the dispositions that are essential to the capacity for a given sort of mental state are reliable indicators of the truth of certain normative propositions concerning that sort of mental state…For example, I suggested above that the disposition that is essential to the capacity for admiration is a disposition to respond to a set of antecedent mental states that makes it highly likely that it is correct to admire a certain object, by having an attitude of admiration for that object. Thus, this essential disposition is a reliable indicator of which sets of antecedent mental states make it likely that it is correct to admire various objects. (2007, p. 239)
It seems that the disposition in virtue of which one possesses a concept is one that involves moving from, say, antecedent mental states to the attitude of admiration and that those antecedent mental states reliably track what is actually admirable. And this reliability plays a role in his arguments for the epistemic status of various normative intuitions (see for example 2007, pp. 239-242).
If I understand Wedgwood correctly, this complication does not affect my conclusion, for the following reason. If these dispositions are constitutive of a thinker’s concept being what it is—that is, for example, the dispositions to move from certain admirability-tracking mental states to the attitude of admiration are what makes the thinker posses that attitude rather than some other attitude—Wedgwood’s account will struggle to account for shared moral concepts across disagreements, because in cases of widespread moral disagreement both parties cannot be linked in the necessary ways to the normative truth.
Wedgwood considers cases of disagreements in moral intuitions caused by “moral evil demons,” or situations where context or influence causes moral error that cannot be detected through, say, reflection on one’s beliefs (pp. 257-263). He argues that in spite of the possibility of these sorts of errors, we can rationally trust our own normative intuitions in light of disagreement with peers. But this argument for the epistemic force of our intuitions does not explain why our intuitions and those of our interlocutors are intuitions about the same thing, given that reliable ties between various mental states and normative truths seem to be constitutive of normative concepts generally and moral concepts in particular.
Here I simplify Wedgwood’s account in ways that do not affect my argument.
Wedgwood is neutral between several ways this sort of account might be developed. Note that the concept in question is really [morally right] since there are non-moral judgments (“the right way to fix the flat tire,” e.g.) that are orthogonal to moral judgments.
It’s also worth noting that in The Nature of Normativity Wedgwood devotes little attention to an account of the narrowly moral concepts, so whether or not he continues to endorse this sort of extension of his view is unclear.
For further discussion of this point see Wedgwood 2007, p. 24.
For a more detailed argument focusing on moral sentiments and their use in expressivist accounts in particular, see (Merli 2008). The general line of argument goes this way: specifications of the relevant conceptual role, or the relevant moral sentiments, will be either too broad or too narrow. A broad specification will fail to distinguish moral judgments from other sorts of evaluation, while a narrow specification will rule out speakers who, intuitively, seem to be making moral judgments.
Compare one of Fodor’s objections to generic Concept Pragmatism. CP must specify which c-involving inferences are constitutive of possessing the concept. Simple versions of holism claim that all of them are, which seems to rule out sharing concepts. Molecularism claims that only some inferences play this role, but this answer seems to require the sort of analytic/synthetic distinction Fodor finds implausible. Whether or not Fodor is right about this, there are, as we’ve seen, reasons to be particularly suspicious in the moral case, since the issues of contestability and interpersonal variation are more pronounced in moral discourse than elsewhere.
I thank an anonymous reviewer for a helpful revision to my formulation of the open question.
RM Hare argues against ‘descriptivism’ in ways that rely on the idea that differences in normative standard entail differences in concepts and meanings. See (Hare 1952).
See for example (Jackson and Pettit 1995).
This strategy has been developed in (Sayre-McCord 1997) and is suggested by some of the discussion in (Railton 1986). Sayre-McCord’s view is that we can interpret diverse speakers as sharing our terms and concepts when we can see their use of moral vocabulary (e.g., “right”) as regulated by the same moral kind (e.g., rightness).
Or, to put the point in Sayre-McCord’s language, it would be more difficult to argue that our use of moral terminology is regulated by a single moral kind, rather than different kinds. Jacksonian moral functionalism can be viewed as a kind of conceptual role view in which the relevant role is picked out by “mature folk morality” and moral properties are those which play the rightness role in that network of descriptions.
It’s also worth noting that convergence is something the naturalistic realist tends to like for other reasons as well. See for example Brink (1989) esp. pp. 197–210.
