A. Self-Referential Problems

Now consider (CP) itself. First, it isn’t properly basic according to the classical foundationalist’s lights. To be properly basic, it would have to be self-evident, incorrigible, or Lockeanly evident to the senses. But first, it isn’t self-evident for the foundationalist (or for the rest of us). Even if someone claims it has some intuitive support, one couldn’t with a straight face claim that it has enough intuitive support to be self-evident. For if it were self-evident, it would be such that it isn’t even possible for a properly functioning human being to understand it without seeing that it is true.¹⁰⁴¹⁰⁴   See WPF, p. 109. Clearly (CP) isn’t like that at all; for example, I understand it, and I don’t see that it is true; and I’ll bet the same goes for you. In this regard (CP) is wholly unlike 2 + 1 = 3 or If all cats are animals and Maynard is a cat, then Maynard is an animal. Second, it isn’t about anyone’s mental states and therefore isn’t incorrigible for the foundationalist (or any of the rest of us). And third, it obviously isn’t evident to the senses.

According to (CP) itself, therefore, (CP) is not properly basic. That means that if (CP) is true, those who are within their rights in believing (CP) must believe it on the evidential basis of other propositions—propositions that are properly basic and that evidentially support it. And if they do, in fact, believe it in that way, then there will be good inductive, deductive, or abductive arguments to (CP) from propositions that are properly basic according to (CP). As far as I know, there aren’t any such arguments. As far as I know, no classical foundationalist has produced any such arguments or proposed some properly basic propositions that support (CP). It is of course possible that there are such arguments, even if so far no one has produced them; but the probabilities seem to be against it. So probably one who accepts (CP) does so in a way that violates (CP); (CP) lays down a condition for being justified, dutiful, which is such that one who accepts it probably violates it. If it is true, therefore, the devotee of (CP) is probably going contrary to duty in believing it. So it is either false or such that one goes contrary to duty in accepting it; either way, one shouldn’t accept it.

But couldn’t one who accepts (CP) perhaps find a sort of inductive argument for it?¹⁰⁵¹⁰⁵   See Philip Quinn’s “In Search of the Foundations of Theism,” Faith and Philosophy 2 (1985), pp. 474ff.; my response, “The Foundations of Theism: A Reply,” Faith and Philosophy 3 (1986), p. 298; and Quinn’s rejoinder, “The Foundations of Theism Again,” in Rational Faith: Catholic Responses to Reformed Epistemology, ed. Linda Zagzebski (Notre Dame: University of Notre Dame Press, 1993), pp. 22ff. I am grateful to Quinn for showing that this possibility needs to be taken much more seriously than I had been taking it. Perhaps the defender of (CP) (‘the classicist’, as I’ll call her) reads Roderick Chisholm¹⁰⁶¹⁰⁶   Theory of Knowledge, 3d ed. (Englewood Cliffs, N. J.: Prentice Hall, 1989), p. 7. See also “Reason and Belief in God,” pp. 75ff. and embraces ‘particularism’; she proposes to develop a criterion of justified belief by assembling samples of justified and samples of unjustified belief and finding a criterion that best fits them. She assembles a reasonably large and representative sample J of cases of beliefs that, as she thinks, are justified, such that the believer is dutiful in accepting them, and another such sample U of beliefs that she takes to be unjustified, accepted in such a way as to flout intellectual duty. Then perhaps she notes that all of the beliefs in J but none in U conform to (CP); she conjectures that a belief is justified if and only if it conforms to (CP). This would be an inductive argument, of sorts, for (CP).

Here is the question, however: are its premises properly basic according to the classical picture? The premises include, crucially, the propositions with respect to each member of J that it is justified and of each member of U that it is not justified. What form do such beliefs take? Well, presumably the sample classes would include such propositions 96as S₁ is justified in believing B₁ in circumstances C₁ and S₂ is not justified in believing P₂ in circumstances C₂. (The sample classes need not include only actual beliefs, so to speak; they should also include clear cases of beliefs that would be justified in certain circumstances, whether or not anyone has ever actually held such beliefs in those circumstances.) And presumably these are beliefs she accepts in the basic way. (She can’t, of course, use (CP) to arrive at them; that would be blatantly circular.) Clearly, beliefs of these sorts aren’t either incorrigible or evident to the senses for her; so if they are properly basic, then, according to (CP), they would have to be self-evident.

