PERCEIVING GOD: THE EPISTEMOLOGY OF RELIGIOUS EXPERIENCE, by William P

PERCEIVING GOD: THE EPISTEMOLOGY OF RELIGIOUS EXPERIENCE, by William P. Alston. Ithaca, NY: Cornell University Press, 1991. Pp. xii and 320. $36.95 (Cloth).

JONATHAN L. KVANVIG, Texas A&M University.

This work addresses the epistemic value of a kind of religious experience, what Alston calls mystical perception. Alston argues that mystical experiences are (often) perceptual, though they do not count as sensorily perceptual, and his aim is to show that mystical perception plays much the same role in justifying beliefs about God that sensory perception plays in justifying beliefs about the external world. The route to this conclusion involves, first, a defense of the perceptual character of (some) mystical experience and, second, an attack on the epistemic pretensions of sense perception. Regarding the first point, Alston argues primarily against the view that mystical experience is a subjective mode of consciousness that the individual interprets as due to some transcendent cause, and claims that on any adequate theory of perception, the best way to understand (some) mystical experience is in terms of a perceptual model.

It might seem that once the perceptual character of mystical experience has been defended, the epistemic value of it would be the immediate pay-off. Such is not the case, however, for Alston's theory of justification has a reliability component to it. Because of this feature, it would be too facile to maintain that prima facie justification is conveyed by perceptual takings, say, in the manner of Chisholm.[i] Instead, the question must always be asked whether a given practice of forming beliefs is reliable, sufficiently likely to produce true beliefs. Alston admits that mystical perception cannot be shown in any non-circular way to be reliable, but argues at great length that this admission is not damaging. For, he argues, sense perception is no better off; it, too, can be shown to be reliable only circularly. Alston suggests further that this result can be generalized to any of the ways in which we standardly form beliefs (introspection, memory, reasoning, etc.).

Given our inability to show that such practices are reliable, the question arises how to sort between those practices we cherish and those, such as reading tea leaves or the entrails of sacrificial animals, we castigate. Alston defends a "doxastic practice" approach to this question, on which there is no epistemological court of appeal beyond our firmly establish doxastic practices. Thus, we are prima facie reasonable when we form beliefs based on perception but not on reading tea leaves, because the former but not the latter is a firmly established social practice.

The crux of the argument is found in the application of this epistemological machinery to the case of mystical experience. Alston argues that Christian mystical perception does constitute a doxastic practice, and thus that it is prima facie reasonable to think that beliefs formed on this basis are justified. Furthermore, he claims the inconsistencies within this doxastic practice and between this doxastic practice and other practices are not so great as to override this prima facie rationality. Finally, Alston argues that the problem of religious diversity--the difficulty that arises from noticing that adherents of different religious traditions engage in alternative practices that lead to the formation of beliefs incompatible with those of Christianity--is not sufficient to undermine the rationality of engaging in any one of the differing practices belonging to the genre of mystical experience.

Alston's epistemological machinery is at work in defending each of these claims. Alston argues that attempts at undermining the epistemic value of mystical perception hinge primarily on two important fallacies, the fallacy of epistemic imperialism and the fallacy of the double standard. The fallacy of epistemic imperialism occurs when an opponent argues that Christian mystical perception is unreasonable because it does not meet certain criteria that are required of sense perception, requirements such as the ability to predict when the experiences would occur, or some intersubjective requirement on objectivity. Alston notes that such imperialism, if applied to other doxastic practices such as introspection, would imply a lack of rationality there as well. The fallacy of the double standard occurs when it is insisted that mystical perception meet standards that sense perception has been shown to be incapable of meeting.

Nonetheless, Alston admits in the end that mystical perception does not fare quite as well as sense perception from the epistemologist's point of view. The reason mystical perception loses out to sense perception is because of the problem of religious diversity. Alston defends that it is rational to engage in the practice of forming beliefs on the basis of mystical perception in spite of the problem of religious diversity, but he grants that the problem weakens the rationality of engaging in the practice as compared to that of the practice of forming beliefs on the basis of sense perception.

In all, Alston has gone a considerable distance toward undermining rational support for the dominant, negative view concerning the epistemic value of mystical experience. His arguments for the perceptual character of (some) mystical experience are persuasive, and his challenge to the skeptic about the epistemic value of these perceptions will be difficult to counter. In regard to this latter point, Alston is clearly at his best when arguing against negative assessments of religious experience, and his discussion of the arguments against the epistemic value of religious experience is at least half a Prudential-Bache commercial: rock solid, even if not market wise. Unfortunately, the same cannot be said of his attempt to carry out the positive task of defending the epistemic value of religious experience. The qualms I have center around the epistemology that drives the argument.

