Ecumenical expressivism: Finessing frege

Ethics

Volumn 116, Issue 2, 2006, Pages 302-336

  Ridge, Michael ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33644761122     PISSN: 00141704     EISSN: None     Source Type: Journal    
DOI: 10.1086/498462     Document Type: Review

References (79)

1. See Simon Blackburn, Essays in Quasi-Realism (Oxford: Oxford University Press, 1993).
2. In book 3 of An Essay concerning Human Understanding. See http://www.ilt.columbia.edu/publications/Projects/digitexts/locke/understanding/ chapter0301.html.
3. See H. P. Grice, "Utterer's Meaning, Sentence Meaning, and Word Meaning," Foundations of Language 4 (1968): 225-42. The literature on this topic is enormous. For a useful bibliography, see
4. Wayne Davis, Meaning, Expression and Thought (Cambridge: Cambridge University Press, 2003).
5. The Lockean framework is, of course, not uncontroversial, and on certain alternative approaches to semantics, meaning should be understood in terms of the expression of intensional entities (propositions, questions, imperatives) which are not themselves states of mind. One important challenge for the Lockean approach is to explain how a Lockean account can plausibly explain the distinction between semantics and pragmatics, a challenge I shall not explore here. For a contemporary Lockean take on this difficult challenge, see Wayne Davis, Non-descriptive Meaning and Reference (Oxford: Oxford University Press, 2005). Thanks to an editor at Ethics for helpfully pressing me on this point.
6. Yet others have been led by these considerations to reject the Humean philosophy of mind which insists that no mere belief can motivate. On these accounts, moral judgments are unitary states which are belief-like in that they are representational yet desire-like in that they can explain action without the presence of an independent desire. These states are sometimes referred to as "besires." See J. E.J. Altham, "The Legacy of Emotivism, " in Fact, Science and Morality: Essays on A. J. Ayer's Language, Truth and Logic, ed. Graham MacDonald and Crispin Wright (Oxford: Blackwell, 1986), 275-88. Besires are peculiar because intuitively any representation can exist without the motivation allegedly essential to that representation. An adequate appraisal of besire-based theories goes beyond the scope of this article. Here I simply take a broadly Humean philosophy of mind as given.
7. David Brink, "Moral Motivation," Ethics 108 (1997): 4-32, 9; emphasis added.
8. David Brink, Ethics Ibid., 5; emphasis added.
9. Frank Jackson and Philip Pettit, "A Problem for Expressivists, " Analysis 58 (1998): 239-51.
10. Some cognitivists also argue that this role is overstated; see, e.g., Sigrun Svavarsdottir, "Moral Cognitivism and Motivation," Philosophical Review 108 (1999): 161-219.
11. G. E. Moore, Principia Ethica (Cambridge: Cambridge University Press, 1903).
12. This distinction is problematic and has been glossed in a wide variety of ways; see Michael Ridge, "Moral Non-naturalism," Stanford Encyclopedia of Philosophy, available at http://plato.stanford.edu/entries/moral-non-naturalism/.
13. For a classic presentation of the Cornell-style view, see Richard Boyd, "How to Be a Moral Realist," in Essays in Moral Realism, ed. Geoff Sayre-McCord (Ithaca, NY: Cornell University Press, 1998), 181-228.
14. For some very useful discussions of how Moore's open question argument might be extended to deal with these interesting forms of moral realism, see Terence Horgan and Mark Timmons, "Troubles for New Wave Moral Semantics: The 'Open Question Argument' Revived," Philosophical Papers 21 (1992): 153-75,
15. and "Troubles on Moral Twin Earth: Moral Queerness Revived," Synthese 124 (1992): 224-60.
16. See also Nicholas Sturgeon, "Moore on Ethical Naturalism," Ethics 113 (2003): 528-56.
17. Paul Edwards and David Wiggins have defended such views. See Paul Edwards, The Logic of Moral Discourse (Glencoe, IL: Free Press, 1955). Thanks to an anonymous referee for bringing this to my attention.
18. One might argue that such hybrid views should be classified as forms of cognitivism since they vindicate the idea that there are objective moral facts. However, hybrid views (like cognitivism itself) leave open the possibility that all nontrivial moral claims are false. John Mackie's error theory is compatible with cognitivism and indeed with the hybrid view discussed in the text. Moreover, on a quasi-realist account, expressivism is compatible with the idea that there are "objective moral facts" but gives that phrase a deflationary reading. The debate between expressivists and cognitivists is not therefore well construed in terms of the question of whether there are objective moral facts. The debate concerns the question of whether moral utterances express moral beliefs which provide the truth-conditions for those utterances, and on that point the hybrid view is neither fully cognitivist nor fully expressivist. Thanks to an anonymous referee for pressing me on this point.
