Abstract
Mooreans claim that intrinsic goodness is a conceptual primitive. Fitting-attitude theorists object: they say that goodness should be defined in terms of what it is fitting for us to value. The Moorean view is often considered a relic; the fitting-attitude view is increasingly popular. I think this unfortunate. Though the fitting-attitude analysis is powerful, the Moorean view is still attractive. I dedicate myself to the influential arguments marshaled against Moore’s program, including those advanced by Scanlon, Stratton-Lake and Hooker, and Jacobson; I argue that they do not succeed.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
See his (1986: 53).
Some claim that Brentano’s concept of correctness is identical with the concept of moral rightness (indeed he uses the German ‘richtig’ to express it). In this case, Brentano’s view would hardly be mysterious; he proposes only to understand the good in terms of the right. However I worry that this interpretation is ultimately untenable. First, note that Brentano’s account of moral rightness is defined in terms of correctness [see his (2009: 13, 32–34), Olson (2017)]. But if correctness just is rightness, this account is, at best, uninformative. Second, Brentano defines truth in terms of correctness: what is true is what is correct to affirm (2009: 18). But what is true is not merely what it is morally right for us to believe. So I think correctness is not the familiar concept of moral rightness (although it may be identical with some more opaque, all-things-considered rightness).
Scanlon also appeals to certain types of ‘pro-actions,’ such as maintenance and preservation. For ease of exposition, I will not mention pro-actions, but this omission is not intended to be philosophically significant.
For the sake of variety, I will sometimes speak of ‘favor’ or ‘care’. However, as I use these terms, to care about a thing, or favor it, is simply to hold some pro-attitude towards it.
See his (2011: 52).
This restriction may also enable us to avoid the wrong kind of reason problem and its ilk. The argument usually appeals to an example; we imagine that if we do not favor a saucer of mud, a demon will punish us. Thus we have reason to favor the mud, but it is not good. [For further discussion see Jacobson (2011) and Rønnow-Rasmussen (2011: 33–42)]. No such objection applies to our account; the fact that a demon will punish us is, of course, extrinsic to the mud.
Emphasis mine. Though I will make use of his examples in what follows, I am admittedly hesitant about the cases Scanlon gives; it is, I think, unclear whether resorts or research projects can have intrinsic value. Further, while Scanlon commits himself to the view that reasons are provided only by natural properties, this is not a necessary consequence of this ‘reasons first’ position; we could instead claim e.g. that we have a reason to perform some act because it would be kind to do so. However, as I understand Scanlon and his allies, they hope to reduce morality to the concept of reasons. This reduction would be made more complicated if our reasons are themselves generated by some of the moral facts they wish to analyze in terms of reasons. Further, as long as the moral supervenes upon the non-moral, I suspect that there will always be some natural property that can do the same work. (For example, suppose we say that we have reason to perform some action because it would be kind to do so, and that it would be kind to do so because it would make some person happy. Then Scanlon could, I believe, simply appeal to the fact that this act would make someone happy as a reason to do it.).
I borrow this description of the problem from Schroeder (2007).
We may think this view has unacceptable consequences; it entails that we have intrinsic reason to value only what is good. But this does not imply that an individual might not have sufficient reason to value something bad (such as e.g. their taking pleasure in the pains of others). As I understand things, the Moorean view is presumably the opposite of the reasons fundamentalist. If this is so, then the Moorean view is compatible with e.g. the claim that each person has sufficient reason to value their own pleasure, even if it is taken in the pain of others. What cannot be true is that we all have sufficient reason to value what is bad. And this claim is, I think, quite plausible.
The Moorean may claim this argument invalid because of equivocation. We say that (1) is a fact about metaphysical explanation: it specifies the good-making features of S—the properties that ground and explain its value. But (2) is a fact about reasons; it says that the fact that S is intrinsically good provides a reason to value it. Thus the argument is, in fact:
(1*) The fact that S is pleasant grounds the fact that S is intrinsically good.