First, Alston's negative critique of the epistemic respectability of sense perception is flawed by an inordinately high standard regarding what it would take to show that sense perception is reliable. It is clear from a variety of passages that Alston is intent to argue that there is no sound and convincing deductive argument for the reliability of sense perception, but he gives no defense that such a high standard is the appropriate one. A weaker standard would be that showing that sense perception is reliable requires defending the claim that we are justified (perhaps in Alston's sense of `justification') in taking sense perception to be reliable. Perhaps an argument for Alston's negative assessment of sense perception could be carried through using this weaker standard, but Alston does not do it. This fact is important, for much hinges on Alston's insistence that sense perception not be assumed to have some exalted epistemic status from the outset. By imposing such a high standard on the task of showing that sense perception is reliable, the work fails to defend parity between mystical perception and sense perception. After all, my pole-vaulting equality with Sergei Bubka is not defended by showing that neither of us can vault over the moon.

A more important worry, however, concerns Alston's notion of rationality, which he employs in claiming that it is rational or reasonable to engage in firmly rooted doxastic practices. Given the important role the concept plays throughout the entire work, it is surprising that Alston spends only two pages of the entire work telling us about this notion. Even more surprising, his characterization is primarily negative: "I said in section ii that it is rational for us to engage in [established doxastic] practices. ... It is a kind of practical rationality that is in question here. ... I call this rationality "practical" to differentiate it from the rationality we would show to attach to a belief if solid grounds for its truth were adduced..." (p. 168). What Alston tells us is that practical rationality is not, let us term it, epistemic rationality: the kind of rationality that requires reliability for it to characterize a belief. The problem with this characterization is that there would seem to be a number of conceptions of rationality that fit this description. There is true practical rationality, the kind that accrues to a practice given our practical goals of survival and well-being. We might say that doing X is truly practically rational when it is the best means available for furthering our survival and well-being.[ii] There is also what we might term subjective rationality, the kind of rationality that accrues to a belief when persons are following those practices that seem, from a particular point of view, the best to follow in the pursuit of truth and avoidance of error. When the point of view in question is the point of view of the person holding the belief being evaluated, we might term the type of rationality involved egocentric rationality. Egocentric rationality is, however, not the only kind of subjective rationality. We can evaluate a belief from the point of view of some third party, or from our own point of view, or from the point of view of "the common man". In any case, the point is that merely indicating that his sense of rationality is not the epistemic sense does not serve to identify the kind of rationality Alston has in mind.

Moreover, there is some reason to doubt there is any one understanding of rationality that can do all the work to which Alston puts the notion. In defending the rationality of engaging in firmly rooted doxastic practices, Alston seems to emphasize truly practical concerns. He says,

First, consider the rootedness of these practices in our lives. Our basic doxastic practices are firmly entrenched long before the age of reflection... . Thus they are a much more ineluctable part of our lives than are habits, dispositions, and practices that are acquired by deliberate effort later in life. Even if it were possible to abandon or alter them, it would be a very arduous task. Hence, in the absence of extremely good reason to do so, the effort would be ill advised (p. 169).

Here it is the fact that the effort to change would be arduous that makes for the rationality of engaging in the practice.[iii] This emphasis on the arduousness of the task lends itself most readily to an interpretation in terms of true practical rationality, for the more difficult a task, the more costly, in terms of our well-being or survival, to undertake. Alston also says,

Let me sum up . . . the treatment we have provided in this chapter. It is obvious that the argument to and from practical rationality depends on our focus on practices. A question of practical rationality arises only when we are dealing with what we do. If we were speaking of the truth (validity, acceptability . . .) of principles or the adequacy of grounds, then there could be no question of what it is rational to do, since we would not be discussing doings (p. 174).

Here the emphasis is on the relation between rationality and action. According to Alston, only regarding things that are doable can a question of practical rationality arise.