19. David Copp, "Realist-Expressivism: A Neglected Option for Moral Realism," Social Philosophy and Policy 18 (2001): 1-43.
20. James Dreier, "Internalism and Speaker Relativism," Ethics 101 (1990): 6-26. Thanks to Joshua Gert for useful discussion of the relevance of Dreier's work in this context.
21. See Daniel Boisvert, "Expressive Assertivism and the Embedding Objection," http://www.csub.edu/~dboisvert/Online_Papers/ EA_and_The_Embedding_Objection-Long_Version.pdf.
22. Jon Tresan, "De Dicto Internalist Cognitivism" (unpublished manuscript, University of Florida);
23. Matthew Millar, "Making Moral Judgements: Internalism and Moral Motivation" (PhD diss., University of Edinburgh, 2005).
24. Frank Jackson, From Metaphysics to Ethics (Oxford: Oxford University Press, 1998), 161.
25. Actually, some cognitivists may instead claim that moral judgments are irreducibly indexical. A crude version of subjectivism would hold that "I ought to X" just means "Xing would promote the satisfaction of my desires." On this sort of view, no particular content is essential to moral belief as such. My moral beliefs will be about me, while your moral beliefs will be about you. In David Kaplan's terminology, this species of cognitivism insists that moral beliefs as such must have a certain character rather than a certain content, where character is a function from a context of utterance to a content (see David Kaplan, "Demonstratives," in Themes from Kaplan, ed. Joseph Almog, John Perry, and Howard Wettstein [Oxford: Oxford University Press, 1989], 481-563). Your belief that you are tall and my belief that I am tall are identical in character but differ in content. Whether the cognitivist holds that moral beliefs as such must have a particular content or a particular character, the basic idea remains that we have a constraint on which beliefs can count as moral for a given moral agent which is independent of the agent's motivations. The content of the independently fixed candidate belief then determines which motivations are necessary for a given belief to count as moral. Ecumenical Expressivism rejects all such prior constraints on which beliefs can count as moral.
26. Or rather, someone who approves "in the right way"; expressivists must explain what is distinctive of moral proattitudes as such, but this difficult issue must also be put to one side here. For some discussion, see Gibbard's distinction between accepting a norm and being in the grip of one (Allan Gibbard, Wise Choices, Apt Feelings [Oxford: Oxford University Press, 1990], 60-61).
27. A. J. Ayer, Language, Truth, and Logic (New York: Dover, 1952), 107.
28. R. M. Hare, The Language of Morals (New York: Oxford University Press, 1952), 146.
29. Hare himself does not distinguish conversational implicature from other sorts of meaning, but this seems like the best interpretation of what he has in mind.
30. It might be objected that the connection between attitude and belief expressed in a single moral judgment is contingent according to Ecumenical Expressivism too, so that if contingency is a problem for Hare's account then it is a problem for my own account as well. The reply is that on my own account it is not contingent that a moral utterance always expresses a belief in which anaphoric reference is made to the property in virtue of which the speaker approves of actions. Moreover, it is this belief that does the work in avoiding the Frege-Geach problem. There is no analogously noncontingent belief involved in moral judgment on Hare's account. On Hare's account there will be cases in which no belief at all is expressed by a moral utterance, in which case the machinery of Ecumenical Expressivism will be unavailable. It is in this sense that the connections Hare draws between moral judgment and belief are too contingent to do the needed work. Thanks to an anonymous referee for discussion.
31. Perhaps the deflationist about truth-aptness should also require that the sentence be meaningful and not too vague to be truth-apt. This depends on whether being truth-apt is simply being the sort of thing that it makes sense to call true or false or whether being truth-apt also requires being determinately true or false. Deflationists may well have in mind the former. Lewis Carroll's "'Twas brillig, and the slithy toves / Did gyre and gimble in the wabe" is perhaps in this deflationist sense truth-apt yet has no determinate truth-value. Someone who heard a recitation of "Jabberwocky" but did not realize that it was a nonsense poem might reasonably wonder whether it was really true that the slithy toves gyred and gimbled in the wabe. Yet if someone heard a poem in which the question "Do the slithy toves gyre and gimble in the wabe?" was asked and this person then wondered whether that question was true then this would betray a kind of confusion-in more old-fashioned terms, a category mistake. Perhaps we should distinguish different conceptions of truth-aptness here, with some more deflationary than others. Thanks to an anonymous referee for pressing this point.