(2*) The fact that S is intrinsically good gives us reason to value S.
(3*) So, the fact that S is pleasant gives us reason to value S.
And this inference, we say, is invalid. However I worry that the argument might be repaired relatively easily. If some fact F gives us a reason, then F grounds the fact that we have this reason. So, e.g. if the fact that the resort is pleasant gives us reason to visit, then the fact that the resort is pleasant also grounds the fact that we have a reason to visit. Thus (2*) entails:
(2**): The fact that S is intrinsically good grounds the fact that we have reason to value S.
But the grounding relation is transitive. Thus (2**) and (1*) entail that:
(3**) The fact that S is pleasant grounds the fact that we have reason to value S.
However, if some fact grounds our reason to have some attitude, then this fact gives us reason to have that attitude. E.g. if the fact that the resort is pleasant grounds the further fact that we have reason to visit, then I believe the pleasantness of the resort gives us a reason to visit. So I think (3**) entails (3*). However I would be happy to be mistaken about this—if I am, then it would only be easier to defend the Moorean position.
Schroeder (2007) proposes an interesting and novel account that may explain why reasons are not additive. However his approach has costs; it requires that it is not particular reasons but sets of reasons that have strengths or weights. I find this difficult to accept. But those untroubled by this consequence will have another attractive way of solving this puzzle. See Stratton-Lake (2017) for discussion.
See his (2017).
I am thankful to Bradford Skow for suggesting this kind of defense.
This example is due to Yablo (1992).
The test of proportionality is in fact much more complex—and there is debate about how it should be understood; see Woodward (2010) for discussion. However I believe this very crude form of the test will be clear enough for our purposes.
Of course, if we think the transitivity argument is unsound instead, then this would be just fine for the Moorean; it would make any defense unnecessary.
Or as we do when we look at some organic unity and see that the value of the whole is not equal to the sum of the values of its parts.
In fact, I think this argument is somewhat poorly stated. As Tucker (2016) argues, value pluralism cannot be understood as the view that there are many different kinds of intrinsic goods. But for the sake of argument, I will not object here.
There is another argument about pluralism lurking—but this concerns not value pluralism but pluralism about reasons. It seems on the Moorean view we are committed to a kind of monism about intrinsic reasons; we must claim that we have intrinsic reason to value things only because they are valuable. But it seems that e.g. our reasons to favor a resort are very different from our reasons to favor some line of research. (I am thankful to Peter Graham for making this objection clear to me.) However, this concern can, I believe, be ameliorated in three ways. First, it is not clear that either of these things are in fact examples of intrinsic goods, and our reasons to value clear examples of intrinsic goods (like different instances of pleasure) appear more homogenous. Second, the Moorean view is committed only to monism about non-derivative reasons. Thus our derivative reasons to favor a resort may be very different from our reasons to value some line of research. Third, though fundamentally we have reason to value things only because they are good, things may be good for very different reasons. Thus there is still a kind of pluralism here—it is simply a little bit deeper in our explanatory chain.
The fundamental concept of moral supervenience is simple enough: we claim that there cannot be a difference in the moral facts without a difference in the non-moral (i.e. natural) facts. This slogan is perhaps best understood as a claim of strong supervenience, i.e.:
Necessarily, if a thing x has some moral property F, then there is some non-moral property G (which may be disjunctive, conjunctive, negative, etc.) such that, necessarily, anything that has G has F.
For an exhaustive survey of the various interpretations of moral supervenience, see McPherson (2015).
In fact, it is unclear just what Blackburn’s ultimate concern was—and there is still significant debate. But I am interested only in the supervenience argument as Stratton-Lake and Hooker present it; I believe this is the only version of the supervenience argument that has been advanced in the debate between the Moorean and the fitting-attitude theorist. Thus if I can undermine their concern, then I will have defended the Moorean view adequately for my purposes. But I do not pretend that this is the only version of the supervenience argument, nor do I insist that Hooker and Stratton-Lake’s argument is related to Blackburn’s concern in the way they claim.
