III. A Killer Argument?

Richard Gale asks whether religious experience is ‘cognitive’, as he puts it.⁴²⁷⁴²⁷   On the Nature and Existence of God (Cambridge: Cambridge University Press, 1991), pp. 285ff. In the remainder of this section, page references are to this work. What precisely does he mean? I think he means to ask whether religious experience is or could be part of a cognitive process that puts us in epistemic touch with God. The question is 336whether religious experience resembles sense experience in being part of a cognitive process that issues in knowledge of or warranted belief about an independent reality: God, for example. Gale argues that in fact religious experience is not cognitive, in this sense. He gives this argument in the course of addressing what he calls “the analogical argument for cognitivity,” which he associates with William Alston, Gary Gutting, Richard Swinburne, and William Wainwright. Their argument, he says, has two premises:

1. Religious experiences are analogous to sense experiences.

2. Sense experiences are cognitive.

Therefore:

3. Religious experiences are cognitive. (p. 288)

Gale directs most of his fire at the first premise.

Here we must issue a couple of caveats. First, strictly speaking, Gale is objecting to this argument by objecting to premise 1; so, strictly speaking, his conclusion would not be that religious experience or belief is not cognitive, but only that this particular argument for its cognitivity fails. For present purposes, this doesn’t matter: in fact Gale does much more than merely object to premise 1. What he really offers and what I want to consider is an argument for the conclusion that religious experience and belief are necessarily not cognitive. This argument, if successful, would show that it isn’t possible that we should have anything like a perceptual awareness of God. Second, he also seems to believe or assume that any experiential awareness of God would have to be or be like perceptual awareness of God (that any experience of God that was part of a cognitive process yielding knowledge of or warranted belief about God would have to play the same role, in that process, as perceptual experience plays in perception); he therefore concludes, I think, that it is not possible to have knowledge of God by way of experience—that is, that religious experience is not cognitive.

Now it isn’t entirely clear just how Gale’s argument bears on my argument for the conclusion that theistic and Christian belief can have warrant by way of the sensus divinitatis and the IIHS (that, given the truth of these beliefs, there are no cogent objections to their having warrant in that way, so that any objection to these beliefs will have to be to their truth rather than to their rationality or reasonability). First of all, Gale didn’t have my argument in mind, an omission that is entirely excusable, given that he proposed his argument well before I proposed mine. But second, the bearing of his argument on mine isn’t clear because it isn’t clear whether knowledge by way of the sensus divinitatis and the IIHS is properly thought of as knowledge by way of experience. It would be a shame, however, to pass up the chance to consider an argument as engaging as Gale’s; so suppose we 337assume, for purposes of argument, that if there is any such thing as knowledge by way of these processes, that knowledge is knowledge by way of experience. Then we can consider Gale’s argument as an argument against my conclusion. (Of course it could turn out that his argument is only dubiously relevant to mine precisely because it is dubious that knowledge by way of the sensus divinitatis and the IIHS, if there is any such knowledge, should be thought of as knowledge by experience.)

So how does the argument go? It begins with a playful threat directed at those who, as Gale thinks, accept the analogical argument for the cognitivity of religious experience:

We have yet to unearth a deep disanalogy between sense and religious experience that will totally destroy the analogical premise of its analogical argument. This “big disanalogy” will prove to be the shipwreck of this defense of cognitivity, a time for Alston, Gutting, Swinburne, and Wainwright to join their fellow analogical arguers on the deck for a few heart-felt choruses of “Nearer My God to Thee.” (pp. 326)

Brave words! Does Gale speak with the tongue of an angel, or is he only a clanging cymbal? Precisely what is this killer argument?

Necessarily, any cognitive perception is a veridical perception of an objective reality. It now will be argued that it is conceptually impossible for there to be a veridical perception of God . . . from which it follows by modus tollens that it is impossible that there be a cognitive religious experience. . . . A veridical sense perception must have an object that is able to exist when not actually perceived and be the common object of different sense perceptions. For this to be possible, the object must be housed in a space and time that includes both the object and perceiver. It then is shown that there is no religious experience analogue to this concept of objective existence, there being no analogous dimensions to space and time in which God, along with the perceiver, is housed and which can be invoked to make sense of God existing when not actually perceived and being the common object of different religious experiences. Because of this big disanalogy, God is categoreally unsuited to serve as the object of a veridical perception, whether sensory or nonsensory. (pp. 326–27)