32. See Jackson and Pettit, "A Problem for Expressivists."
33. See Daniel Stoljar, "Emotivism and Truth Conditions," Philosophical Studies 70 (1993): 81-101;
34. and Huw Price, "Semantic Deflationism and the Frege Point," in Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives, ed. S. L. Tsohatzidis (London: Routledge, 1994), 132-55.
35. Here my discussion dovetails with Gibbard's, which itself picks up on some nice points developed by James Dreier. See Allan Gibbard, Thinking How to Live (Cambridge, MA: Harvard University Press, 2003), 61-68;
36. and James Dreier, "Expressivist Embeddings and Deflationist Truth," Philosophical Studies 83 (1996): 29-51.
37. See Walter Sinnott-Armstrong, "Expressivism and Embedding," Philosophy and Phenomenological Research 61 (2000): 677-93, for a more extensive argument that deflationism can solve at most some but not all of the problems in this area.
38. In the text, I suppose for expository reasons that the most basic moral predicate is "is a moral reason," with other moral predicates like "is morally good" and "is a moral duty" being analyzable in terms of moral reasons. The moves made in the text can just as easily be made in metaethical theories which reject this assumption.
39. Arguably, the proposed account relies on a broader notion of anaphora than the standard one invoked in the philosophy of language. For often the idea in those contexts seems to be that anaphora must refer back to something to which the speaker has referred previously. It is not clear on the account on offer, though, whether any such reference has occurred. For it is not clear that expressions of attitude in favor of actions with a given property literally refer to those properties. However, it seems clear enough that we can use anaphoric devices to refer back to properties introduced into conversation through expressions of attitude. For example, I can say "Hooray for the property of being red. That property is my favorite property." The second sentence makes anaphoric back reference to the property of being red, and any speaker of English would understand that this is what "That property" picks out. If the first sentence does not literally refer to the property of being red (and I do not mean to concede that it does not but just want to explore the implications of this view) then what follows is that we must broaden our conception of anaphora to fit better with ordinary language and not that ordinary language is mistaken because anaphora is limited to context in which reference has already taken place. Perhaps it is enough for anaphora that a property has been introduced into the conversation in some much more minimal sense that need not involve reference. Developing a full account of how this notion of "being introduced into the conversation" should be understood is an important task but not one I shall undertake here. Thanks to an anonymous referee for pressing me on this point.
40. Here I briefly abandon my stated neutrality about deflationism about truth to explore this version of expressivism; I am not committed to this version, though the argument of my article as a whole remains neutral on whether deflationism is correct.
41. Thanks to an anonymous referee for pressing me on this point.
42. Gibbard, Thinking How to Live, 94.
43. Gibbard, Thinking How to Live Ibid., 119.
44. Gibbard, Thinking How to Live Ibid.
45. Gibbard, Thinking How to Live Ibid., 130. Moreover, it is hard to see how any expressivist who does not follow maverick expressivists like Ayer in dismissing talk of reference to normative properties altogether can avoid a conclusion along these lines. For expressivists aspire to accommodate the intuitions which underlie Moore's open question argument. This makes it hard to see how any expressivist could allow that claims about the reference of normative predicates are not normative claims. For it would then turn out to be an ordinary descriptive fact that 'right' refers, e.g., to maximizing happiness, in which case Moore's concerns about open questions will become as salient for the expressivist as they are for any form of reductive cognitivism.
46. In some contexts, such anaphoric reference might be to the property to which one's interlocutor has speaker referred, although this will be unusual. For example, suppose you say "Wrongness just is being forbidden by the Ten Commandments. Working on the Sabbath is wrong." I, the village atheist, sardonically reply by saying that "working on the Sabbath has that property, all right, but it isn't wrong." In this sort of case it may be clear enough that my anaphora refers back to the property to which you have speaker referred rather than the property to which your use of the word 'wrong' refers. However, such cases are exceptional, and the more standard case will be one in which my use of 'that property' refers back to the word reference of your use of 'wrong'.
47. Nothing metaphysical hangs on this; if one does not like the term 'property' here then everything can be recast in terms of falling under descriptions to which the agent's preferences are sensitive.
