Keywords

It does not follow immediately from the perceptual nature of religious experience that the beliefs formed on the basis of those experiences are justified. After all, hallucinations are also perceptual experiences, and illusions cause people to form false beliefs about their environments. Why should we think that religious experiences provide even prima facie justification to the beliefs based on them? Many of the critics of the idea of mystical knowledge think that the objects of religious experience, especially God on the monotheistic model, are by their very nature not the kind of things we can make contact with perceptually, whereas the ordinary experiences we have of our physical environment are unproblematic, so even if religious experiences are perceptions of a sort, they can’t really reveal any objective realities. In other words, there is some important difference between sense perception and purported religious perception that disqualifies the latter as a justifying ground of religious belief. Even if it can be shown that religious experiences provide prima facie justification for religious beliefs, they may nevertheless be unjustified ultima facie (to use Senor ‘s (1996) felicitous distinction), because some defeater occurs that overrides that justification.

Purported Defeaters to the Justification of Perception

One way to approach the question of the justificatory force of religious experience, once we grant that it is perceptual in kind, is to ask what justifies other kinds of perceptions, and look to see whether religious perceptions share those features. A thorough undertaking of that task is clearly beyond the scope of this inquiry, but perhaps we can say a few general things in that direction. While it is a contested matter what justifies our perceptual beliefs, it may be that, whatever it is, religious perceptions also have that feature. To put the idea negatively, we could argue that one cannot rule out religious perception without ruling out ordinary perception; that whatever it is that allegedly disables religious perception is in fact a feature of ordinary perception, too. The two main exponents of this line of argument are Alston and Plantinga .

Plantinga and the De Jure Question

Plantinga (2000) approaches the question of justification for beliefs about religious experiences in a defensive mode. That is, instead of asking what justifies these beliefs, he asks what reason there is to deny their justificatory force. Famously, he theorizes (1981) that many beliefs about how God appears to believers are properly basic; that is, they are properly held, and not because they are inferentially related to any other beliefs. If these beliefs are properly basic, then they cannot be criticized on the grounds that they are not grounded in good evidence. His view of justification makes it the case that any belief formed by the use of a faculty that is functioning properly in its proper environment, working according to a design plan that is successfully aimed at truth, is justified (1993); so if there is a faculty at work here, and it is functioning properly (to put it briefly), then the output beliefs are justified. Of course, believers think that their faculties are functioning properly, while non-believers might hold that there is no such faculty, or that the beliefs result from improper function of some faculty. How can we decide between them? It turns out to be quite difficult to answer that question in a noncircular way, as believers will appeal to resources that are reasonable resources, but only if they are the beneficiaries of a properly functioning faculty. A central question then is whether there is any reason to think that those who have religious beliefs are somehow damaged or defective, or behaving improperly. Plantinga calls that question the de jure question; is there any reason to think that religious belief is disreputable simply in virtue of being religious belief? He considers some reasons that people have offered for the positive answer, and finds them all wanting. Having found no in-principle defect in religious belief, he concludes that the discussion of the epistemic status of religious belief depends on an answer to the de facto question, that is, the whether the religious beliefs are true.

Alston and Doxastic-Practice Coherentism

Alston approaches the problem in a slightly different way, by making a direct case for the rationality of religious belief. Instead of asking directly whether it is rational to form religious beliefs on religious experiences, he asks the more general question as to what makes it rational to form any beliefs. In particular, since we are talking about a kind of perceptual experience, what makes it rational for us to accept the deliverances of our senses? It’s not that some sound, noncircular argument can be made for the reliability of the senses; every valid argument for the reliability of the senses appeals to the evidence of the senses at some point, and so they all exhibit epistemic circularity (1986). The same can be said of our other ways of forming beliefs (Alston calls them ‘doxastic practices’), like deduction, induction, memory, and so on. Rather than embrace a skeptical conclusion, Alston argues that we are rational to engage in these doxastic practices because they are the only game in town for forming beliefs about their particular subject matter, and that they don’t produce massively inconsistent outputs, either internally or with the other, equally well-established doxastic practices; those inconsistencies it does produce are either eliminable or not on central matters. Since this is all we can say in favor of our basic package of doxastic practices, it must be sufficient to show that we are rational to engage in them. Practices can be shown to be unreliable if they run into persistent, ineliminable inconsistency, but they can’t be shown directly to be reliable. Alston then argues that what he calls the ‘Christian Mystical Practice’ exhibits all the same features. It is socially established, and does not produce important, ineliminable, massive inconsistencies, either internally or with respect to our other practices. Moreover, if it is the truth, it is the only game in town for coming into perceptual contact with the divine. It is impossible to produce a direct argument for the reliability of the practice without appealing to its own outputs, but the same is true of sense perception, so that is no disability.Footnote 1

