IV. Problems with the Classical Picture

The classical picture has been enormously influential in guiding thought about the de jure question; its near relatives still dominate discussion of it; in particular, the evidentialism of the classical picture persists. This picture, however, like some other big pictures, doesn’t survive close examination; it is subject to powerful, indeed, fatal objections. After pointing out some of the problems, I’ll consider contemporary analogical extensions of the various elements of the classical picture to see whether any of them supports the evidentialism that is still widely popular and finds such a comfortable home in the classical picture. I’ll conclude that in fact there is no reason at all to think that Christian belief requires argument or propositional evidence, if it is to be justified. Christians—indeed, well-educated, contemporary, and culturally aware Christians—can be justified, so I shall argue, even if they don’t hold their beliefs on the basis of arguments or evidence, even if they aren’t aware of any good arguments for their beliefs, and even if, indeed, there aren’t any. Indeed, it is obvious that they can be justified in this way; as I shall argue, that suggests that the de jure question we seek is not this question of justification; that question is too easy to answer.

So, first, what are these problems attaching to the classical picture? Recent philosophy has not been kind to classical foundationalism; many objections have been raised, many problems pointed out. I shall confine my attention to two objections, both fatal. First, as I’ve argued elsewhere,¹⁰²¹⁰²   “Reason and Belief in God,” pp. 61ff. classical foundationalism appears to be self-referentially incoherent: it lays down a standard for justified belief that it doesn’t itself meet. More exactly, the classical foundationalist, in asserting (and presumably believing) his classical foundationalism, lays down a standard for being justified, blameless, within one’s intellectual rights: a standard which his own belief in the classical picture doesn’t meet. Stated at slightly greater length, what he claims is that

(CP) A person S is justified in accepting a belief p if and only if either (1) p is properly basic for S, that is, self-evident, incorrigible, or Lockeanly evident to the senses for S,¹⁰³¹⁰³   Here I am reading Locke (see above, pp. 76–77) as claiming that what I know immediately is only that my sensations are caused by external objects of some kind or other, not that those objects have the properties of trees, horses, or the other sorts of objects we think there are. or

(2) S believes p on the evidential basis of propositions that are properly basic and that evidentially support p deductively, inductively, or abductively.

Here I ignore the fact that the ‘believes on the basis of’ relation is not transitive. The classical picture doesn’t really require that all of one’s nonbasic beliefs be believed on the evidential basis of basic beliefs; some nonbasic beliefs may be believed on the basis of other nonbasic beliefs that support them, provided those others are believed on the basis of still other beliefs that support them, provided those others. . . . To put this more accurately, say that a nonbasic belief is properly based if and only if it is believed on the evidential basis of beliefs that are either properly basic or properly based. Then, according to the classical picture, every nonbasic belief must be properly based.

Further, I ignore another condition that is really part of the classical picture. Suppose I believe p on the basis of propositions q₁, q₂ . . . q_n where the q_i in fact support p, but I can’t see that they do. (Perhaps I believe that there is no greatest cardinal on the basis of ordinary axioms for set theory, but don’t know, can’t see, and have no reason to believe that the latter support the former.) Then, presumably, on the classical picture I am not justified in this belief. My duty is to believe a nonbasic proposition on the basis of propositions that I can see support it, not just any old propositions that happen to support it, whether or not I can see it. Hence perhaps we should add what Locke and Descartes take for granted here: if S is justified in believing p on the basis of other propositions, it must be that those other propositions support p, of course; further, S must also recognize that they do so.