In fact, my strategy echoes Scanlon (2014). Though Scanlon identifies only certain facts about reasons as being necessary, we both agree that (1) we should divide moral facts into a pure or privileged class and an impure class and that (2) the pure claims may, in some sense, explain the supervenience of the impure moral facts.
See his (2005).
Of course, I do not expect the naturalist or anti-realist to be moved by such concerns. But this should be enough, I think, for the non-naturalist—regardless of whether he subscribes to the Moorean view or Scanlon’s alternative. I should note also that the claim that moral principles are a priori is not necessary for the suggested defense; it is merely, I think, a reasonable posit for the non-naturalist, regardless of the structure they assign to moral philosophy.
In fact, there is a kind of gap in this argument, for there may be moral claims that are not principles but are not contingent (for example, the disjunction of the true theory of right action and the claim that some particular action is wrong). But if such claims hold necessarily then, again, it is not clear that there is anything to explain, since their supervenience on the natural facts is also vacuous. Thus the gap is, I hope, easily bridged.
We might object: what if the facts that the true theory of right action mentions are moral facts, namely about goodness or virtue? Then we have an explanation only of the moral in terms of the moral. However, this dependence cannot continue indefinitely. The right may be explained in terms of the virtues or the good, but unless our total moral theory is to be circular, eventually a non-moral property must be invoked.
See Scanlon (2014: 2).
We may think the explanation of supervenience given by Hooker and Stratton-Lake simpler than the alternative I suggest. But I think they are on roughly equal ground. This becomes evident when we think not of non-naturalism in general, but instead adopt some particular view of the structure of moral philosophy. E.g. suppose we accept the Moorean position; we ground all moral notions in the concept of intrinsic goodness. Then to explain moral supervenience we need only appeal to a theory about what things are intrinsically good and why. Such a theory will both explain why supervenience holds and (if it is necessarily true) guarantee that it does. To explain the supervenience of the moral upon the non-moral, Hooker and Stratton-Lake need the same thing—namely a reduction of all other moral notions to the concept of reason and a necessarily true principle that specifies our reasons.
References
Blackburn, S. (1971). Moral realism. In J. Casey (Ed.), Morality and moral reasoning. London: Methuen.
Brandt, R. (1946). Moral valuation. Ethics,56, 106–121.
Brentano, F. (2009). The origin of our knowledge of right and wrong (R. Chisholm, and E. Schneewind, Trans.). Routledge: New York.
Broad, C. D. (1930). Five types of ethical theory. London: Routledge & Kegan Paul.
Chisholm, R. (1986). Brentano and intrinsic value. New York: Cambridge University Press.
Ewing, A. C. (1948). The definition of good. Oxford: Routledge & Kegan Paul.
Hooker, B., & Stratton-Lake, P. (2006). Scanlon versus Moore on goodness. In T. Horgan & M. Timmons (Eds.), Metaethics after Moore (pp. 149–168). New York: Oxford University Press.
Jacobson, D. (2011). Fitting attitude theories of value. In E. N. Zalta, (Ed.), The stanford encyclopedia of philosophy (Spring 2011 Edition). Retrieved November 2017 http://plato.stanford.edu/archives/spr2011/entries/fitting-attitude-theories/.
McPherson, T. (2015). Supervenience in ethics. In E. N. Zalta, (Ed.), The stanford encyclopedia of philosophy (Winter 2015 Edition). Retrieved November 2017 http://plato.stanford.edu/entries/supervenience-ethics/.
Moore, G. E. (1993). Principia ethica (revised edition, T. Baldwin, ed.). New York: Cambridge University Press.
Olson, J. (2017). Brentano’s metaethics. In U. Kriegel (Ed.), Routledge Handbook of Brentano and the Brentano School (pp. 187–195). New York: Routledge.