Many have argued that Alston ‘s argument fails because he has overlooked some crucial disanalogy between religious experience and sensory experience. Louis Pojman , for example, argues (2008, 130–131) that religious experiences cannot provide justification for belief in God for three reasons: (1) religious experience is “amorphous and varied”; that is, the objects of religious experiences don’t appear the same to everyone; (2) justification of religious belief by religious experience is circular, in that it depends on premises that are not self-evident to everyone; and (3) religious experiences are not confirmed in the same way that sensory experiences are. Pojman is clearly right that sense-experience is richer and more unanimous than religious experience, and that sensory experiences are confirmed in a way that religious experiences are not; but notice that the evidence that confirms sensory experience is all derived from sensory experience, so the justification for sense experience is circular, too. In any case, no grounds for the validity of sensory experience are “self-evident to everyone.” While a great many philosophers have attempted to prove the reliability of sense experience on some ground or other, they all argue from different grounds, and none of them is generally recognized as successful in the attempt. If that is a disability for religious experience, it is equally a disability for sense experience. Pojman has shown that the two types of experience are different, but he has not shown that those differences underwrite different epistemic evaluation.

A more ambitious attempt at refutation comes from Matthew Bagger (1999). He starts from a theory of perception according to which a belief is justified just in case it is the best explanation for our experiences, and argues against supernaturalism generally. The argument involves a general repudiation of Alston ‘s view of perception, and all views that make room for any externalist component.

Likewise, Ulf Zackariasson (2006) argues that, while Alston has shown that there is a structural similarity between sense perception and religious experience, the two fail to be similar in another way that is crucial for an argument from analogy. The important dissimilarity is that sense perception plays a certain functional role in our wider practices, and religious perception plays a quite different role. Sense perception serves a critical function and “frequently forces us to reconstruct our beliefs” (2006, 330). It provides us with a rich and varied range of information so that we are constantly updating and revising our picture of the world. Religious experience, on the other hand, serves primarily to confirm the previously arrived-at picture of its object. Religious dogma provides the final arbiter of genuineness in religious experiences. If your experience contradicts established dogma, then it was not genuine. This dissimilarity, Zackariasson claims, seriously weakens the analogy.

Alston has resources to answer this kind of argument. For one thing, the differences between the two kinds of perception can be accounted for by differences in their objects. The Christian picture of the world, built up in large part from the results of religious experiences, includes claims that, if true, make it likely that religious experiences will not provide very much new information, and will tend to support the previously arrived-at picture the believer has, just as Zackariasson says. Since these features are not unexpected, on the Christian view, they cannot count as evidence against it. The same can be said of the other monotheistic practices. Zackariasson is aware of this response; he’s just not very impressed with it.

It will not help to appeal, at this point, to the fact that the doctrines used to assess numinous experiences partially have their origin in the output of CMP [Christian mystical practice]. Besides the fact that this may be a dubious empirical claim, it does not remedy the difference between CMP and SP [sense perception] as they function presently; in SP, received opinion has not, as in CMP, been allowed to take the status of unquestionable dogma. (Zackariasson 2006, 336)

First, as to the dubiousness of the claim that Christian dogma has its origin partially in the deliverances of religious experience: unless Zackariasson is claiming massive fraud—and he wouldn’t be the first—where does he think the doctrines came from? People’s purported experiences of God, including experiences of him revealing doctrines or inspiring writings, are certainly the foundations of Jewish, Christian, and Islamic doctrines. Take any religious doctrine, and trace its justification back to foundational beliefs, and you will find that the ‘axioms’ on which it depends are either a priori logical or metaphysical principles, on which all reasoning relies, or claims about something that has been experienced. This is true even of those esoteric doctrines that came from the application of abstract philosophical reasoning to Christian doctrine; experience provides the raw material to which the reasoning is applied. According to their own stories, Judaism begins with Moses ’ experiences at Sinai, Christianity begins with the experiences of the disciples in Judea and Samaria, and Islam begins with the prophet Muhammad ‘s experiences at Mecca. Second, he certainly underestimates the degree to which the deliverances of sense perception function as unquestionable dogma. How would we respond to someone who insisted she saw rocks falling up, or reported that the sky is bright red? The well-established and repeated deliverances of sense perception function as practically unquestionable defeaters for claims to such an extent that persistent claims to experiences to the contrary are taken to be symptoms of psychological disorder. The difference is much smaller than Zackariasson thinks.