At other places, however, a different notion of rationality or reasonability seems to be at work. One objection to Alston's view regards his insistence that practical rationality accrues only to socially established practices and not to idiosyncratic practices, and in responding to this objection, Alston says,

When a doxastic practice has persisted over a number of generations, it has earned a right to be considered seriously in a way that Cedric's consultation of sun-dried tomatoes has not. It is a reasonable supposition that a practice would not have persisted over large segments of the population unless it was putting people into effective touch with some aspect(s) of reality and proving itself as such by its fruits. But there are no such grounds for presumption in the case of idiosyncratic practices (p. 170).

The issue I want to focus on from this passage concerns how we are to interpret the phrase `It is a reasonable supposition'. I suggest that we have a different sense of rationality here than true practical rationality. First, a supposition is not a doing, and in the previous quotation Alston limited the scope of practical rational precisely to doings.[iv] Second, the kinds of interests and concerns that underlie the claim of reasonability seem to have to do, not with our well-being and survival, but rather with the truth. To persuade the reader not to take seriously idiosyncratic practices, Alston needs to convince us that if our interests are in the truth of the matter, socially established practices have something going for them that idiosyncratic practices do not.

One might try to interpret the concept of reasonability here in terms of epistemic rationality, the kind that accrues to a belief only when it is objectively likely to be true. Perhaps one thinks that Alston intends to claim that it is epistemically rational to believe that any socially established doxastic practice "puts people into effective touch with some aspect(s) of reality"? Surely he does not, for he notes elsewhere (p. 176) that many socially established practices have died out precisely because they were unreliable. Might he then think that there is a reliable though not exceptionless connection between being a socially established practice and "putting people into effective touch with reality"? He may think this, but we need an argument for that claim. Furthermore, if we had such an argument, we would not need the remainder of the book. For, if we had such an argument, we would have a very quick argument for the epistemic adequacy of mystical perception, one that bypasses entirely the notion of practical rationality. For it would follow from the fact that this supposition of reliable connection is itself justified, that it is objectively likely that sense perception and mystical perception are reliable. So it would seem that we are left only with the hypothesis that we have here a notion of reasonability that is not to be equated either with true practical rationality or with epistemic rationality (the kind that involves reliability as a component).

This difficulty of interpretation is exacerbated by considering what conception of rationality might generate the conclusion that poor Cedric is irrational. If one thinks of true practical rationality, there is nothing to be found in the fact that a practice is not socially established that would warrant an inference to the claim that it is not good for Cedric in terms of well-being and survival to form beliefs as he does. In addition, as Alston is ready to point out, there is no connection to be found between the reliability of a practice and what percentage of the population employs it, so the fact that Cedric's practice is idiosyncratic fails to show that he is epistemically irrational in forming beliefs in the fashion he does. Finally, there is no connection either between what appears from Cedric's own point of view to be the best way to satisfy the epistemic goal of finding the truth and avoiding error, and what practices are followed by his or any other society,[v] so idiosyncracy alone does not imply egocentric irrationality. The best that can be offered here, I think, is if we take the point of view of something like "the common man" in Cedric's culture, and argue that from that point of view, Cedric is subjectively irrational. In any case, it would seem that whatever sense can be found to undergird Alston's claim about the irrationality of poor Cedric, it will have to be a different sense than that which undergirds the rationality accruing to a practice because of the difficulty of altering socially established ones.

Alston's use of the notions of rationality and reasonability appear frequently throughout the work; in fact, they lie at the heart of the primary argument for the thesis of the work. Alston wishes to defend the claim that it is reasonable to regard mystical perception as reliable. I discuss that argument below, where I will note in passing that the notion of rationality involved would seem to be neither true practical rationality nor epistemic rationality. Since so much hangs on the epistemology involved in the book, it is a serious drawback to have left such an important concept obscure.

I turn now to Alston's major argument on behalf of his that it is reasonable to regard the Christian mystical practice as resulting in beliefs that are justified and hence reliable. Alston claims that this thesis follows from the fact that the practice is rationally engaged in. The argument, I will argue, is deeply flawed in a variety of ways, and in order to evaluate his argument, I will set aside the above concerns and grant that Christian mystical practice is rationally engaged in, as Alston claims.

