Arguments from Experience

Over at Prosblogion, they are having a discussion of the argument from religious experience. One of the comments says the following:

I’ve always thought that the argument from religious experience had a certain strength to it. In order for it to be cogent, only one of the billions of religious experiences needs to be veridical. In order for the argument to be proven false, every one of the billions of religious experiences must be false. With these large numbers, and the remarkable burden of proof that rests on those who wish to reject the argument, it seems that the argument from religious experience has a certain initial plausibility to it.

This point is surely mistaken.

The truth of the conclusion would be guaranteed by the veridicality of the experience, but that doesn’t make the argument cogent. It’s like one of Plantinga’s arguments for the existence of God: look how many sets there are; there couldn’t be that many sets without a God; so God exists. Even if the argument is sound, it should still be found wanting from an epistemic point of view. (Even though it is a hard question to say what is needed of an argument so that it is not deficient from this point of view).

This discussion raises an interesting question in epistemology about the epistemic value of experience and the role of testimony. Suppose we think of it in a non-religious context. I’m thinking of the following kinds of cases: empirical disputes, political disputes, moral disputes, etc. There are two, er, bizarre views here: the first is that unanimity of testimony is required for rational acceptance of the content of the testimony, and the other is the position in the quote above that no amount of disagreement matters.

How about this viewpoint? Testimony creates a presumption in its favor that requires defeating information to block rational acceptance, and the defeating information must itself be something more than that there is contrary testimony. What else? I can’t say, but I can list some things: some sketch of an explanation of why the testifier is mistaken or unreliable; the existence of contrary testimony plus a claim to the effect that there are no rational grounds on which to prefer one piece of testimony to another; etc. Central to the position is that testimony is a generative source of justification (or rationality or (non-Plantingian) warrant). Many here have worked on testimony much more than I, so I ask: does this position withstand scrutiny (or better: is there is precisified version of it that withstands scrutiny)?


Arguments from Experience — 19 Comments

  1. It can’t be right that merely contrary testimony isn’t relevant here. If testimony creates a presumption in favor of the view to which it testifies, then contrary testimony creates a presumption in favor of the negation of the claim, which seems to be sufficient to cast doubt on the value of the first instance of testimony. Of course, the numbers and the relative epistemic position of the testifiers is going to be relevant here.

    In any case, I would have thought that what people who want to deny the value of a particular instance of (sincere) testimony owe us is, inter alia, an explantion of the experience testified to which does as good a job of explaining it as the rival claim. In the religious case, there are plenty of prima facie plausible explantions of the experiences invoked as evidence for religious claims (from stimulation of the temporal lobe to cultural expectations).

  2. Neil, that’s right, though by the phrase “the existence of contrary testimony” I meant only that someone is saying something contrary to someone, not that the subject of the example has been told the contrary testimony. I’m assuming that the subject knows that there are people out there who think the opposite of the received testimony, and also would testify to the opposite if asked. If the contrary testimony has been received as well, then we have counterbalancing in the absence of further information (whichm, in most cases there will be–one testifier strikes one as more honest and reliable, etc.). But the bottom line is that I was thinking of cases where one has received testimony that p, and the contrary testimony has not been received but merely exists.

    There’s another intermediate case. You’ve been told that p in circumstances where coming to believe p is thereby reasonable. So you believe it. You act on the belief, use it in constructing explanations of other things, etc. Then you come to receive contrary testimony. The timing issue, and the intervening events, have to bar the contrary testimony from having the same epistemic power it would have had, had one received it at the same time as the original testimony, I would think.

  3. Jon, I don’t think I agree with the idea that receiving the contrary testimony (as opposed to learning of its existence) makes a difference. Suppose that A has told me that p, and that I know that B would tell me that ~p, but B has not actually told me. If I don’t have a reason to favor A over B, doesn’t the fact that I know what B would say cancel my justification for believing A?

    Now, there’s this: If I am confident that someone would tell me that ~p, that doesn’t cancel A’s testimony. Even if 99% of all people would tell me that p, it is likely that there will be someone who would tell me that ~p–the fact that there is at least one person who would tell me ~p doesn’t outweigh the testimony that p I have received from one specific person. Is that like what you’re thinking of?