Note first the implied claim in the second sentence: if it is conceptually impossible that there be a veridical perception of God, then it follows that it is impossible that there be “a cognitive religious experience”; Gale seems to believe that any variety of religious experience would be cognitive only if it were a part of a veridical perception of God. This seems wrong: as I argued in chapters 6 and 8, we can appropriately think of the sensus divinitatis and IIHS, on the extended A/C model, as providing knowledge of God that is knowledge by way of experience, but not perceptual knowledge. Consider the sensus 338divinitatis: you are in grave danger and form the belief that God is able to help; there needn’t be anything here we can sensibly refer to as perception. You suddenly realize that what you did was despicable; you form the belief that God disapproves, acknowledging to him that you did it; again, there need be nothing present that is properly called perception. On the model, there is knowledge of God here, and experience plays a crucial role—both the doxastic experience and also the experience that goes with feeling afraid or guilty or ashamed. This experience is closely associated with the operation of the sensus divinitatis, and perhaps triggers the production of the relevant belief. But the result is not, I should think, a perceptual belief. Furthermore, it is if anything clearer yet that knowledge of God and of the great things of the gospel by the IIHS is not perceptual knowledge (above, p. 288). If we don’t think of the sensus divinitatis and IIHS as issuing in perceptual knowledge, however, then Gale’s objection would be irrelevant.

For present purposes then, suppose we also temporarily concede that knowledge of God by way of sensus divinitatis or IIHS would be perceptual knowledge of God, at least in an appropriately analogical sense. Alternatively, suppose we consider Gale’s objection to perceptual knowledge of God, bracketing, for the moment, the question of its bearing on the extended A/C model. How does this objection go? “A veridical sense perception,” he says, “must have an object that is able to exist when not actually perceived and be the common object of different sense perceptions.” That seems right, or at any rate plausible; and if there is such a thing as experience of God, God will, of course, be able to exist when⁴²⁸⁴²⁸   At any rate, if God is indeed in time. And even if he isn’t, someone who says, at a given time when (per impossible) no one is experiencing God, that God exists, speaks the truth. See my “On Ockham’s Way Out,” Faith and Philosophy (July 1986). not being experienced by any human cognizer (even if in fact his presence is always experienced by some human being or other). Furthermore, God himself, the very same being, would be grasped, cognized, or apprehended by many different persons. So what exactly is supposed to be the problem with supposing that you and I are both aware of God, and that God continues to exist when neither of us is aware of him? The problem, he thinks, is that if an object can exist when unperceived, and can be perceived by different perceivers, then “the object must be housed in a space and time that includes both the object and perceiver.” Why so? Gale’s strongest argument here is contained in the following passages:

Another invidious consequence of their nondimensionality is that no analogous explanation can be given of how they can exist unperceived and be common objects of different perceptions to that 339which was previously given for empirical particulars. Whereas we could explain our failure to perceive an empirical particular, as well as our perceiving numerically one and the same empirical particular, in terms of our relationship to it in some nonempirical dimension, no such analogous explanation can be offered for our failure to perceive God and the like, or our perceiving numerically one and the same God. . . .

Similarly, how is Gutting going to decide when two religious experiences of a very powerful and loving nonhuman person had by two people at one time or by one person at two different times are of numerically one and the same being or only qualitatively similar ones? (pp. 341–42)

The alleged problem, therefore, seems to be twofold: it concerns (a) perceiving God at one time but not at another and (b) two different people’s both perceiving God—that is, perceiving the very same all-powerful, all-knowing, and loving nonhuman person. It isn’t as easy to see, however, just what Gale is claiming about (a) and (b). Two possibilities present themselves for (a) and three for (b). With respect to (a), Gale could be claiming that (1) if God is not in time and space, then no explanation could be given of our perceiving him at one time but not at another, and (2) if God is not in time and space, then it would not be possible that we experience him at one time but not at another. With respect to (b), the same two possibilities present themselves, and in addition there is the possibility that (3) if God is not in time and space, then we couldn’t decide (tell) when two people are both perceiving God—that is, both perceiving the very same divine person.

Well, consider (1) with respect to (a). Strictly speaking, I suppose, it wouldn’t damage the sensus divinitatis model or the claim that it is possible to perceive God if there were no explanation of our perceiving God at one time but not at another. All the model says is that such knowledge occurs; it doesn’t go on to add that there is an explanation of it, unless the model itself is an explanation. Still, I believe Gale is really claiming, in this first possibility, that we can’t see any way in which it could happen that we should perceive God at one time but not at another. More strongly, I think he means to claim that we can see that a person could perceive God at one time and not at another only if he and we were in the same time and space. We can see that this is the only condition under which such a thing could happen. So really, (1) and (2) with respect to (a) come to the same claim: it would be possible for us to perceive or experience God at one time but not at another only if he were in the same space and time (spacetime) as we.

So says Gale; is he right? I think not. Even with respect to things that are in space and time, there are a wide variety of explanations for my perceiving a thing at one time and not at another. I might perceive the thing at t₁, but then at t₂ have my eyes closed, or be asleep, 340or thinking of something else, or have lost my glasses, or have a brown paper bag over my head, or be under water, or be suffering from cognitive malfunction. Similarly for perceiving or cognizing God. God is, of course, always existent; furthermore, because (as we may assume for purposes of argument) he isn’t in space, he is never related to me spatially in different ways at different times. Still, why should that mean that I couldn’t perceive him or experience him at one time and not at another? I might perceive or experience him at t₁ but not at t₂ because at the latter I am asleep, or my attention might be elsewhere (I have just hit my thumb with a hammer or shot myself in the foot), or I might be suffering from cognitive malfunction, or I might be angry with God because of my friend’s suffering and thus not in the right frame of mind. There are plenty of ways in which I might perceive him at one time but not at another, even if it’s not possible that I be spatially related to him in different ways at different times.