Alston wishes to show that granting the rationality of engaging in any practice commits one to granting the rationality of taking that practice to be reliable. The notion of commitment involved in this claim Alston understands as follows: "When I say that in judging that p I am thereby committing myself to its being the case that q, ... I mean ... it would be irrational (incoherent . . .) for me to judge (assert, believe) that p and deny that q, or even to abstain from judging that q, if the question arises" (p. 179). He begins by first arguing that engaging in a practice commits one to regarding the practice as reliable, just as holding a belief commits one to holding that the belief is true. The crucial argument then follows:

But if one cannot engage in the practice and refuse to admit that the practice is reliable if the question arises, then in judging that the former is rational one has committed oneself to the latter's being rational... For I cannot hold that X is rational and coherently deny (or abstain from judging) that Y is rational, where accepting (engaging in) X commits me to accepting Y. If pursuing a Ph.D. commits me to the belief that it is possible for me to get a Ph.D., then I can't rationally hold both that it is rational to pursue a Ph.D. and that it is not rational to suppose that I can get a Ph.D. The rationality of a practice (action, belief, judgment . . .) extends to whatever that practice . . . commits me to (p. 179).

The idea here seems to be, first, to show what engaging in a practice commits one to, and then to claim that rationality distributes over commitment. As the last sentence implies, the rationality of a practice extends to anything to which that practice commits one.

Note that this argument employs a notion of rationality that Alston glosses in terms of incoherence. How this notion of rationality is related to true practical rationality, or epistemic rationality, or the various types of subjective rationality, I do not know. If the gloss is meant to be taken as a way of clarifying Alston's concept of practical rationality, perhaps we might expect him to embrace a coherence account of rationality. However, Alston has already dismissed coherentism as a theory of justification (p. 73), claiming that it suffers from "crippling disabilities." It would be quite surprising if Alston, without explanation, intends us to presume that coherentism has partaken of the miraculous and is now able to carry the load required. Here, as elsewhere, the reader is left frustrated by a lack of information about the crucial concept of rationality.

My main concern here, however, is with the argument itself. Let us be a bit more careful about the structure of this argument, for in doing so we can see how it fails. The argument begins by making claims about what engaging in a practice commits one to, namely, to taking that practice to be reliable. So if we employ Alston's account of commitment, we first get:

(1) It is irrational to (engage in a practice & (deny that that practice is reliable or withhold on the question of its reliability, if that question arises)).

Alston then claims that "the rationality of a practice... extends to whatever that practice commits me to." I take it that what this claim means is that if we add the further premise that engaging in a certain practice is rational, we can infer that it is rational to regard that practice as reliable. In other words, if we represent (1) above as

(1) ~R(A & B),

and add the further premise

(2) R(A),

the claim that it is rational to engage in the practice in question, we can infer

(3) ~R(B),

the claim that it is irrational to believe the practice to be unreliable or to withhold on the question of the reliability of the practice if the question should arise.

Though Alston does not make the following explicit in the above quotation, we can complete the argument by adding the following principle:

(4) If it is irrational to believe the practice to be unreliable or to withhold on the question of the reliability of the practice if the question should arise, then it is rational to regard that practice as reliable.

It then follows from (3) and (4) that

(5) It is rational to take the practice in question to be reliable.

In sum, applying this argument to the case of Christian Mystical Perception (CMP), we begin with an instance of claim (2), that it is rational to engage in CMP. We then add two crucial principles, namely, (1) and (4). From (1) and (2), we infer and instance of (3), that it is irrational either to deny that CMP is reliable or withhold on the question of its reliability should it arise. Finally, from this instance of (3) together with (4), we infer the conclusion of the argument, an instance of (5), that it is rational to regard CMP as reliable. As Alston puts the argument regarding sense perception (SP),

But then, if I have shown, by my practical argument, that it is rational to engage in SP I have thereby shown that it is rational to take SP to be reliable. For since the acknowledgement of the rationality of the practice commits one to the rationality of its reliability, to provide an adequate argument for the former will be to provide an adequate argument for the latter. Hence our argument from practical rationality, though it does not show that SP is reliable, does show that it is rational to take it to be reliable (p. 180).

The argument summarized here is at the heart of Alston's entire project, and unfortunately it is multiply defective. First, the above quotation commits Alston to the view that rationality is closed under known deduction. In the middle of that passage, he affirms an instance of the following:

(6) If X commits one to Y, then any adequate argument for X is an adequate argument for Y.

One way for X to commit one to Y is for one to know that X to entail Y, for in such a case one cannot coherently suppose both that X is true and that Y is not.[vi] But then, according to (6), any adequate argument for X is also an adequate argument for Y. This claim may be true if we gloss `adequate argument' as `valid deductive argument', but it seems false otherwise. In particular, one thinks of inductive and epistemic arguments: even if Descartes was right that the existence of God implies the non-existence of the omnideceptive demon, it would seem to be a mistake to think that the reasons given in the Third Meditation constitute by themselves adequate reasons for the non-existence of that demon.