48. An alternative would be to hold that such comparisons can be determinately true when one option dominates the worse option by a great deal on one dimension and is at most slightly worse on the remaining dimensions. One obvious problem with this suggestion is that it is irreducibly vague. A deeper problem is that even putting such vagueness to one side it seems to have counterintuitive implications in the present context for at least two reasons. First, suppose I hold a pluralistic theory with an extremely wide variety of types of reasons. Now suppose that A scores much better than B along one dimension but that B scores just slightly better than A on a whole host of the remaining dimensions. If the number of dimensions is great enough then we might intuitively want to say that there is more reason for B than A, but that contradicts the proposed semantics. Second, we may think that some sorts of reasons are simply more important so that doing better along some dimensions matters more than doing better along others. For example, the fact that my action scores very well indeed along the dimension of preventing mild pains (e.g., it prevents thousands of toe stubbings) may not make it better than an alternative course of action which scores only slightly better in terms of lives saved (the other action saves one life whereas my crusade against toe stubbings would save none). Intuitively, we might well want to say that there is more reason for the second action than the first, but the proposed semantics entails the opposite verdict. Thanks to an anonymous referee for pressing me on this point.
49. T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), 17;
50. and Jonathan Dancy, Ethics without Principles (Oxford: Oxford University Press, 2004), 29.
51. 2. However, in that case the argument given here will go through once again, and the nonnaturalist will be committed to incommen-surability after all. Or at least, she will be committed to incommensurability if she accepts the platitude that one thing is better than another just in case it instantiates more goodness. Thanks to Elinor Mason for useful discussion here.
52. David Hume, A Treatise of Human Nature, 2nd ed., ed. L. A. Selby Bigge, rev. P. H. Nidditch (Oxford: Clarendon, 1978);
53. Immanuel Kant, The Groundwork of the Metaphysics of Morals, trans. Lewis White Beck (New York: Macmillan, 1990);
54. and Adam Smith, The Theory of Moral Sentiments (Indianapolis: Liberty Fund, 1984).
55. Roderick Firth, "Ethical Absolutism and the Ideal Observer," Philosophy and Phenomenological Research 12 (1952): 317-45;
56. David Lewis, "Dispositional Theories of Value: II," Mark Johnston, "Dispositional Theories of Value: III," and Michael Smith, "Dispositional Theories of Value: I," Supplement to the Proceedings of the Aristotelian Society 63 (1989): 113-37, 139-74, and 89-111.
57. An alternative approach here would be to maintain that one reason is stronger than another just in case the relevant sort of idealized subject would approve of acting on the former when it conflicts with the latter. This approach faces two problems. First, consider a case in which there are several reasons for an action and only one reason against it. Suppose that the weaker reasons, when taken together, outweigh the stronger reason. An ideal subject presumably would approve of acting on the weaker reasons because of their aggregate weight. In that case, though, the proposed account seems to entail that each of those weaker reasons is actually stronger than the stronger reasons, which contradicts our initial description of the case. Second, there may be reasons which cannot coexist, in which case the proposed account provides no account of how one reason could be stronger than the other. To take an exotic example, suppose that I promise to meet you for lunch so long as there is no reason whatsoever not to do so. We might want to hold that my promise does generate a reason for me to meet you for lunch but only in the extremely rare situation in which there is no reason whatsoever against meeting you. How can we compare the reason generated by this promise with other reasons? The proposed account seems to imply that we imagine a counterfactual situation in which the two reasons conflict and let the approval of an ideal subject in that case decide the matter. By hypothesis, though, the reason generated by such an odd promise can never conflict with any other reasons. This, however, intuitively does not entail that the reason cannot be weighed against other reasons. Intuitively, the reason generated by my promise is much weaker than our reasons not to kill innocent people or cause gratuitous suffering, for example. Thanks to an anonymous referee for pressing me on this point.
