The Epistemology of Religious Experience.
In this review I offer some criticisms, mainly to do with Yandell's four provisos, which, as I understand them, are as follows. First, any numinous experience E must be one that its subject S does not believe he would have anyway even if there were no possible object for E. Second, it must be true that if E were nonveridical, S could find this out. Third, E must not be a token of a type of experience such that every experience of that type is nonveridical. And fourth, the proposition for the truth of which E gives evidence must be amenable to "collegial" and "lateral" disconfirmation, without actually being disconfirmed in either of these ways. Collegial disconfirmation, in the case of religious experience, is what we would have if religious experience suggested that there were a being very like God, yet lacking something essential to God's character as classically conceived, and if it seemed unlikely that both such a being and God could exist. Lateral disconfirmation is what we would have if religious experience suggested that there were a being very like God, yet having some feature essentially incompatible with God's character as classically conceived, and if it seemed unlikely that both such a being and God could exist.
The third of these provisos seems all right. Clearly anyone who wants to put any evidential weight at all on religious experience cannot allow a priori moves like "No religious experience is veridical because there is no God". The other three provisos seem more dubious. The first proviso does not allow the religious believer to admit the possibility of overdetermination. I might cheerfully admit that my mystical experiences might have occurred even if God had not caused them, since something else might have caused them; and yet insist that, as a matter of fact, it was God that caused them. (Yandell can meet this, perhaps, by pointing out the difference between saying that one case, or some smallish class of cases, of religious experience was overdetermined - which does not excite scepticism about such experiences' evidential value - and saying that all such cases are overdetermined - which does look fishy.)
But the second proviso also seems rather too strong. If a numinous experience is nonveridical, one possible cause of this is what Descartes calls "a malicious demon of the utmost power and cunning". But (as every undergraduate knows) the malin genie hypothesis is not falsifiable - except perhaps by the malin genie himself. Since the malin genie has the utmost power, he is always able to prevent us detecting his activity. It follows that there is at least one logically possible cause of nonveridicality which it is impossible for a subject of religious experiences ever to detect.
Yandell's first and second provisos thus seem to exclude too much of typical religious experience. People quite often use, as evidence for religious hypotheses, religious experiences which come nowhere near satisfying these two provisos. But it is not as obvious as Yandell thinks that they are not justified in doing so. As for Yandell's fourth proviso, this not only seems to exclude typical uses of the evidence of religious experience, it actually seems to threaten Yandeli's own procedure. For Yandell (rightly) stresses the ways in which the religious experiences of different traditions of faith seem to buttress one another. But are there not also ways in which these traditions seem to oppose one another? Ways which, moreover, look very much like cases of Yandellesque collegial or lateral disconfirmation?
Compare, for example, the Muslim and Christian conceptions of God. Isn't something like the following true? Christian religious experience suggests that there is a being very like the Muslim's God, yet lacking something essential (specifically, unity of a sort which excludes Trinitarianism) to the Muslim's God's character as classically conceived. Moveover, it seems unlikely that both Allah and Yahweh can exist. So Christian religious experience provides collegial disconfirmation of Muslim religious experience (and, of course, vice versa). Again, Christian religious experience suggests that there is a being very like the Muslim's God, yet having some feature essentially incompatible with the Muslim's God's character as classically conceived (for example, being a Trinity, having a tendency to become incarnate, and other features). Now, once more, it seems unlikely that both such a being and the Muslim's God could exist. So Christian religious experience provides not only collegial but also lateral disconfirmation of Muslim religious experience (and, of course, vice versa).
On Yandell's view such experiences must cancel out, just as the opposed Jain and Buddhist views about persons which he considers cancel out. So neither Christian religious nor Muslim religious experience provides any evidence at all for God's existence - except insofar as Christian or Muslim religious experience is separable from the points of conflict between the Christian and Muslim doctrines of God. But Muslims and Christians alike deny strongly the possibility of any such separation. One is reminded of what Yandell himself points out when discussing Jainism and Buddhism: "The meditative traditions themselves are shaped by the doctrines in dispute, and what counts as a religiously genuine experience in part is decided by whether it is an experience in which the correct doctrine is 'seen to be so"'.
If one is sympathetic to the idea that religious experience ought to be taken as some sort of datum that cannot simply be ignored, this cancelling-out is an unfortunate result. For by way of it, Yandell seems to have provided those philosophers who would much prefer simply to ignore the data of religious experience with a perfect excuse for doing so. The moral seems to be that we should abandon the fourth proviso, along with the first and the third, if we want to allow religious experience to have a role in arguments for God's existence. This, of course, means that religious experience will have to take a humbler and less exciting role in such arguments, perhaps merely as providing some prima facie presumption in favour of the idea that some God, or gods, probably exist(s).
Two other, more general, criticisms. First, as Yandell himself concedes, there are many types of religious experience besides direct subject/object (or I/Thou) experience of God. But perhaps Yandell underestimates the importance of this point. It may be that most religious people's experience, and for the most part, is not directly of God, but is rather (as Hopkins put it) of a world "charged with the grandeur of God". That is, the nature of their religious awareness is that it gives a certain pervasive tinge to their entire experience. So, on the one hand, their whole experience is religious experience (of varying intensity, for sure), not just certain privileged bits of it. But, on the other, they would perhaps feel it presumptuous of them to claim that the object of their humble awareness was ever, directly and essentially, God. Second, they might also question the idea that religious, or at any rate mystical, experience should be used as evidence for God's existence, as a confusion of means and ends. The point of arguments for God's existence (they might say) is to clear one sort of obstacle out of the way of more or less direct experience of God: not vice versa. To seize hold of mystical experiences as raw material for philosophical arguments is indeed, to use Yandell's word, "impolite": for this use of mystical experience is incompatible with giving such experience the attention that it deserves in its own right.
These criticisms aside, I heartily recommend Yandell's book.
TIMOTHY CHAPPELL School of Economic and Social Studies University of East Anglia Norwich NR4 7JT UK