  4. Matt, I think it depends on what you know about B, but I think you recognize that since you require having no reason to favor A over B. In that case, I agree with you. But just having a particular instantiation for the claim that someone disagrees isn’t enough–you have to know something further about the instantiation in order for the disagreement to undermine the testimony of A. From what you say I think you agree with that.

    And I agree with the last paragraph, but I’m inclined to run a slippery slope argument on it to conclude that there isn’t a percentage of agreement that is required for justification to be conferred. I’m inclined to think that there’s something going on in cases of testimony that is not simply a matter of agreement or disagreement, so that both agreement and disagreement get overrated. Maybe this commits me to the view that testimony is not a basic source of warrant, but I can’t tell for sure.

  5. Jon, you raise a good point about my comment at the Prosblogion. I definitely was thinking like a metaphysician, not an epistemologist. This is not to say that I haven’t given this any epistemic thought, though.

    Some recent work has been done that fits eyewitness testimony into a Bayesian format. If you are fond of Bayes’s theorem, it is possible to give a precise account of the value of eyewitness testimony. There are essentially two variables in this Bayesian assessment: (1) the prior likelihood of this event occurring (not counting testimony) and (2) the likelihood of that the testimony is true than false. The value given to (1) will depend on the evidence one has for the event. If the event is highly implausible for scant evidence, etc., then it will be difficult for testimony alone to give justification to assent one to belief. However, as long as the value for (2) grants that it is more likely that a fair witness would give the testimony if it is true rather than false, the Bayesian format allows that given enough multiple independent eyewitness testimony, the event can be epistemically confirmed. (I’m giving a paper on this at the midwest SCP meeting, if anyone is interested.)

    For some more reputable sources on this matter, I would recommend:
    Rodney D. Holder, �Hume on Miracles: Bayesian Interpretation, Multiple Testimony, and the Existence of God,� British Journal for the Philosophy of Science 49 (1998): 49-65.
    John Earman, Hume�s Abject Failure: The Argument Against Miracles (New York: Oxford University Press, 2000).

  6. Jon, I’m not sure whether this helps, but it seems to me that a further factor when evaluating the cogency of a piece of testimony would have to be its relation to a set of relevant background assumptions pertaining to the issue addressed. And maybe what differentiates defeating information from mere contrary testimony is that while the latter is not always properly related to the background assumptions of the epistemic community in question, the former is. By way of example, I take it that the reason people in secular societies do not in general consider religious testimonies to be cogent, is that the kind of background story these testimonies presuppose does in some central assumptions not mesh with what secular people believe about the world. (Hence, one might be tempted to try to make religious testimonies more “scientific”.) What do you think?

  7. Isn’t the problem of religious experience the problem that the interpretation (or even nature) of the experience is so theory laden? It becomes an issue of how to separate out the assertion from the theory or theories it is contained within. Now I think that within science we’re able to do that do some degree. Even if there are still problems inherent to it, we can translate claims from say old astronomy to modern astronomy. However with respect to religion it seems like we ignore the whole notion of theory and discount all testimony. I’m not entirely sure that is fair.

    Put more explicitly, consider a Evangelical and a Hindu having a very similar experience of God. How they will describe the experience will differ. But that doesn’t mean that the experience can’t be evidence, any more than the same thing in science.

    Now there may well be many other problems. But it seems to me that most discussions never get that far.

  8. Excellent comments here, and I’ll try to say something about each of them. First, on Clark’s point. I’m suspicious of claims to the effect that religious experience can be sorted into the theory-laden part and the remainder. When I try to report my color experience the way artists are said to see the world, so that I attend very carefully to the phenomenology of the experience, I end up having a different experience rather than clarifying some non-theory-laden core of the original experience. I think the same thing happens when we try to sort religious experience into the phenomenological part and the theory-laden part. I think a better approach is to focus on what is experienced versus inferred, identifying what is experienced with the taking rather than the inferring, and then the theory-ladenness issue disappears, I think.

    John and Kristoffer, I take it that the heart of the Bayesian account is going to make the evidential value of experience depend centrally on background information. With that I wholeheartedly agree. Bayesianism is to epistemology what Brazil is to soccer: the beautiful game! My only reservation is that I think rationality is not a community matter, but an individual one, in the following sense. It’s your own background information that matters, not that of your community (though it is presumably true that most of your background information has that status because of the community to which one belongs).

