¹ Philosophical Works of Descartes, Vol. I (translated by Haldane, and Ross, ), Dover Publications, Inc., 1955, p. 174Google Scholar.

² This is certainly the view expressed in Meditation IV. But in the Principles, Vol. I, p. 233, Descartes says that our will can extend as far as God's.

³ Vol. I, pp. 175–176.

⁴ In fact, Descartes seems to draw an explicit analogy between ‘intellectual’ errors and sins in Meditation IV. See Vol. I, p. 177.

⁵ Descartes does, in Meditation IV, sometimes suggest that one might be constrained to believe by forces of which one is unaware. But none of the obvious interpretations of this makes his view about belief and choice any less paradoxical. On the other hand, there are some indications in other works of Descartes' of a partial recognition of the paradoxical nature of this view. But Descartes never scceeds in bringing himself to face the paradox head-on.

⁶ This should make it clear that the present experiment is quite unlike, and does not involve, the kind of thing at which King Canute proved to be such a miserable and embarrassing failure.

⁷ This connects with the fact that, while I can say of someone that he can believe what he wants, the only obvious way of taking this is as an expression of my complete indifference to whatever it is that he happens to believe.

⁸ There might be certain trivial exceptions to this. For example, I might, I suppose, want to believe that I want to believe something or other. But I am going to ignore such cases.

⁹ We sometimes talk of self-deception where what is at issue is nothing more than stupidity, for example. But I am not concerned with such cases.

¹⁰ I said earlier that I couldn't imagine myself ‘simply opting for insanity’. But there are cases in which we say of people that they chose to let go (of their sanity)—even that they decided to go mad. I would argue, however, that the choice involved in such cases—if there is one—must be of a rather dubious kind. It might be, for example, similar in many respects to the (parody of a) choice involved in certain cases of self-deception.

¹¹ There are, of course, quite unproblematic cases of ignoring evidence. But they do not involve what is being discussed here.

¹² I recognize that this might be thought to be a controversial remark. But it is considerably less controversial if it is understood as applying only to those cases of unintelligibility that are also cases—and are seen to be—of propositions that cannot be true. And this is all that I require.

¹³ See note 8.

¹⁴ Neither of the possibilities mentioned in this paragraph should be thought of as involving any genuine or worthwhile kind of freedom with respect to my beliefs. The first might involve a kind of freedom with respect to the facts, but is, at best, only misleadingly thought of as involving a kind of freedom with respect to my beliefs. And the second not only involves the having of an unintelligible belief; it is, anyway, a matter of my believing in accordance with what I see as rational considerations.

¹⁵ In fact, there are. But there is no need to describe any such cases here.

¹⁶ The peculiar remark ‘Unlikely, but true!’, familiar to frequenters of freak shows and readers of Ripley's Believe It Or Not, trades on a piece of sleight-of-hand with respect to the evidence, for example, that is presupposed. Compare the even more blatantly outrageous remark ‘Impossible, but true!’

¹⁷ Not obviously non-rational considerations include hunches, intuitions, ‘funny feelings’, and so on. To be sure, a person might think of a hunch, for example, as non-rational. But if he thinks of it as non-rational in the way in which most of us think of our wants as non-rational (in connection with beliefs), certain restrictions must be placed on the possibility of his believing on the basis of it.

¹⁸ I am not going to try to defend this view here—any more than I am going to try to defend the view that the having of an unintelligible belief requires some element of blindness. But I have tried to defend these views elsewhere.

¹⁹ There is, I think, some reason for saying that the answer to this should be ‘No’. But I don't think that the point is nearly decisive enough to rely on here.

²⁰ There are (justifying) reasons for beliefs that are not happily thought of as evidence. And I suppose that there might be cases in which evidence cannot play the role that is played by such reasons. If so, there will be two kinds of direct roles with respect to beliefs, neither of which can be played (with epistemological transparency) by wants. But I am going to ignore this complication.

²¹ Or about any given candidate for a belief. And the same kind of point applies, in what follows, to what I say about actions.

²² Plato, of course, has a certain implausible view of the second of these relations, and so would disagree with this. But this is hardly the place to discuss his view.

²³ It is this that provides the contrast to, and so makes clear what is, in the present context, involved in, the possibility of my wants ‘producing’ an action of mine indirectly. Consequently, it would be quite beside the present point to say that my wants ‘produce’ (most of) my actions indirectly if what is meant by this is, for example, that (most of) my actions are not ‘basic actions’.

²⁴ What I have said in this paper has a fairly direct bearing on certain ‘large’ philosophical issues. For example, it provides a crucial part of the response that should be made to the scepticism with respect to reason that Hume takes so seriously and that hovers dangerously near the surface of the Meditations at certain points. Again, what I have said has an obvious bearing on some of the problems of Freedom and Determinism. But this is not the place to go into these matters in any detail.