Realism and emotivism

The moral content question is, of course, a question in philosophi­cal semantics. As we saw in Chapter 3, one plausible way to say what a sentence means is to say what the world would have to be like for it to be true—that is, to give its truth conditions.

K: Killing innocent people is wrong (or I ought not go about killing innocent people)

the words “killing innocent people” have the same sense and refer­ence as they do in the factual sentence

Killing innocent people is common (or I have seen someone killing innocent people).

The new questions, therefore, are about the meaning of “I ought not to” and “is wrong.” Let's try to see what a truth-conditional seman­tics for “is wrong” might look like.

Our explanation of what a predicate such as “is red” meant involved saying what it referred to. We said that its reference was its extension, which was a class of objects. Since K is equivalent to

K': Every action that is a killing of an innocent person is

wrong

the class of things in the extension of “is wrong” is a class of actions. So far, so good.

But we then went on to give the sense of “is red” by saying that it was a way of determining that reference. How are we to determine which acts are in the extension of “is right”?

Anyone who believes that the way this extension, in particular, and the truth values of moral claims, in general, are determined is not importantly different from the way the truth values of factual claims are determined, we call a “moral realist.” Moral realists think that, just as there are ordinary facts “out there” in the real world that deter­mine whether factual claims are true or false, so there are moral facts in the world that determine the truth values of moral claims.

We certainly do act on the basis of factual beliefs: in Chapter 1, I suggested a functionalist theory of beliefs that explained why that was. But when we act on the basis of a factual belief we do so because we already have preferences or desires that make the belief relevant to deciding what to do. If I want to eat a strawberry, then I need to find out where there are strawberries, which is a matter of fact, before I can set about the action of eating them. But believing that there are strawberries in the kitchen doesn't commit me to going there to eat them. What does commit me to that action is the combination of the belief that there are strawberries in the kitchen and the desire to get strawberries to eat.

If, however, for some bizarre reason, I decided that I ought, all things considered, to eat the strawberries in the kitchen, then I would be committed to doing so whether I wanted to or not. Whereas factual beliefs commit us to action only in conjunction with our preferences or desires, moral beliefs commit us to action what­ever our preferences or desires. The terminology I shall use to mark this difference is that moral beliefs are action-guiding, while fac­tual beliefs are not. Always remember, however, that beliefs guide action too, but in a different way. The moral realist's view—that there's no difference between the ways the truth values of factual and of moral beliefs are determined—has to explain why there is, nevertheless, this important difference between them.

Immanuel Kant, the great German philosopher of the Enlightenment, was one of the first people to identify this sort of action-guiding “ought.” He called it a “categorical imperative” and contrasted it with what he called “hypothetical imperative,” such as the “ought” in the sentence

If you want to get there quickly, you ought not to walk but to take a taxi.

This “ought” is hypothetical because it depends on a hypothesis about what you want. Even if someone just said:

You should not walk.

the “ought” would still be hypothetical because it would still be based on this hypothesis about your wants. So you cannot identify a hypo­thetical imperative simply by seeing whether it is preceded by “If you want to... ” Instead you must consider whether the speaker would withdraw the “ought” sentence if you said that you didn't have the desire he or she seemed to be supposing you to have. If someone would still say you ought to do something whatever you said your wants and desires were, then the “ought” would be categorical.

We can express one challenge for moral realism simply by asking how it is to explain the categorical nature of moral imperatives. The force of this challenge becomes clearer if we recall the way in which we connected the idea of a truth condition with the idea of commu­nication at the end of Chapter 3. Because of the connection between the truth conditions of sentences and the contents of beliefs, we were able to say that we use the speech act of assertion to communicate our beliefs. Thus, we said that someone who understands “It is raining” uses it to get other people to believe that the truth conditions of the sentence hold.

In the normal case of the speech act of assertion, I get you to believe that it is raining because you think that I believe it and that I am in a position to know. That is why we call Mary's asserting that it is raining the expression of her belief that it is raining, for she gets us to believe it by giving us reason to think that she does. The moral realist, then, regards the assertion of K as a way of expressing the belief that killing innocent people is wrong. The problem is that if it is an ordinary belief that is being expressed, it is hard to see how it can also be action-guiding: beliefs, as I said, guide action only in concert with desires.

So the fact that moral assertion commits us directly to action might lead you to suppose that moral sentences do not express beliefs but feelings, preferences, or desires.

I shall call the view that moral sentences express not beliefs but feelings, preferences, or desires, “emotivism.” Strictly, as the term suggests, emotivism would be the view that moral sentences express feelings or emotions. But the view that moral sentences express action-guiding states of mind rather than beliefs is the core of emo- tivism even in this stricter sense. I shall call action-guiding mental states that dispose you towards doing something “pro-attitudes.” Those that dispose you against some action, I shall call “con- attitudes.” Pro-attitudes and con-attitudes together I shall call just “attitudes.”

Moral realism and emotivism represent the extreme poles of views on the moral content question, and these views tend to pro­duce polar positions in moral epistemology. The moral realist will say that since moral sentences express beliefs that can be true or false, and since they can be justified or unjustified, moral beliefs are candidates for knowledge. The emotivist, on the other hand, will say that, since moral sentences express attitudes, which cannot be true or false, they are not candidates. So in moral epistemology, realism and emotivism, as views about the moral content issue, tend respec­tively to go with cognitivism—the view that we can have moral knowledge—and noncognitivism—the view that we cannot.

In Chapter 3, we saw that issues about the sense of words and sentences were cognitive: they had to do with knowledge. So it is not surprising that different views about the content of moral judgments are associated with different views about moral epistemology. Now we have characterized the range of views on the moral content ques­tion, we can look in more detail at the views about moral episte­mology that are associated with them.

5.4