Parfit, D. (2011). On what matters. New York: Oxford University Press.
Parfit, D. (2017). Responses. In S. Kirchin (Ed.), Reading Parfit on what matters (pp. 189–236). New York: Routledge.
Rønnow-Rasmussen, T. (2011). Personal value. New York: Oxford University Press.
Scanlon, T. M. (1998). What we owe to each other. Cambridge, MA: Harvard University Press.
Scanlon, T. M. (2014). Being realistic about reasons. New York: Oxford University Press.
Schroeder, M. (2007). Weighting for a plausible humean theory of reasons. Nous,41(1), 110–132.
Shafer-Landau, R. (2005). Ethics as philosophy: A defense of ethical nonnaturalism. In T. Horgan & M. Timmons (Eds.), Metaethics after Moore (pp. 209–222). New York: Oxford University Press.
Stratton-Lake, P. (2017). The buck-passing account of value: Assessing the negative thesis. In S. Kirchin (Ed.), Reading Parfit on what matters (pp. 82–95). New York: Routledge.
Tucker, M. (2016). Two kinds of value pluralism. Utilitas,28(3), 333–346.
Woodward, J. (2010). Causation in biology: Stability, specificity, and the choice of levels of explanation. Biology and Philosophy,25, 287–318.
Yablo, S. (1992). Mental causation. Philosophical Review,101(2), 245–280.
Zimmerman, M. (2001). The nature of intrinsic value. Lanham, MD: Rowman & Littlefield Publishers.
Acknowledgements
I want to thank Phillip Bricker and Peter Graham for their careful and thoughtful comments. I am especially grateful to Bradford Skow, Fred Feldman, Lisa Tucker, and an anonymous referee from Philosophical Studies for all of their help with this essay.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Alternatives to Scanlon’s view
Appendix: Alternatives to Scanlon’s view
There are many fitting-attitude views. I have attended to Scanlon’s: it is the most popular, and, I think, the most attractive. But some may prefer views like Brentano’s, which appeal to correctness, or views like Ewing’s, which appeal to fittingness.
Begin with the former. According to Brentano:
x is intrinsically good = df. it is correct to intrinsically love x.
The double counting argument may thus be recast as follows: according to Brentano, the natural properties of things make them correct to love; this makes such things good. According to the Moorean, the natural properties of things make them good and this makes them correct to love.
But again, this short description belies the options available to the Moorean: in fact, he may accept that both the natural and evaluative properties of things make them correct to love. He may do so by claiming that the goodness of things makes them non-derivatively correct to love, and that the natural properties of things make them derivatively correct to love. Thus our verdict of this argument should, I think, mirror our verdict of Scanlon’s argument. (This response runs similarly for the accounts advanced by philosophers like Ewing: we simply replace “correct to love” with “fitting to favor.”)
Our reply to the concern about pluralism is also familiar: Brentano can claim that plural goods are alike in that they are all correct to love. It is unclear that this is a true explanation, however. If it is an explanation, then we must admit that the Moorean cannot give such an answer—but he can explain what plural goods have in common that makes them correct to love. The result is, at worst, a draw. (And mutatis mutandis for other fitting-attitude views.)
Finally, supervenience. It is perhaps analytic that, if we have a reason to do something, then there is something that provides that reason; reasons are, in this way, relational. But fittingness and correctness are not relational. Thus, the claim that,
(i) If something is correct to intrinsically love, then there must be something that makes this so
is no stronger than the claim that
(ii) If something is intrinsically good, there must be something that makes this so.
I conclude the supervenience argument cannot be advanced if we do not accept Scanlon’s program. It is, in this way, different from the other concerns examined.
Rights and permissions
About this article
Cite this article
Tucker, M. Moore, Brentano, and Scanlon: a defense of indefinability. Philos Stud 177, 2261–2276 (2020). https://doi.org/10.1007/s11098-019-01309-w
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-019-01309-w