Note, though, that Brink himself is unsure of whether causal regulation is the right way to understand reference. See (2001, p. 170). He tentatively offers an alternative view that requires speakers to share a “concept” of morality—e.g., thinking about the moral point of view in terms of interpersonal justifiability—that is compatible with competing “conceptions” of how to understand substantive moral questions, e.g. Kantian or utilitarian conceptions. Then he suggests that sharing meanings is a matter of shared referential intentions to use moral terms to pick out properties that play an important role in “the interpersonal justification of people’s characters, their actions, and their institutions” (2001 p. 174). As Brink notes,
this appeal to the role of referential intentions within the theory of direct reference in order to defend a semantics for moral realism makes no use of the causal regulation thesis, which Boyd embraces…(2001 p. 175)
This is an interesting proposal that cannot be evaluated here. What is relevant for our purposes is that this view is no longer a locking view of the sort we’re discussing, since it doesn’t rely on a locking relationship between speakers and moral properties to ensure shared concepts.
One might wonder if our commitment to hearing others as picking out the same property would result in convergence under ideal circumstances simply because we would, under those conditions, take others’ views as relevant inputs to our reasoning. I suspect that the upshot is, instead, that a certain kind of idealized agreement simply wouldn’t take place. Yablo expands his point in this way:
The [interpretive] commitment comes at a price: a certain kind of anti-dogmatism. The readier we are to claim conceptual authority for our own moral views—to say they follow from what ‘right’ means in our mouths—the harder it becomes to hold onto the idea of coreference as between disputing parties. But then, far from its being a condition of moral communication that we expect to arrive at a single moral truth, the proper condition is that nothing will ever be regarded as the point of arrival: the point at which reference is finally fixed and moral theory acquires a conceptual imprimatur. (Yablo 2000, p. 19)
Even if this is wrong, the sort of convergence that would occur as a result of our commitment to share terms with other speakers is not the sort of convergence that would help ensure shared meanings. That way of convergence seems to be best described as a shift in the properties guiding our use of terms, rather than the revelation of what we were talking about all along. This point is developed in my second elaboration of the argument against the locking view, below. I thank an anonymous reviewer for bringing this issue to my attention.
This argument is developed in more detail in (Merli 2007).
It’s worth noting that in cases like this there’s little practical advantage to having a shared vocabulary that permits bedrock disagreement, as opposed to having a vocabulary that excludes some speakers but ensures agreement at the limit. Yablo’s remark about the importance of using evaluative language suggests that there’s some benefit to sharing terms, but, while this is generally true, it’s not particularly useful in the cases we’re considering, since, ex hypothesi, the language allows for the sort of disagreement that cannot be reasonably resolved.
For a more detailed development of this argument, see (Merli 2007). The suggestion is that we might converge on moral norms the way we converge on technological standards: some possibilities are ruled out on straightforward epistemic grounds, but these considerations aren’t robust enough to single out one alternative.
Gary Ebbs argues that our theoretical developments are often like this: our patterns of use underdetermine the extension of our predicates, and our eventual position is affected by historical contingencies, even if, after the fact, the course of our theory looks inevitable. See (Ebbs 2000).
Some arguments for its plausibility are discussed in more detail in (Merli 2007). I should also note that my argument here has some relationship with the “Moral Twin Earth” argument developed by Terence Horgan and Mark Timmons. (See, e.g., Horgan and Timmons 1993.) Horgan and Timmons argue, essentially, that we can have robust moral disagreement between speech communities whose ostensibly moral terms are regulated by different natural properties even at the end of the idealized day.
Though I find their argument unconvincing, I agree with the general opposition to causal regulation or locking views. Here, I have argued that locking views ask too much of speakers: they require a continuous relationship with a common set of moral properties, and it’s more plausible to see speakers as lacking that relationship.
Two tempting objections deserve replies. First, one might say that moral speakers are locked on to some generic property somehow held in common by competing normative theories. This won’t do because it fails to explain how interlocutors could be genuinely at odds with each other, if the apparently contradictory propositions expressed via their concepts were both true. Second, one might say that if the modest way of understanding convergence is right, the concept is empty. But this too has counterintuitive results for our interpretations of moral discourse. Perhaps we’ll be forced to accept them, but a more plausible hope is that a better view of moral concepts will allow us to do a better job saving the appearances.
