1.1 Understanding Normative Explanations
MORAL THEORIES DO NOT PURPORT merely to tell us which things we ought to do. They also try to tell us why we ought to do them. Moral theories, that is, generally have explanatory ambitions. What they try to explain to us is not why we think we ought to do certain things, of course, or why we do some things, but why we ought to do things. Little has been said, however, in a general vein, about how moral or, more generally, normative, explanations work--what sort of things they are, in what ways they are like and unlike explanations of non-normative or descriptive phenomena, and so on. And that is unfortunate--for given the importance of explanatory ambitions in moral theorizing, differences in expectations about how moral explanations can and cannot work could potentially be playing an important role in underwriting disagreement about many other questions in moral theory. In fact, I think that this is the case. But the only way to see whether this is so is to look hard for implicit theories about how moral or normative explanations must work.
The best way to look for implicit theories about normative explanations is to look for arguments which, once spelled out carefully, turn out to need assumptions about such explanations in order to work. In this paper I want to closely examine such an argument. It is originally due to Ralph Cudworth, and it is one of at least four different arguments that he offered against voluntaristic ethical theories. Cudworth's argument has since been widely held to conclusively establish its result; it was very influential in the 18th century, and arguments like it have recently been reiterated or endorsed by philosophers like Jean Hampton and Christine Korsgaard. (1) Voluntaristic theories, as Cudworth understood them, say that obligations derive from commands or decisions. Those commands or decisions may be those of God (as with Ockham, Descartes and the Calvinists), those of a temporal sovereign (as with Hobbes or Protagoras) or even those of anyone whatsoever (as Cudworth understood Epicurus to claim). Cudworth held that his argument worked against all of these views and, following Cudworth, Richard Price held that it worked against many other views as well. (2) But for concreteness, it is easier to focus on a single view.
1.2 Cudworth and Voluntarism
So consider theological voluntarism: the theory that every obligation derives from one of God's commands.
Voluntarism: For any person x and action-type a, if x ought to do a, that is because God has commanded x to do a.
Cudworth argues like this: (3) the voluntarist has to admit that in order for his theory to be true, God--or at least His commands--have to be pretty special. After all, we can all agree that when I command you to do something, it does not become the case that you ought to do it. So God's commands have to be different in some way from mine--they have to have authority, as Cudworth puts it. But what does the authority of God's commands consist in? Surely just this: that you ought to do what God commands. It is surely because you ought to do what God commands, while it is not the case that you ought to do what I command, that when God commands you to love your neighbor as yourself, you ought to do that, while when I command you to bring me my slippers, it has no such effect. And surely even the voluntarist has to agree with that much:
Authority Vol: [for all]x (x ought to do what God commands)
But that is exactly what we need to get the voluntarist into trouble. For according to voluntarism, every time that you ought to do something, it is because God has commanded it. But why ought you to do what God commands? According to the theory, this would have to be because God has commanded it. But that is surely incoherent. God could not make it the case that you ought to do what He commands simply by commanding it--if it were not already the case that you ought to do what He commands, then such a command would make no difference, and if it were already the case that you ought to do what He commands, then it would be beside the point. …