But Alston need not answer these claims at all. Both Pojman and Zackariasson address Alston’s argument as if it were a straightforward argument from analogy, when in fact it is not one, or even an ‘indirect’ one. Alston does not argue “CMP is like SP; SP is rationally taken to be reliable; therefore CMP is rationally taken to be reliable,” so pointing out disanalogies is beside the point. The structure of Alston’s argument is to show that if a doxastic practice is socially established and free from massive contradiction, either internally or conjoined with other established practices, then it is rational to engage in it. If it is rational to engage in the practice, then it is rational to take its deliverances to be justified. He then argues that Christian mystical practice is a socially established doxastic practice, and argues that it is free from massive contradiction, and so concludes that its practitioners are rational to engage in that practice. While he does draw analogies with sense perception, those analogies serve primarily to illustrate the general argument, which is not an argument from analogy. Early in his article, Zackariasson shows that he is aware of this feature of Alston’s argument.

The reason for characterizing Alston ‘s argument as indirect is that it involves no direct comparisons between instances of sense perception and numinous experience, but rather reflections on the general characteristics of reliable doxastic practices. (2006, 333)

But he misses the consequence of this point: disanalogy alone is not enough to undermine the argument, since neither Plantinga nor Alston makes a simple argument from analogy. Alston argues for a general epistemic principle for evaluating doxastic practices. To show that he is wrong, it is necessary to show that the principle is wrong, or at least that his argument for it is faulty. Disanalogies may be able to play an important role in showing that the general principle is false if it is, but disanalogy alone shows nothing. To see that this is true, one need only consider some of the other doxastic practices (like memory, deductive reasoning, and so on) and how dissimilar they are from sense perception. If disanalogy undermines the credibility of religious experience, it also undermines the credibility of memory and induction.

Some have argued that the principle is too permissive, and when appropriately amended, it no longer countenances religious experience as a rationally engaged-in doxastic practice. In Webb (1996), I argued that Alston ‘s principle was too permissive, in that it would count as rationally engaged-in a practice that was completely untethered from reality, provided only that it could gain adherents over time, and could avoid inconsistency by incorporating a defeater system that forced consistency. The only way we could rule those practices out while preserving things like sense perception, memory, and the rest, was to add that justified practices, besides being established and free from inconsistency, must be natural and inescapable. Since religious doxastic practices are avoidable—one can always quit one and join another, or not join one at all—they don’t fill the bill. But this is clearly not an adequate reformulation of the principle. After all, we can’t define ‘natural’ in a way that excludes all religious practices without begging the question. If people are capable of experiencing God, or nirvana, or whatever, it is because of something in their natures, so that capacity is natural. It is also not at all clear that avoidability should be epistemically decisive. While it is true of our basic package of doxastic practices that they are the only game in town for learning about their characteristic objects, that is not the same as their being unavoidable. Even if we could opt out of sense perception, it would still be the only game in town for discovering things about the dispositions of physical objects in our vicinities. The fact that a practice completely untethered from any reality could qualify as rationally engaged in remains as a worry for the view, but the amendment to the view that requires that practices be natural won’t do.

Critics of Alston and Plantinga point out that the argument they make for the rationality of religious belief based on religious experience is potentially available to the adherents of all religions,Footnote 2 and so provides no special support for the Christian beliefs they favor. G.A. Cohen states the problem nicely:

Suppose that identical twins are separated at birth. Twenty years later, they meet. One was raised as, and remains, a devout Presbyterian. The other was raised as, and remains, a devout Roman Catholic. They argue against each other’s views, but they’ve heard those arguments before, they’ve learned how to reply to them, and their opposed convictions consequently remain firm.