Now here the classicist will be told that she has a really nasty problem: she will be told that there aren’t any cases at all where it is self-evident that a belief is unjustified, such that the believer has gone contrary to duty and in fact warrants disapproval and blame. The alleged reason is that our beliefs are not within our direct control; one can’t just decide to hold or withhold a belief. If you offer me $1,000,000 to believe that I am under 30, or even to stop believing that I am over 30, there is no way (short of mind-altering drugs, say) I can collect. Still, this is by no means the whole story. A full examination of this question would take us too far afield, but first some of my beliefs are indirectly within my control (in the way in which, for example, my weight is), even if I can’t simply decide what to believe and what not to. I can train myself not to assume automatically that people in white coats know what they are talking about; I can train myself to pay more attention to the evidence, to be less credulous and gullible (or less cynical and skeptical), and so on.

Furthermore, some of my beliefs or belief states are, in a way, within my direct control. I don’t at the moment have a belief on the question of the year of George Washington’s birth; a quick look at my encyclopedia or a call to my eighth-grade history teacher would remedy this deficiency. It is therefore directly within my power to bring it about that I have a belief on that topic. We might even go on to say that there is a belief on that topic (the one the encyclopedia reports) such that it is directly within my power to bring it about that I have that belief. Still further, I can be in a state of epistemic sin by virtue of failing to have a certain belief. If it is my responsibility to care for a child and I see her playing with a suspicious looking bottle but don’t take the trouble to examine its label, I can’t expect to deflect blame by claiming that I didn’t know the bottle contained poison. I should have known. (“I didn’t know the gun was loaded” doesn’t always suffice for self-exculpation; it might be my responsibility to know.) And there are plenty of other ways to be in epistemic sin by virtue of the beliefs you hold or don’t hold. I believe that you failed to pay your income taxes last year because X, whom I would have known to be irresponsible had I made any inquiries, said so; and I was in the wrong not to make further inquiries.) I am malicious and wish you ill; the speaker says your thought is deep and rigorous; by virtue of my ill will, I form the belief that what she said is that your thought is weak and frivolous. Out of vanity and pride, I may form the belief that my work is unduly neglected when the fact is it gets more attention 97than it deserves. And so on. Further, in these cases it is perhaps self-evident that the beliefs in question are unjustified, formed in a way contrary to duty; at any rate I am not prepared to dispute the claim.

So suppose we accommodate the classical foundationalist by stipulating that at any rate there are some cases of self-evidently unjustified belief: there still remains a real problem for the classicist. That is because these cases, at least the ones I can think of, lend no support to the claim that it is unjustified to form a belief that is neither properly basic (according to classical standards) nor believed on the basis of such propositions. More important, aren’t there cases where a belief is formed according to (CP), but is nevertheless unjustified? I shouldn’t form the belief that you failed to pay your taxes last year on the basis of merely casual inquiry; the stakes are too high. But suppose I do just that: your false friend Myrtle tells me you didn’t pay them; I believe this in the usual way, a way, let us assume, that conforms to (CP); I am nevertheless unjustified in that belief.

And on the other side, aren’t there any number of cases where it is self-evident that a belief not formed in accord with (CP) is justified? Someone asks you what you had for breakfast; you reply that it was an orange and some cornflakes. You can’t really think of any propositions that are properly basic according to (CP) and support your memory belief; but isn’t it self-evident that you are not guilty, not worthy of reproof or blame, in so believing? And of course there will be an enormous number of examples of this sort. And the relevance of this is as follows: if the samples are chosen in any responsible and plausible way (if they are appropriately ‘random’), they will not support that conjecture that a person is conforming to intellectual duty if and only if her beliefs conform to (CP). Hence, I can’t see how a devotee of (CP) could responsibly argue for it by way of such an inductive, particularist procedure; and hence I conclude that there is probably no way in which the classicist can argue for (CP). If so, however, then (because she also holds that (CP) is not properly basic) she will be unjustified in believing (CP) if it is true; it is therefore self-referentially incoherent for her.