Earlier versions of this work were presented at the Ohio State/Maribor/Rijeka Conference in Dubrovnik and at the University of Nebraska-Omaha. I am grateful to both audiences. I am particularly indebted to Paul Audi, Justin D’Arms, Pamela Hieronymi, Bill Melanson, and Geoff Sayre-McCord. I also thank Barbara Sykes for several helpful conversations about these issues. Finally, I owe a special debt to Tyler Hower and William Taschek for their help and encouragement.
References
Abou el-Fadl, K. (2001). And God knows the soldiers: The authoritative and authoritarian in Islamic discourses. Lanham, MD: Rowman and Littlefield.
Annas, J. (2004). Being virtuous and doing the right thing. Proceedings and Addresses of the American Philosophical Association, 78(2), 61–75.
Boyd, R. (1997). How to be a moral realist. In S. Darwall, A. Gibbard, & P. Railton (Eds.), Moral discourse and practice (pp.105–136). Oxford: Oxford University Press.
Brink, D. (1989). Moral Realism and the Foundations of Ethics. Cambridge: Cambridge University Press.
Brink, D. (2001). Realism, naturalism, and moral semantics. Social Philosophy and Policy, 18, 154–176.
Burge, T. (1979). Individualism and the mental. Midwest Studies in Philosophy, 4, 73–121.
Ebbs, G. (2000). The very idea of sameness of extension across time. American Philosophical Quarterly, 37(3), 245–268.
Fodor, J. (1998). Concepts: Where cognitive science went wrong. Oxford: Clarendon.
Fodor, J. (2004). Having concepts: a brief refutation of the twentieth century. Mind and Language, 19(1), 29–47.
Foot, P. (1967). Moral beliefs. In P. Foot (Ed.), Theories of ethics. Oxford: Oxford University Press.
Gibbard, A. (1990). Wise choices, apt feelings. Cambridge MA: Harvard University Press.
Hare, R. M. (1952). The language of morals. Oxford: Clarendon.
Horgan, T., & Timmons, M. (1993). Troubles on Moral Twin Earth: moral queerness revived. Synthese, 92(2), 221–260.
Jackson, F. (1998). From metaphysics to ethics. New York: Clarendon.
Jackson, F., & Pettit, P. (1995). Moral functionalism and moral motivation. Philosophical Quarterly, 45, 20–40.
Mackie, J. L. (1977). Ethics: Inventing right and wrong. Hamondsworth: Penguin.
Merli, D. (2007). Moral convergence and the univocity problem. American Philosophical Quarterly, 44(4), 297–315.
Merli, D. (2008). Expressivism and the limits of moral disagreement. Journal of Ethics, 12, 25–55.
Nichols, S. (2002). How psychopaths threaten moral rationalism: is it irrational to be immoral? Monist, 85(2), 285–303.
Peacocke, C. (1992). A study of concepts. Boston: MIT.
Peacocke, C. (1998). Implicit conceptions, understanding, and rationality. In E. Villaneuva (Ed.), Concepts: Philosophical Issues, vol 9 (pp. 43–48). Atascadero: Ridgeview.
Putnam, H. (1975). The meaning of ‘meaning,’ in Mind, Language and Reality. Cambridge: Cambridge University Press.
Railton, P. (1986). Moral realism. Philosophical Review, 95, 163–207.
Rey, G. (1983). Concepts and stereotypes. Cognition, 15, 237–262.
Rey, G. (1998). What implicit conceptions are unlikely to do. Concepts: Philosophical Issues, 9, 93–104.
Sayre-McCord, G. (1997). ‘Good’ on Twin Earth. Philosophical Issues, 8, 267–292.
Scanlon, T. M. (2003). Thickness and theory. Journal of Philosophy, 100(6), 275–287.
Smith, M. (1995). The moral problem. Cambridge MA: Blackwell.
Sturgeon, N. (1994). Moral disagreement and moral relativism. Social Philosophy and Policy, 11(1), 80–115.
Wedgwood, R. (2001). Conceptual role semantics for moral terms. Philosophical Review, 110(1), 1–30.
Wedgwood, R. (2007). The nature of normativity. Oxford: Oxford University Press.
Williams, B. (1993). Who needs ethical knowledge? Philosophy, 35(Supplement), 213–222.
Williams, B. (1995). Truth in ethics. Ratio, 8(3), 227–242.
Wilson, M. (1983). Predicate meets property. Philosophical Review, 91(4), 549–589.
Yablo, S. (2000). Red, bitter, best. Philosophical Books, 41(2000), 13–22.