Objectivity

2. Anti-realist views of objectivity

Suppose we admit that two types of claim are truth-apt and yet do not have an anthropocentric subject matter. Do we have to view them as equally objective? Anti-realists, such as Wright (1986, 1992) and Dummett (1978), think not. They try to show how differences in objectivity can be captured, without adopting either non-cognitivism or subjectivism. How can we distinguish between ‘It is right to help those in distress’ and ‘The table is square’ if we allow that both are truth-apt, and that neither is susceptible to a subjectivist analysis? One way would be to develop an analogue of the distinction between primary and secondary qualities, the locus classicus for which is Locke (1689) (see Primary–secondary distinction; Secondary qualities). The standard example of a secondary quality is colour, while shape is usually held to be primary. What is involved in the claim that redness, for example, is a secondary quality? The idea is that there is a close relationship between facts about certain of our subjective states and statements like ‘The mailbox is red’. In order to explain this relationship, we need to explain the notion of a best judgment. A judgment that an object is red is best when it is made in conditions that are optimally good for appraising whether or not the mailbox is red. These conditions could be specified very roughly as those that obtain out of doors, in the shade, at lunch time on a lightly overcast summer afternoon. To say that redness is a secondary quality is to say that our best judgments concerning the redness of objects determine the extension of the concept red. The extension of a concept is the class of things to which that concept can correctly be applied. So our best judgments about whether objects are red determine which class of objects the concept red can properly be applied to. In contrast, a primary quality like squareness is one whose extension is determined independently of facts about best judgments concerning squareness. Best judgments in this case merely detect an independently determined extension.

This way of saying how ‘The mailbox is red’ is less objective than ‘The table is square’ is different from the subjectivist analysis. To say that redness is secondary is not to say that ‘The mailbox is red’ literally means that ‘If conditions were best, we would judge that the mailbox is red’. The claim is the weaker one, that our best judgments determine whether the mailbox falls in the extension of red. The proposal that redness is a secondary quality can therefore avoid the problems associated with subjectivism. And since it does not deny that ‘The mailbox is red’ is truth-apt, it avoids the problems associated with non-cognitivism. One way, then, to claim that ethical claims are less objective than claims like ‘The table is square’ is to argue that ethical qualities such as right, wrong, good and bad are secondary rather than primary.

According to the anti-realist, there can be more than one way of saying that one type of claim is less objective than another. Even if we cannot show that ethical qualities are secondary, there might be other ways of contrasting ethical claims with different claims. One way concerns what Wright (1986) calls ‘objectivity of truth’. This notion of objectivity features prominently in Dummett’s discussions of realism. To say that a class of statements exhibits objectivity of truth is to say that ‘[they] may be fully intelligible to us even though resolving their truth-values may defeat our cognitive powers (even when idealized)’ (Wright 1986: 5). Their truth might be ‘evidence-transcendent’: they are determinately either true or false, even if we are in principle incapable of citing evidence for or against them. Take Goldbach’s Conjecture (that every even number greater than two is the sum of two primes), or some claim about past happenings in some far distant galaxy. We understand these claims, but we are in principle incapable of resolving their truth-values. This gives us one way of claiming that mathematical or cosmological claims are more objective than claims about morals or comedy. There is little temptation to think that claims about the moral status of an action, or about the comic quality of a joke, may in principle transcend our best cognitive efforts. There are no difficulties akin to those affecting non-cognitivism and subjectivism: there is no denial of truth-aptitude, nor an assignment of an anthropocentric subject matter.