Harman's Argument for Moral Nihilism (Moral Anti-Realism)

The Analogy to Scientific Theories. We believe them because they explain empirical observations.

What about Moral Theories? Do we believe them because they explain observations?

Are there moral observations that are explained by moral theories?

Yes and no. It depends on the sense in which we use the term "observation".

Observation₁ = "an opinion [that] is a direct result of perception"(519). Harman agrees that there are moral observations₁ that are explained by moral theories. For example, on seeing the children set the cat on fire, you immediately judge that it is wrong. Your observation₁: The children's setting the cat on fire is wrong.

Observation₂ = the psychological fact of making an observational₁ judgment. For example, the fact that, on seeing the children set the cat on fire, you immediately judged that it was wrong. Your observation₂ is: You judged that the children's setting the cat on fire was wrong.

Harman's Thesis: Both moral theories and scientific theories explain observations₁; but only scientific theories, not moral theories, explain observations₂.

The example of the proton causing a vapor trail in a cloud chamber. In this case, the observation₁ is part of the explanation of the observation₂.

This is not true of moral observations.

"A moral [observation₁] does not seem to be observational evidence for or against any moral theory, since the truth or falsity of the moral [observation₁] seems to be completely irrelevant to any reasonable explanation of [the moral observation₂]"(520).

"Observational evidence plays a part in science it does not appear to play in ethics, because scientific principles can be justified ultimately by their role in explaining [observations₂] . . . . Apparently, moral principles cannot be justified in the same way"(521).

What is the implicit principle in Harman's argument?

Harman's Explanationist Principle of Epistemic Rationality (EPER): "We can have evidence for hypotheses of a certain sort only if such hypotheses sometimes help explain [observations₂]"(522). This is a principle of rational belief, related to Occam's Razor ('Don't multiply entities beyond necessity.') It is a normative principle of to-be-believedness.

Harman agrees that this principle is too strong. But none of the exceptions helps to make it rational to believe moral theories. Why not? Because all of Harman's exceptions are examples of purely descriptive theories that are reducible to other purely descriptive theories that do explain observations₂.

So Harman's Challenge is this: How can we be justified in accepting moral theories? He thinks there are only two alternatives, neither very promising: That ethical theories explain observations₂ or that they are reducible to theories that explain observations₂.

What is Sturgeon's Reply to Harman's Challenge?

Moral theories do explain observations₂!

Sturgeon claims that moral theories (in combination with auxiliary hypotheses) have empirical implications.

For example:

(1) Hitler was a morally admirable person.

(2) A morally admirable person would not instigate the death of millions of [innocent] persons.

Prediction: Hitler would not have instigated the death of millions of [innocent] persons.

The failure of the prediction requires us to give up one of the premises. The obvious premise to give up is (1).

To answer Harman's Challenge, Sturgeon must argue that there is no difference in principle between moral and scientific explanation. Both kinds of explanation provide explanations for both kinds of observations.

Argumentative Strategy: Assume the existence of moral facts and see whether they are play a role in explaining any observations₂. If not, discard them.

What is Sturgeon's reply to Harman's bubble chamber example[WJT1] ?

Sturgeon provides several examples in which moral facts and moral theories do seem to play a role in explaining observations₂.

Sturgeon's examples: Hitler, the Donnner Party, and the burning cat.

Resuscitating Harman's Challenge

Harman could agree with Sturgeon that on our ordinary understanding, moral facts and theories do seem to explain observations₂. It is only when we consider the full evolutionary history, both biological and social, of our moral practices that we find moral truths to be explanatorily irrelevant.

Revised Harman Argument for Moral Nihilism

Suppose it is true that there is a complete explanation of our capacity for moral judgment from evolutionary biology and that there is a complete social psychological explanation of our moral practices (the results of developing and employing those capacities). Biological evolution proceeds by natural selection. Moral theories play no role in the theory of natural selection. Harman could argue that there is no role for moral theory to play in social psychology either. This is because moral theories are normative theories, but biological and psychological theories are purely descriptive. But if there is a complete explanation of our development and exercise of moral judgment that makes no reference to moral truths, then moral truths are explanatorily irrelevant to our moral beliefs and thus to our moral observations₂.

On this account, Sturgeon's moral explanations would be regarded as useful heuristics, but not as true explanations. For example, we can explain the orientation of morning shadows, by the "fact" that the sun rises in the East, even though it is false.

A Problem for the Revised Argument: It applies to all normative beliefs, beliefs about epistemic rationality (to-be-believedness) and non-moral practical rationality (non-moral to-be-doneness), as well as to morality (moral to-be-doneness). But for Harman to make his argument, he at least needs to be able to appeal to judgments of epistemic rationality (e.g., his EPER). His argument supports nihilism about all normative beliefs, including the very normative principles (EPER and deductive logic) that he needs to make the argument.

So Harman's Challenge has led us to a deep puzzle: How can it be rational to believe any normative truths?