Then each of them realizes that, had she been brought up where her sister was, and vice versa, then it is overwhelmingly likely that (as one of them expresses the realization) she would now be Roman Catholic and her sister would be Presbyterian. That realization might, and, I think, should, make it more difficult for the sisters to sustain their opposed religious convictions. (2000, 8)

The reaction of these hypothetical twins is appropriate because they realize that they continue to hold their convictions in spite of neither of them having better grounds for her belief than the other (2000, 13). Both Alston and Plantinga acknowledge this, but do not think this result is a problem for their arguments. Suppose that their arguments do show that Christians are rational in their Christian beliefs, and that the same argument can be used to show that Buddhists are rational in their Buddhist beliefs. It is tempting to think that a style of argument that can justify anything justifies nothing. In other words, the mere fact that equal evidence can be adduced for different, mutually inconsistent belief systems undermines our justification for believing either; each acts as a defeater to the other. It counts as exactly the kind of interpractice massive inconsistency that shows a practice to be unreliable, and so not rationally engaged in.

There are a variety of possible principles here, ranging from the radical claim that diversity completely defeats whatever force religious experiences might have to the grudging admission that religious diversity should reduce our confidence, but not enough to make any real difference in the epistemic status of our religious beliefs. Robert McKim (2001) offers a pair of modest principles that are close to the ‘grudging admission’ end of the spectrum. He argues that disagreements in general (under certain carefully defined conditions) trigger an epistemic situation in which a burden is placed on the believer. The awareness of disagreement makes it the case that we have “an obligation to examine” (140) our beliefs on the matter in question, and is reason for our beliefs in that area to be “tentative” (141). David Basinger argues that if one sincerely wants to maximize truth and avoid error, and also wishes to maintain an exclusivist religious belief, one takes on an obligation to try to resolve the disagreement (1991). It’s hard to know what to make of these kinds of proposals, especially since the facts about the epistemic status of any given belief depends on the particular intelligence, education, and other circumstances of the believer.Footnote 3 It’s also unclear what an obligation to behave in a certain way (e.g., to try to resolve disagreement, to hold one’s beliefs tentatively, or to examine one’s beliefs) entails as regards the actual epistemic status of the belief. There’s no logical barrier to having an obligation to examine a belief, even though the belief is on perfectly solid epistemic ground. Suppose I re-examine my belief and its grounds, and find that I have made no mistakes; am I not then justified in retaining my original belief, even in the face of disagreement? In at least some cases, I can’t look at the other person’s grounds and “check his work,” since he is ex hypothesi a member of another practice, with resources I don’t have access to. This is particularly clear on externalist views; since part of what justifies the belief is the fact that makes the belief true, I can’t check someone else’s beliefs against my own without knowing first what is objectively true.

Neither Alston nor Plantinga admits that mere awareness of religious diversity removes the religious believer’s justification for her beliefs. Plantinga argues (2000, 437–442) that if you have adequate reason for your own belief, the presence of beliefs inconsistent with your own, even among apparently rational people, does not decrease your belief’s justification, and should not decrease your confidence in your belief. This is because whatever warrant your belief enjoyed before you found out about rival beliefs remains intact; if your belief was properly formed, that fact about its origin is still a fact. What is wrong with Cohen ’s argument is that neither of the hypothetical twins should grant that her reasons are no better than her sister’s. A thoughtful religious believer, one who actually has reasons for his or her belief, need not concede that people in other religions are in exactly the same epistemic condition. This is in line with our other doxastic practices; the mere fact that we are aware that other people disagree with us (consider the case of those who believe in a flat earth, for example) is rarely taken to count against our justification. The plurality of religious beliefs does not even constitute a probabilistic defeater; since probability would have to be assessed with respect to background information, and the believer’s background information includes beliefs that support the belief in question, then the probability of the belief should remain unchanged, even in the face of a plurality of beliefs. Alston admits that the fact of religious diversity should lower a believer’s subjective confidence, but denies that it should be taken to “undermine the credibility” of the religious practice (1991, 266–278). Every mystical practice is internally coherent, and insofar as they are separate practices, there is no noncircular way to decide among them. But that means that each one, by itself, is consistent with our basic package of doxastic practices, and so each one is rationally engaged in. Since they are independent, they cannot act as defeaters for one another.