58. One might have thought that we could capture all of these advantages within the framework of the original nondispositional version of Ecumenical Expressivism. The idea would be that dispositional theories can simply be understood as constituting another substantive normative perspective which someone could hold within the framework of our original version of Ecumenical Expressivism. However, this approach unfortunately would not really preserve pluralism about reasons. The crucial move, when it comes to preserving pluralism in this context, is to avoid being forced to the conclusion that there is really only one reason for action, namely, that such-and-such sort of subject would want one to perform the action. The dispositional version of Ecumenical Expressivism avoids this conclusion by holding that for a fact to be a reason for a given action is for that fact to be a fact to the effect that the action has some feature such that an idealized subject would approve of the action qua its having that feature. However, it is hard to see how a dispositional first-order normative view situated within the nondispositional version of Ecumenical Expressivism could avoid this conclusion. For recall that on the nondispositional version of Ecumenical Expressivism, for a fact to be a reason for action just was for that fact to be a fact to the effect that the action has "that property" where 'that property' anaphorically refers to the feature in virtue of which the speaker approves of actions. If, however, the speaker approves of actions just insofar as they would garner the approval of such-and-such a subject, as a speaker would if she accepted a dispositional first-order normative view, then it would follow that on her view there really is only one reason for action - namely, that an action would garner the approval of such-and-such a subject. It is hard to see how to preserve pluralism without building a dispositional element into the analysis of normative concepts itself.
59. Another interesting issue is what, on the dispositional version of Ecumenical Expressivism, we should say about someone who approves in the right way of more than one observer at the same time. The possibility of such complex normative psychologies is in one sense an advantage of the dispositional account. For it provides a sort of structure with which we could try to capture what Nagel memorably calls the "fragmentation of value" and Sidgwick's "dualism of practical reason." However, the possibility of such normative psychologies also raises a problem for the semantics. For now we can no longer comfortably hold that normative utterances express beliefs which make anaphoric reference back to the relevant sort of observer - that may simply be a failed definite description in the case in which a speaker approves of more than one observer in the right (and nonderivative) way. Fortunately, it seems to me that the dispositional account can be amended in a natural way to deal with this sort of phenomenon. For we can instead hold that normative utterances express (a) approval (of the right sort) of a set of observers (this set may or may not include more than one member) and (b) a belief which makes anaphoric reference back to what the set of all such observers would converge in approving of, disapproving of, insisting on, or whatever (depending on what the normative predicate being analyzed is). Such an analysis seems to fit well with how we could understand the all-things-considered judgments such a fragmented agent might nonetheless make and will also allow us to avoid the Frege-Geach puzzle in pretty much the same way developed in the text; we will get contradictions in belief in the right cases in just the same way, and the only difference will be in the contents of the beliefs which are contradictory. However, developing the details of the needed semantics would take us too far afield, and so for present purposes I must set these issues to one side. Many thanks to Mark van Roojen for useful discussion.
60. Actually, there is an important caveat to my claim in the text. For someone could suspend judgment in normative matters more globally, not taking on any substantive first-order normative views at all. I discuss this interesting possibility in the appendix.
61. Walter Sinnott-Armstrong raises a similar objection to Simon Blackburn's version of expressivism. See Walter Sinnott-Armstrong, "Nihilism and Scepticism about Moral Obligations," Utilitas 7 (1995): 217-36;
62. Nicholas Unwin, "Quasi-Realism, Negation, and the Frege-Geach Problem," Philosophical Quarterly 49 (1999): 337-52.
63. Actually, I do not need the strong thesis that someone must have some idea of what would be sufficient for something to be F in order to have some conception of F for purposes of the argument in the text. For in order to count as a nihilist someone needs only to judge that some necessary condition for there being a reason to perform an action could never be met. So in order to accommodate the possibility of nihilism I need only to suppose that having a conception of F involves at least some conception of necessary conditions on something's being F. While I think that the stronger thesis is plausible, I do not need to defend it for present purposes. Thanks to an anonymous referee for pressing me on this point.
64. The 'thereby' in this definition can be read in stronger or weaker ways. On a strong reading, both the premises and the denial of the conclusion must figure in the explanation of why the person has beliefs with inconsistent contents. This reading of the 'thereby' yields a logic closer to traditional relevance logics than classical logic, since it will not entail that a contradiction entails absolutely everything. On a weaker reading, it is enough that accepting the premises and denying the conclusion are sufficient for having beliefs with inconsistent contents. This reading of 'thereby' yields something much more like classical logic. Thanks to Peter Milne for bringing this to my attention.
65. It is worth pausing to clarify the notion of "accepts" in play here. One natural but misguided reading of the relevant notion is that someone accepts a sentence 'p' just in case they are poised for sincere utterance of 'p'. However, the discussion of conventional implicature reveals that this cannot be the right reading of 'accepts' in this context. For if someone says "even philosophers get this joke" and believes that philosophers get the joke but also does not think that it is surprising that philosophers would get a joke then this smacks of insincerity. Yet in the sense of 'accepts' in play here one need not accept the implicature to count as accepting that even philosophers get the joke. Simply believing that philosophers get the joke is enough. The idea of acceptance in play is one we can get via disquotation and asking whether someone believes that which is disquoted. Putting indexicals to one side, someone accepts a given sentence 'p' just in case they believe that p.