  9. Jon, from an epistemological point of view, I’m not sure we can easily sort them. Thus the distinction with theory ladenness in science. The way science gets at the common truth is by repeating experiments. In religious experiments that’s harder.

    But my point is simply to point out a neoKantian view that the experience will be given in terms of our categories of thought. That means that getting at the pure experience is impossible. But that means that assuming there is no real experience is also impossible. Now I’m not a neoKantian, so I’ll say it’s just real hard, beyond the potential of doing what scientists do and repeat the experiment. The alternative is to also do what scientists do and try to “demythologize” texts. However that’s a very subjective approach. And perhaps not to be trusted too much.

  10. Clark, I see; I misunderstood what you were suggesting. I agree with the general point that we can’t get out of the theory-ladenness problem, and that we should figure out how to move on without some way of isolating what is given from what is brought to an experience. A suitably subjective theory of justification will be able to do that!

  11. I have recently published an article in Philo, â??A Defeater of the Claim that Belief in Godâ??s Existence is Properly Basic,â?? Philo 7: 57-70 that addresses the issue of theory laden perception and religious experience. The main point about disentangling the subjective content of perception from the objective part that is being adressed by the discussion seems to be correct. From the perspective of content individuation of perceptual states it seems that we are entitled to make existential claims only if we have some robust account of the perceptual mechanism at work that is evidentially sufficient for ruling out the possibility that what is alleged to be perceived is just narrowly individuated by reference to some concepts or background knowledge. This sort of existential appeal based on experience then seems to require an externalist approach to content individualtion of perceptual states. So, due to the theory ladenness issue, what is required to accept an existential claim based on experience is a (broadly) causal account of how such a perceptual state arose in virtue of being caused by the alleged object of that perception. As I see it this seriously undermines religious appeals of this sort (like those of Plantinga, Alston, et al.) as we have no causal account of how such entities cause perceptual states and so they cannot be individuated from circumstantially identical perceptions that do not have such content.

  12. Michael, you say,

    From the perspective of content individuation of perceptual states it seems that we are entitled to make existential claims only if we have some robust account of the perceptual mechanism at work that is evidentially sufficient for ruling out the possibility that what is alleged to be perceived is just narrowly individuated by reference to some concepts or background knowledge.

    I think this claim radically overestimates what is required of perceptual knowledge and perceptual entitlement. Small children and animals are entitled to make existential claims based on what they see, and they have no account of the the perceptual mechanism at work.

  13. Jon,

    I agree that an individual need not be necessarily aware of the details of the causal account of the faculty, but unless there is some in principle possibility of giving such an account then there is no way to rule out the possibility that such perceptions are just subjective interpretations of ordinary perceptions and, hence they are not adequately justified. This constitutes a serious defeater for the existential claim in the case of God because in the case of mystical religious expereince we have little or no hope, in principle or in actuality, of explaining how non-physical entities enter causally into perceptual expereince. So, the case of a child seeing a tree is appreciably different from that of, say, Plantinga seeing God when he sees the same tree. The child is justified in her belief that there is a tree not just because she claims to see a tree on the basis of a tree expereince, but in part because her tree seeing expereince can be causally explained to a sufficient degree. Consider the historical analogy with accepting beliefs based on the use of a microscope (or the telescope a la Galileo). It certainly was a matter of great controversy whether microscopes really revealed the existence of microscopic entities and much of the support for such existential claims took the form of optical explanations of how the microscope works. It was, nevertheless, a legitimate question whether they were allowing us to perceive ordinarily undetectable entities or whether these were just subjective features of the users and this skeptical response was quashed by offering the causal account of how a microscope works. The same principled requirement, I take it, applies to how the human eye works. First, there does not seem to be any special prima facie entitlement associated with the output of the human eye qua the existential import of those experiences, and , second, we need to be especially careful about this requirement in the context of “special” sensory acuities like that which the “reformed” epistemologists claim to possess.

  14. Michael, I still think your demands are too high. I don’t see why we should need to be able to construct a theory of the sort you describe, except when faced with defeaters for the warrant provided by our senses. I take it from your last three lines that you don’t think there is such, but that looks obviously false. That’s just what the evidence of the senses provides.