If known religious diversity is to count as a defeater for religious belief, then it will have to be because of some special features of the religious case. There are, of course, differences between religious disagreement and other kinds. People who disagree about the shape of the earth share sensory access to the same physical evidence, and so their difference of opinion can be adjudicated, even if they can’t be brought to agreement. In the religious case, the resources that exist for adjudicating disagreements are internal to the different practices, so no disagreement between practices can be adjudicated by neutral principles. If Alston and Plantinga are right, the mere fact of religious diversity does not make it irrational to maintain religious belief. The argument then has to turn on the facts about the individual practices. Can we find actual evidence of unreliability for any of the religious practices?

Perceptual Error and Psychology

Defeaters come in two kinds: those that show that the belief in question is false by providing evidence that supports a proposition inconsistent with it, and those that show that the evidence supporting the belief does not, in this case, actually support it. In the case of religious belief based on religious experience, the first kind of defeater would be, for example, independent reason to think that there is no God (for Judaism, Christianity, and Islam), or reason to think that a permanent self does exist (for Buddhism). Defeaters of the second sort would be reasons to think that religious experiences are not good evidence for religious belief, even though experiences in general do have that kind of force. Our subject in this inquiry is whether there are defeaters of the second sort. If we could show that religious believers are unreliable perceivers, that fact would be such a defeater.

The likelihood of perceptual error by self-deception, or self-delusion, or wishful thinking in the case of religious experience is precisely such evidence. As even the proponents of knowledge by religious experience admit, the kind of information presented in religious experiences tends to be meager, vague, and lacking in informational richness, and so is open to a wide range of interpretations. This interpretive latitude leaves a lot of room for the subject’s preferences, prejudices, and expectations to enter into the resultant beliefs. As J.J. Gibson put it,

[W]hat happens to perception when the information is inadequate? In general, the answer seems to be that the perceptual system hunts. It tries to find meaning, to make sense from what little information it can get. (Gibson 1966, 303)

The more inadequate the information, the more the perceiver has to supply, which allows an opportunity for all kinds of psychological biasing to introduce error.

One important way that perception can be biased is by wish-fulfilling self-deception, seeing what we want to see. Self-deception seems paradoxical, for reasons that lots of people have pointed out, but for our purposes, we don’t need to give an account of the phenomenon, so we don’t need to dissolve the paradox. If self-deception takes place, that is, if people form beliefs that in some sense they know are unfounded, that is enough. If we can then show that self-deception is not only possible in religious experience to a larger degree than other ways of forming beliefs, but even is prevalent, then we have shown that the subjects of religious experience are unreliable perceivers, and so their justification for the resultant beliefs is undermined.

Henry C. Triandis finds the telltale signs of self-deception in all kinds of features of religious belief and practice. Rather than looking at religious believers themselves and measuring whatever psychological propensities they have, he finds the signs of self-deception in the contents of religions themselves. Since the various religions have features that fulfill our wishes, if true, then they are probably wish-fulfillment self-deceptions. Some of those features are the following.

  1. 1.

    Gods are conceived of as resembling those who believe in them (Triandis 2009, 119);

  2. 2.

    Religions reinforce the powerful’s political position (Triandis 2009, 119–120);

  3. 3.

    Religions correlate with geographic areas (Triandis 2009, 122);

  4. 4.

    Religions have the same structure as magic, which is clearly wish-fulfillment (Triandis 2009, 126–131).Footnote 4

The claim seems to be (though it is not made explicit) that the best explanation for religions’ having these features is that they are self-deceptive fantasies. While this list may offer a persuasive cumulative case, when we look at each of the items on the list, their force evaporates. Item 3 is a version of the problem of religious diversity, and we have seen that the doxastic-practice approach has resources to answer it. Item 1 makes a good point against those religions and practices that have gods and think those gods have something like physical form; that is these days a very small subset of the range of human religions. Religions that assert that God is like human beings in more subtle and abstract ways are less vulnerable to this objection, though any personal being will have to be like us in most of those ways. As for item 2, it is not at all clear that it is even true. Religious belief certainly can have a role in reinforcing political power, but it frequently also has a role in undermining it. Item 4 claims that religion has “the same structure” as magic, but he offers us no account of what those structures are, so no reason to think such a similarity either holds, or would have any epistemic consequences if it did. But most important, Triandis is not really, or not effectively, arguing that religious belief is probably self-deception, and that therefore we have a defeater; rather, he is arguing that since religions are false, the best explanation for why people believe them is that they are wish-fulfilling fantasies.