66. One further complication which I ignore in the text arises over contexts in which normative predicates figure in the contents of a propositional attitude attributed to someone (e.g., when I say "she believes that abortion is wrong"). The point is that in these contexts we are not typically assuming that the person to whom we attribute the propositional attitude associates the same cluster of descriptive properties with a given normative predicate that we do. For example, when I, as a utilitarian, say that Jones believes that abortion is wrong I need not be presuming that Jones believes that abortion fails to maximize happiness. So we should understand such attributions in terms of the attribution of a suitable belief-desire pair without taking a position on whether the speaker shares our conception of the good. So when I say that she believes that abortion is wrong I am making a purely descriptive claim, namely, that she has the belief that abortion is wrong. It turns out (though a given speaker may not realize this) that the belief that abortion is wrong is really in another sense a belief-desire pair - a general proattitude of the right kind and a belief (in a strict Humean sense of "belief") which makes suitable anaphoric reference back to the content of that proattitude. The only real difficulties emerging for Ecumenical Expressivism on this front arise when we combine ascriptions of normative beliefs with claims about the truth of what the subject believes, which should allow us to infer a normative conclusion. For example, we have inferences like "She thinks abortion is wrong, and everything she thinks is true, so abortion is wrong." However, I shall not here go into the details of how Ecumenical Expressivism is best extended to deal with these further cases. For this would require a full theory of truth (for a start) and would therefore take us too far afield from an outline of the basic ideas and advantages of the Ecumenical approach. I hope to return to this important cluster of issues elsewhere, though. Thanks to Timothy Williamson and John Hawthorne for pressing me on this point.
67. See Mark van Roojen, "Expressivism and Irrationality," Philosophical Review 105 (1996): 311-35, 320-21.
68. In his more recent work, Gibbard holds that atomic normative judgments like "believing that p is bad" express plans, but it seems reasonable enough to suppose that a plan not to believe that p amounts to a commitment to avoid believing that p.
69. An anonymous referee suggests that while Ecumenical Expressivism does not need a logic of attitudes to deal with the sorts of arguments canvassed in the text, it still will require such a separate logic for arguments like "I have most reason to keep my promise, I cannot keep my promise if I go fishing, so I do not have most reason to go fishing." If we read 'most reason' here as allowing ties, so that I can have most reason to keep my promise and most reason to do something else then the argument is not valid on any account. So the idea must be that we should read 'most reason' as precluding ties. In that case, though, the argument is valid without any recourse to a logic of attitudes. For someone who accepts the premises and denies the conclusion of this argument must, according to Ecumenical Expressivism, simultaneously believe (a) that keeping his promise is an action available to him which has more of a certain property than any of the other actions available to him, (b) that going fishing is an action available to him which is incompatible with keeping his promise, and (c) that going fishing has the property in question to a greater extent than any of the other alternatives available to him. These beliefs entail that keeping his promise simultaneously instantiates the property in question more and less than going fishing, which is surely contradictory. So the argument is valid according to the account laid out in the text.
70. For Blackburn's earliest account, see Simon Blackburn, Spreading the Word (Oxford: Oxford University Press, 1984). He later switches from talk of what one attitude "involves" to a theory employing higher-order attitudes in Essays in Quasi-Realism. For his most recent account,
71. see Simon Blackburn, Ruling Passions (Oxford: Oxford University Press, 1998).
72. See also Price, "Semantic Deflationism and the Frege Point," as well as Gibbard, Thinking How to Live. For criticisms of Blackburn's accounts,
73. see Robert Hale, "The Compleat Projectivist," Philosophical Quarterly 36 (1986): 65-84,
74. and "Can There Be a Logic of Attitudes?" in Reality, Representation and Projection, ed. J. Haldane and Crispin Wright (Oxford: Oxford University Press, 1993), 337-63;
75. G. F. Schueler, "Modus Ponens and Moral Realism," Ethics (1988): 492-500;
76. van Roojen, "Expressivism and Irrationality";
77. and Nicholas Zangwill, "Moral Modus Ponens," Ratio 5 (1992): 177-93. Blackburn responds to many of these worries in developing and revising his account in all of the works by him cited above.
78. Gibbard, Thinking How to Live, 56.
79. Cian Dorr, "Expressivism and Wishful Thinking," Nous 33 (2002): 558-72.