Alston addresses this problem in footnote 54 at the bottom of pages 180-181, where he claims that (6) is true on his conception of rationality. He says,

If we were thinking of the rationality involved in my discussion as guaranteeing likelihood of truth, this would be a serious objection. But since it is practical rationality that is involved, the situation is different. Dretske's point is that what renders probable the supposition that it's a zebra does not also render probable the supposition that it's not a donkey with stripes painted on it. But what I am thinking of the rationality of SP as committing one to is not the likelihood of the truth of the claim that SP is reliable (or anything like that), but rather the practical rationality of taking SP to be reliable, the thesis that, given what I have to go on, I am well advised to make and act on this assumption. Thus the qualms expressed by Dretske and Nozick (1981, chap. 3) do not apply here.

In one incredibly weak sense, Alston's conclusion follows, for the Dretske and Nozick counterexamples are directed at some concept of truth-conducive support. But surely something more is needed; we need to know not only that the qualms expressed by Dretske and Nozick are not directed at Alston's conception of rationality, but that they cannot be extended to apply to this conception as well. And there seems to be good reason to think they can. First, there is nothing about Dretske's zebra counterexample that requires an objective interpretation of the concept of probability; even on a subjectivist reading, the counterexample has the same convincing appearance. Note that I am not claiming that the example is decisive; instead, I am only claiming that it is no less compelling when one adopts a subjective interpretation of probability. So the counterexample is not limited in significance to conceptions of rationality, justification, or probability that have an objective connection to truth. More generally, it is well-known that the validity of a deductive argument is insensitive to substitution by logically equivalent statements, but it is also well-known that non-deductive support is more fine-grained than that. On at least some theories of propositional content, we can have reason to think that Hesperus is the morning star without having reason to think that Phosphorous is the morning star.[vii] Or, again, suppose traditional theism is correct, and then ask whether it is possible to have reasons for thinking that 2+2=4 that are not also reasons for thinking that God exists.[viii] Most important, we can have such reasons without antecedently committing ourselves on the question of whether this concept of reasons is truth-conducive. Put another way, the fine-grained sensitivity of the concept of rational support seems to depend only on the non-deductive character of such support, rather than on whether the support is in some way truth-conducive. So, in the end, Alston needs to do more than just to find some difference between his concept of rationality and the concept employed by Dretske or Nozick.

Regarding the major argument above, however, the worst is yet to come. In particular, I do not think (3) follows from (1) and (2). The validity of the argument hinges on the following connecting premise:

(7) If it is rational to engage in a practice and it is irrational to (engage in that practice & (deny that the practice is reliable or withhold on the question of its reliability, if that question arises)), then it is irrational to deny that the practice is reliable or withhold on the question of its reliability, if that question arises.

Using `A' and `B' as above to stand for each of the conjuncts in the antecedent, and `R' for the rationality operator, (7) can be represented as:

(7*) (~R(A&B) & R(A)) ® ~R(B),

which, interpreting the arrow as strict implication, is logically equivalent to:

(7**) ~à(~R(A&B) & R(A) & R(B))

But, of course, (7**) is true only if the following is true:

(8) (R(A) & R(B)) ® R(A&B)

There is reason to doubt the truth of (8). Though I do not have the space to argue for the following claim here, it has been cogently argued in the literature that the generalized version of (8), where `A' and `B' are replaced by propositional variables, is the faulty conjunction principle that lies at the heart of both the lottery paradox and the paradox of the preface.[ix] At the very least, then, we have reason to be suspicious of the inference from (1) and (2) to (3). If the arguments against the generalization of (8) are correct, there would seem to be little reason to think that (8) itself is true, and so little reason to think that (3) follows from (1) and (2). So Alston has failed to show that it is reasonable to regard CMP as reliable.

There is a variety of problems with Alston's work, some a bit ancillary to his main purpose but others that cut to the heart of it. Nonetheless, the book is a welcome antidote to the usual treatments of mystical perception, both in its understanding of the character of mystical experience as truly perceptual and in the sophistication of the epistemology that underlies the evaluation of the epistemic value of such experiences. Here I have focused more on the work's vices than its virtues, but I would be remiss to close without noting that, in spite of the flaws, it is an important work.

NOTES