The evidence of other justification-undermining kinds of psychological conditions is equivocal, but suggestive. Gibbons and De Jarnette (1972) found a correlation between religious experience and hypnotizability. This suggests that the real explanation for religious experience is some kind of suggestion, just as people’s sense perceptions can be manipulated by the right kind of preparation. There is apparently no correlation between religiosity and psychoticism. The literature on neuroticism is all over the map; some find no correlation, some a positive correlation, and some a negative correlation (see, for example, Hills 2004). This suggests either that there is a flaw in the design of some of the studies, or that there is some important difference between the variables in the different studies, so they are actually measuring different things. There is a large literature of studies that have found a correlation between religious experiences and a range of personality factors associated with openness to new experiences and ideas, including those that seem irrational to others (see Beit-Hallami and Argyle 1997, 91–92). One of those factors, and in fact one for which the correlation seems to be strongest, is schizotypy. Schizotypy is not itself a mental disorder, but people with schizotypy are more likely to develop schizophrenia. They are also more prone to hallucinate. On the positive side, they tend to be more creative than the general population. This collection of results, while far from conclusive, supports the hypothesis that religious experiences are attributable to imagination and suggestion rather than actual perception.

Still, this defeater does not work equally against all religious experiences. One thing that influences the likelihood of misperception or self-deception is how one is primed for the experience. Balcetis and Dunning (2006) found that visual perception can be influenced by what the perceiver wants, when the visual information is incomplete or ambiguous. Other studies show that motivation can influence how and on what attention is focused, which can determine what is seen.Footnote 5 If these effects show up in visual perception, how much more likely are they to be operative in religious experiences, where incomplete, vague, and ambiguous perceptual data is the norm? The mystical practices of Christianity, Judaism, and Islam involve putting oneself into a certain frame of mind by entering into prayer, or contemplating the nature of God, or the like. It is no surprise that such experiences culminate in beliefs concordant with those beliefs. Similarly, Hindu bhakti practices involve chanting the names of gods, and so it is no surprise that it is those gods who appear to the subject. Some kinds of Theravāda Buddhist meditation do not have that priming feature to the same degree, because there is no particular object that the meditation is directed to discover. Instead, they begin from a determination to calm the mind, and then simply observe, as neutrally as possible, the rising and falling of mental events, or contemplate the nature of the physical body, for example. Certainly, Buddhist meditators come to their practice with expectations, and they begin as primed as anyone else, but the practice of insight meditation is intended to weed those things out, not reinforce them. There is no presupposition in the practice itself as to what you will find.

Theravāda Buddhist literature on meditation emphasizes the need to observe whatever is there, without expectation. Even though there is some expectation as to what you will find in those observations, the technique itself does not involve focusing on or thinking about those things, and the achievement of final liberation is possible only when one sees the world as it is. The goal of insight (vipassanā) meditation is to pierce the illusion of the world as made up of stable objects, including a permanent self who is the witness and observer of that world. Paul Williams describes the goal this way: ‘Thus the meditator comes to deconstruct the apparent stability of things and to see directly the world as a process, a flow.’ (Williams 2000, 85) This effect is achieved by an examination of all elements of the experienced world to discern in them the ‘three marks’; every object of experience is unsatisfactory (dukkha), impermanent (anicca), and not-self (anatta). This examination is conducted by exercising mindfulness (sati), a prolonged and concentrated attention. In his discussion of the etymology and meaning of the word ‘sati,’ Anoloyo identifies as part of the importance of sati that it ‘is required … to fully take in the moment’ (Anoloyo 2003, 48).

Conclusion

Since justification admits of degrees, so does the undermining of justification. For that reason, it is impossible to say, in many cases, whether a defeater has completely removed justification or only weakened it. The same is true in this case. Some reasonable people may conclude that the prevalence of self-deception in religious matters, or among religious people, completely removes any justification for religious belief based on religious experience. But this seems too strong. After all, we could have reason to believe, of a particular subject, that she is less prone to self-deception than most. If a particular perceiver is intellectually responsible and a careful weigher of evidence, then it would make sense to conclude that the beliefs she forms on the basis of religious experience are reasonably grounded. So in the next chapter, I will turn to the question of religious testimony. Even if a particular subject is rational to form religious beliefs on the basis of religious experience, is it rational for a second person to believe on the basis of her testimony?