Online research in philosophy

Cognitivism is the view that the primary function of moral judgements is to express beliefs that purport to say how things are; expressivism is the contrasting view that their primary function is to express some desire-like state of mind. I shall consider what I call the freshman objection to expressivism. It is pretty uncontroversial that this objection rests on simple misunderstandings. There are nevertheless interesting metaethical lessons to learn from the fact that the freshman objection is prevalent among undergraduates and non-philosophers. It leaves for expressivists two awkward explanatory tasks. Number one is that of explaining why natural selection – which, by expressivism's own lights, favoured moral thought and talk because of their socially useful regulative and coordinating functions – did not favour a stance that would make moral thought and talk more effective in fulfilling these functions. Number two is that of explaining how moral thought and talk survive in cultural evolution, despite the prevalence of the freshman objection and related worries. I conclude that expressivism as a theory of actual moral discourse rather than a revisionist theory is either false or committed to an implausible error theory, according to which ordinary speakers are systematically mistaken about what they are up to when they make moral judgements.

Aesthetic judgements are autonomous, as many other judgements are not: for the latter, but not the former, it is sometimes justifiable to change one's mind simply because several others share a different opinion. Why is this? One answer is that claims about beauty are not assertions at all, but expressions of aesthetic response. However, to cover more than just some of the explananda, this expressivism needs combining with some analogue of cognitive command, i.e. the idea that disagreements over beuaty can occur, and when they do it is a priori that one side has infringed the norms governing aesthetic discourse. This combination can be achieved by reading Kant’s aesthetic theory in expressivist terms. The resulting view is a form of quasi-realism about beauty. The position has its merits, but cannot ultimately explain the phenomena which motivate it. This conclusion generalises to quasi-realism about other matters.

The semantic theory of expressivism has been applied within metaethics to evaluative words like ‘good’ and ‘wrong’, within epistemology to words like ‘knows’, and within the philosophy of language, to words like ‘true’, to epistemic modals like ‘might’, ‘must’, and ‘probably’, and to indicative conditionals. For each topic, expressivism promises the advantage of giving us the resources to say what sentences involving these words mean by telling us what it is to believe these things, rather than by telling us what it would be for them to be true. This, in turn, absolves these theories of the burden of holding that there is any general answer to what it is for these sentences to be true. However, expressivism is famously subject to a deep and general problem about how to account for the meanings of complex sentences – a problem variously known as the ‘Frege-Geach’ or ‘embedding’ problem. In this paper I will be interested in whether there are reasons to think that the embedding problem looks less difficult for some of these applications for expressivism, than for others. In particular, in this paper I will be interested in the prospects for expressivism about what I will call epistemics – a class which I take to include epistemic modals like ‘might’ and ‘must’, sentential adverbs like ‘probably’, adjectives like ‘likely’ and ‘improbable’, and so-called ‘open’ indicative conditionals like ‘if the Fed doesn’t intervene, then the economy will enter a deflationary spiral’. There are several reasons to be particularly interested in expressivism about epistemics, relating both to the philosophical payoffs of such a view, and relating to the technical prospects for making it work. In other work I’ve touched on the especially interesting philosophical payoffs which make expressivism about epistemics interesting; in this paper I will be interested primarily in evaluating the possibility that there are better prospects for making expressivism about epistemics work than there are for making expressivism work about other topics. There are two main reasons why one might suspect that expressivism about epistemics will have better prospects than expressivism about many other topics, including in metaethics..

One important but infrequently discussed difficulty with expressivism is the attitude type individuation problem.1 Expressivist theories purport to provide a unified account of normative states. Judgments of moral goodness, beauty, humor, prudence, and the like, are all explicated in the same way: as expressions of attitudes, what Allan Gibbard calls “states of norm-acceptance”. However, expressivism also needs to explain the difference between these different sorts of attitude. It is possible to judge that a thing is both aesthetically good and morally bad. While the realist can explain the difference by suggesting that each judgment makes reference to a different property (or set of properties), the expressivist cannot. She must show that what is expressed by the speaker is different in each case. This has proven to be difficult to do.

Philosophers should consider a hybrid meta-ethical theory that includes elements of both moral expressivism and moral error theory. Proponents of such an expressivist-error theory hold that all moral utterances are either expressions of attitudes or expressions of false beliefs. Such a hybrid theory has two advantages over pure expressivism, because hybrid theorists can offer a more plausible account of the moral utterances that seem to be used to express beliefs, and hybrid theorists can provide a simpler solution to the Frege-Geach problem. The hybrid theory has three advantages over pure error theory, because hybrid theorists can offer a more plausible account of the moral utterances that seem to be used to express attitudes, hybrid theorists can more easily explain moral motivation, and hybrid theorists can avoid the implausible claim that all moral discourse is radically mistaken. Accordingly, such a hybrid theory should be more attractive than pure expressivism or pure error theory to philosophers who are skeptical about moral facts and truth.