    The theory your describing might be weaker–it might be only that there has to be such a true theory about how the belief-forming practice is reliable. I think that’s mistaken, on evil demon grounds, but even if we grant that point here, that’s all Plantinga claims and that’s all he needs for his position. His claim is that if his religious beliefs are for the most part true, then they are warranted (because none of the other conditions for warrant are going to cause problems); moreover, he thinks, there’s no difference on this score with perceptual beliefs.

  15. Jon,

    Why should we believe that the products of our senses are only in need of explanation when faced with a defeater? On what basis is such an implausible a priori entitlement founded and does this not beg the question against the skeptic? I suspect that if Plantinga’s position is as weak as you suppose and it has the conditional form that “if his religious beliefs are mostly true, then they are warranted”, then his position cannot underwrite the unqaulified claim that the objects of religious perceptions (i.e. religious entites) exist. This seems obvious, becuase his religious beliefs may be false and therby unwarranted and so belief in such objects may be the result of theistic wishful thinking that accompanies things like looking at trees. I suppose that I am rather more inclined to be skeptical of perceptual claims than yourself.

  16. Michael, well, that’s what it is for the senses to prima facie justify beliefs. And if you ask for a basis for this, I guess the answer may depend on exactly what question you’re asking, but here’s one answer to at least one possible question: there are epistemic norms or principles, one of which ties the contents of experiences to belief states.

  17. Shaffer’s requirement that a theory of perception is required before one has warrant in forming beliefs on the basis of perception has problems in addition to the obvious point, made by Jon, that the requirement is counterintutively over-intellectual.

    If Shaffer is right then prior to accepting our current theory of optics and how the human eye works, we had no good reason to believe any empirical claim, including theories about optics and how the human eye works. Thus the empricial theories that underwrite our reliance on perception could only have been adopted by an act of faith. And how could a theory that we adopted without any good evidence provide a basis for our reliance on anything? Or if it can, then why can’t the religious epistemologist just cook up a theory that connects God to our minds and adopt it without any justification?

    I take it that those who think there is an a priori entitlement to rely on perception do so because they see this as the only way to avoid skepticism. I think that Shaffer is right to say that this begs the question against the skeptic but requiring that we have a theory of perception will also result in circularity (or skepticism) and so it doesn’t help at all.

  18. I think that the main point of difference here, that between “religious” perception and ordinary perception, is crucial and raises the question of defeat with respect to religious perception when faced with the fact that such perceptual faculties cannot be reliably differentiated from those that are ordinary. The fact that non-normal perceptual abilities are being suggested and employed in driving home the existential claim makes a huge difference. Even if one were comfortable with (say a la Peacocke) claiming that native perceptual abilities carry some a priori entitlement (which is question begging) this surely is not true in cases where we are dealing with “funky” perception. Our standards should be very high in such contexts. Otherwise, we should all simply accept that some loonies can see auras, etc. and so they are prima facie justified in believing this. By the way, this is also crucial because the refomred epistemologists also want to claim that those who do not share such abilities should ALSO accept the relevant religious beliefs as warranted on the basis of the perceptual states. This, I assume, is far too permissive.

  19. I’m not sure which reformed epistemologists you are thinking about. If Plantinga, your claim isn’t quite right. What is right for Plantinga is this: if theistic beliefs are true, then they are warranted; and Plantinga argues that this (non-religious) conclusion should be accepted by everyone. But the conditional is critical here for Plantinga’s view (though, of course, he also thinks the antecedent is true; he just doesn’t think there is an argument or anything else that compels rational assent to the antecedent).

    On the distinction between ordinary (or normal) perceptual abilities and funky ones, I can’t think of any way to draw the distinction except between those that are shared and those that aren’t. That distinction isn’t going to do much work in fundamental epistemology, I think. After all, the mere distinction by itself is irrelevant, so what would have to do the work is our awareness of the distinction. And we have to rely on our abilities to gather that information, abilities which are going to need to generate prima facie justification on pain of skepticism. And then the right model is, at most, to treat the shared/not shared distinction the same way we treat the distinction between agreement/disagreement in general.

Leave a Reply

Your email address will not be published. Required fields are marked *