Objectivity

2. Error-theory and fictionalism

An alternative to expressivist and subjectivist denials of the objectivity of morality is provided by so-called error-theories of morality. According to J. L. Mackie (1977), there are no objective moral properties instantiated in the world: such properties would be metaphysically ‘queer’ and epistemologically problematic (see Moral realism §6). So although claims such as ‘It is right to help those in distress’ are truth-apt and have a nonanthropocentric subject matter, they are systematically and uniformly false because they would require for their truth the instantiation of objective values such as rightness. Since claims like ‘The table is square’ do not require the instantiation of ‘queer’ properties in the world, they are not systematically and uniformly false: so the difference in objective status between moral discourse and discourse about the shapes of middle-sized material objective comes out via the idea that an error-theory is appropriate for the former but not the latter.

Crispin Wright has developed a forceful objection to the error-theory:

Wright notes that Mackie’s error-theorist does have a story about the point of moral discourse, about ‘some norm of appraisal besides truth, at which its statements can be seen as aimed, and which they can satisfy’ (2003: 185): the point of moral discourse is – ‘to simplify’ – to secure the benefits of social cooperation (Mackie 1977: ch. 5). The error-theorist who does not propose to eliminate moral discourse will identify some ‘subsidiary norm’ distinct from truth, will argue that this subsidiary norm governs the practice of making moral claims, and will argue that moral practice being thus governed has practical benefits. However, Wright argues that the error-theorist will not be able to combine this view of the utility of the practice of making moral claims with the central negative claim of the error-theory (that all atomic and positive moral claims are false):

Mackie’s error-theorist combines a factualist account of the semantics of moral claims (according to which they have truth-conditional content) with a cognitivist account of the psychological state expressed by moral judgments (according to which they express beliefs). This idea – that factualism and cognitivism belong together – is shared by many expressivists, who attempt to combine a nonfactualist, expressivist account of the semantics of moral claims (according to which they have no distinctive truth-conditional content) with a noncognitivist account of moral judgment (according to which they express noncognitive sentiments or feelings). However, this shared idea has in recent years been challenged, in particular by various forms of fictionalism about morality(see Moral fictionalism).

According to one such view, recently defended by Mark Kalderon (2005a), although claims such as ‘It is right to help those in distress’ have genuine truth-conditional content, we do not use them to express the belief that those contents are true. Thus a factualist account of the semantics of moral claims can be combined with a noncognitivist view of moral judgment: Kalderon defends a form of noncognitivist factualism about moral practice. (The term’ fictionalism’ is thought to be appropriate because of an analogy with fiction: ‘Bilbo Baggins lived in Hobbiton’ expresses a genuine proposition (that Bilbo Baggins lived in Hobbiton), but when I accept that claim I do not express belief in that proposition, but perhaps belief in some other proposition (such as the proposition that in The Hobbit, Bilbo Baggins lived in Hobbiton).

This idea, that moral claims as we actually make them are ‘useful fictions’ promises a form of denying the objectivity of morals that avoids the Frege-Geach problem (because it does not deny that moral claims have distinctive truth-conditional content) and also the worries that beset subjectivism (because it does not ascribe to moral claims an anthropocentric subject matter in the manner of subjectivism). It also avoids Wright’s worry about error-theories. Although the fictionalist claims – along with the error-theorist – that the property of moral rightness is never instantiated, and thereby implies that moral claims are false, since we do not in the making of moral claims actually aim at the truth there is no imputation of error to the participants in moral practice and therefore no worry about those participants being in ‘bad faith’.

Kalderon’s form of moral fictionalism is a form of what is known as hermeneutic fictionalism, which purports to be a description of our ordinary moral practice: the claim is that in our practice of making moral claims we do not actually aim at the truth but at the satisfaction of some other norm (Kalderon 2005b: 5–7). One problem faced by hermeneutic fictionalism is that in postulating a radical separation between the truth-conditional content of moral claims and what we actually use moral claims to do, it is enforcing an implausible gulf between meaning and use (for this and other objections to hermeneutic fictionalism, see Eklund 2007).

Whereas hermeneutic fictionalism is a descriptive thesis, revolutionary fictionalism is a prescriptive view that advocates a reform in our actual practice. A form of revolutionary moral fictionalism is developed by Richard Joyce (2001). According to Joyce, since we do actually aim at the truth in the making of moral claims, since moral claims have distinctive truth-conditional content and require the instantiation of moral properties for their truth, and since no such properties are instantiated, our actual moral practice is guilty of the error pointed out by Mackie’s error-theorist. However, Joyce continues, there are good practical grounds for thinking that we ought to reform our actual moral practice so that it aims at something other than truth. If we reform our moral practice so that it merely pretends to aim at the truth, so that it is a form of make-believe, the inculcation of such make-belief might still ‘function to bolster self-control against practical irrationality’ (Joyce 2005: 301) such as weakness of will, so that ‘the practical benefits of moral belief may still be gained by an attitude that falls short of belief’ (ibid.).

Revolutionary fictionalism, unlike hermeneutic fictionalism, still incorporates an error-theory, so it might be wondered whether it can avoid Wright’s objection to error-theories as outlined above. If the reform in moral practice envisaged by the revolutionary fictionalist actually works, why not simply construe moral truth in terms of the satisfaction of the norms governing what the fictionalist styles as make-believe? If no answer to this question is forthcoming, it appears that revolutionary fictionalism, if successful, actually undercuts itself.