Noncognitivism is the metaethical view according to which public moral judgements do not express beliefs, in spite of the fact that they are typically formed in the indicative mood. One form of noncognitivism — prescriptivism — holds that moral judgements are really commands.1 Another form, on which we will focus — expressivism — holds that moral judgements function to express desires, emotions, or pro-/con-attitudes (in Simon Blackburn’s words: ‘a stance, or conative state or pressure on choice and action’ [1993: 168]). Making a frequent appearance in the argumentative fray over noncognitivism is a thesis that is usually called ‘motivation internalism’.

Metaethical expressivism takes moral utterances to express non-cognitive attitudes in just the same way that ordinary factual utterances express belief.1 In doing so, it promises to solve three central metaethical problems at a stroke. First, it avoids the worry (most famously expressed by Mackie 1977) that moral facts and properties would be unacceptable additions to our ontology. Second, it promises to explain the seemingly close relation between one’s judging that one ought morally to ! and one’s having a motive to ! by understanding the attitude expressed as ‘conative’, i.e. as an intrinsically motivating desire-like state.2 Finally, and perhaps most importantly, expressivism promises a solution to the univocality problem, the problem of explaining how people with systematically different ethical views are nevertheless concerned with a common topic. Attempts to substantively explain intuitions of ethical agreement and disagreement between parties in terms of their employment of co-referring ethical concepts have faced a series of counterexamples. Expressivism understands univocality in terms of agreement and disagreement in non-cognitive attitude, thus circumventing the need to characterize a univocal cognitive content.3 Expressivism itself faces two central explanatory challenges. Most famously, the expressivist needs to solve the so-called ‘Frege-Geach problem’: to explain why moral beliefs and moral statements seem to have much the same sort of logical and inferential properties as ordinary factual beliefs and statements.4 Much less famously, the expressivist faces the challenge of specifying what sort of attitude is expressed by moral utterances. This task, dubbed the ‘moral attitude problem’ by Alexander Miller (2003, 43), has received surprisingly little attention compared with that lavished on the Frege- Geach problem. However, solving this problem is a central and non-trivial task for the expressivist program in metaethics: the truth of expressivism requires some answer to this question, and a recent paper by David Merli (2008) suggests that, besides its intrinsic....

This paper argues that expressivism faces serious difficulties giving an adequate account of univocal moral disagreements. Expressivist accounts of moral discourse understand moral judgments in terms of various noncognitive mental states, and they interpret moral disagreements as clashes between competing (and incompatible) attitudes. I argue that, for various reasons, expressivists must specify just what mental states are involved in moral judgment. If they do not, we lack a way of distinguishing moral judgments from other sorts of assessment and thus for identifying narrowly moral disagreements. If they do, we can construct cases of intuitively real dispute that do not rest on the theory’s preferred mental states. This strategy is possible because our intuitions about moral concept-ascription do not track speakers’ noncognitive states. I discuss various ways of developing this basic argument, then apply it to the work of the two most sophisticated proponents of expressivism, Allan Gibbard and Simon Blackburn. I argue that neither is successful in meeting the challenge.

Expressivist analyses of evaluative discourse characterize unembedded moral claims as functioning primarily to express noncognitive attitudes. The most thorny problem for this project has been explaining the logical relations between such evaluative judgements and other judgements expressed using evaluative terms in unasserted contexts, such as when moral judgements are embedded in conditionals. One strategy for solving the problem derives logical relations among moral judgements from relations of "consistency" and "inconsistency" which hold between the attitudes they express. This approach has been accused of conflating inconsistency with mere pragmatic incoherence. In reaction such criticisms several recent theorists have attempted to use alternative resources. The most sophisticated noncognitivists have often propounded theories with secondary descriptive components in addition to their primary expressive meanings. Recent independent suggestions by Frank Jackson and Stephen Barker attempt to solve the embedding problem by utilizing such descriptive components of moral utterances. Unfortunately, this strategy fails to handle a certain sort of example using just the descriptive resources available to noncognitivists. For it must rule valid arguments invalid in virtue of equivocation in the secondary descriptive meanings. The present paper explains the problem and suggests a moral for expressivist theories.

One’s account of the meaning of ethical sentences should fit – roughly, as part to whole – with one’s account of the meaning of sentences in general. When we ask, though, where one widely discussed account of the meaning of ethical sentences fits with more general accounts of meaning, the answer is frustratingly unclear. The account I have in mind is the sort of metaethical expressivism inspired by Ayer, Stevenson, and Hare, and defended and worked out in more detail recently by Blackburn, Gibbard, and others. So, my first aim (§1) in this paper is to pose this question about expressivism’s commitments in the theory of meaning and to characterize the answer I think is most natural, given the place expressivist accounts attempt to occupy metaethics. This involves appeal to an ideationalist account of meaning. Unfortunately for the expressivist, however, this answer generates a problem; it’s my second aim (§2) to articulate this problem. Then, my third aim (§3) is to argue that this problem doesn’t extend to the sort of account of the meaning of ethical claims that I favor, which is like expressivism in rejecting a representationalist order of semantic explanation but unlike expressivism in basing an alternative order of semantic explanation on inferential role rather than expressive function.