Even if I am wide awake and eager to feel his presence, eager to receive an answer to prayer, or guidance, or a sense of his love, or a perception of his beauty and grace, I might not get what I am hoping for. This might be because at that time, and for reasons of his own, God doesn’t propose to communicate with me in those ways. Even if I am properly related to you both spatially and temporally and ask you a question, I still might not get an answer. I ask you why you are smiling in that enigmatic way: you make no reply, perhaps because you think the question is an impertinence, or not worth answering, or because answering would interfere with your desire to remain enigmatic. When it comes to knowing and knowing about other persons, their cooperation is often required. Naturally enough, the same would be true of God, and perhaps on many or most occasions he chooses not to be perceived. As I argued in chapter 6, it isn’t easy to say just what the necessary and sufficient conditions for perception are; whatever they are, however, they involve a certain sort of experience, an experience in which the perceived object seems to be present to or given to the percipient. Clearly, this experience could be absent at one time and present at another, whether or not God is in space.

This argument, then—the argument for the claim that if God is not in space and time, then we couldn’t experience or perceive him at one time but not at another—seems to me to be entirely without promise. But what about Gale’s claims with respect to (b)? Here he makes a double assertion: if God is not in time and space, then there could be no explanation of two people’s both perceiving or experiencing him (we can’t conceive of any way in which this could happen), and furthermore there couldn’t be any way of telling that it was God, the very same person, that I experienced on two different occasions. Is there any reason to accept either of the suggestions? How could it 341happen that two different people should both perceive or experience God? Suppose both had the right kind of experience, including that sense of God’s being present, being given. Suppose, furthermore, both formed the right kind of true belief about God—for example, that he is indeed present, given to them. And suppose, finally, the conditions of warrant are met: this belief is produced in them by cognitive faculties functioning properly in a congenial epistemic environment according to a design plan successfully aimed at truth. Then they would both be perceiving God, despite the fact that God is not in space. Is there really a problem here? If so, it is well concealed. Gale or someone else might claim that there is a problem with the very idea of our forming beliefs about God: what would make it God the belief was about? But this is a wholly different kind of objection and is really the objection dealt with in chapters 1 and 2; it is as inconclusive here as it was there.

Similarly for the other part of the suggestion: that it would be impossible to tell whether it was the same divine person one experienced on two different occasions. Because God is not in space, so the claim goes, we could never tell whether we encountered the same divine being on occasion B that we had encountered on occasion A. But is there any reason for thinking this true? This argument, I think, is as fatally flawed as the last. The idea seems to be that with objects not in space, one can’t, in principle, tell when one has encountered the same one again, or only encountered another thing that is appropriately like the first thing. But is this in fact true? What exactly is the problem? Suppose your experience is related to God in the right way and you form beliefs about him on two subsequent occasions, t₁ and t₂; at t₂ you form the true belief that the being about whom you are then forming a belief is the same being about whom you formed a belief at t₁: if this belief is formed under warrant-conferring conditions, then you would have told that the being about whom you formed a belief at t₂ is the very same being as the one about whom you formed the belief at t₁. Indeed, that’s just what it is to tell that the t₂ belief is about the same being as the t₁ belief. Well, perhaps the claim is that one can never tell that a t₂ belief is about the same being as a t₁ belief, if both beliefs are about a nonspatial object. This also seems quite wrong. At t₁ I think of the null set, or the proposition that all men are mortal; I then read Gale’s book or watch a football game on TV, and at t₂ think again about the null set (or that proposition). Is there really a problem with my telling or knowing that it is the same thing I think about at t₂ as at t₁? Should I be in doubt as to whether the set I think of at t₂ really is the same set as that I thought of at t₁?

Could it be that Gale’s suggestion is that if a dispute arose over whether what I experience on a given occasion is the same being as you experience on a given occasion, the dispute might turn out to be 342intractable? Alternatively, might I not (like Teresa of Avila and others) be uncertain, on a given occasion, whether it was God I was in contact with, as opposed to Satan, who appears as an angel of light and is out to deceive me? Couldn’t this happen? Of course it could. But it doesn’t show I could never tell that I am experiencing God; what it shows is only that perhaps on those occasions I can’t tell whether it is God with whom I am in contact. True: it is possible that appearances be just what they are and I not experience God. As we have already seen, however, nothing follows from this: the same is true with respect to perception, memory, and my knowledge that I have existed for a substantial length of time. It is clearly possible that I have the experience that goes with perceiving a horse and yet no horse is there.⁴²⁹⁴²⁹   But aren’t there checks and tests for telling whether you really perceive a horse? Indeed there are; but consider the experiences that go with checking to see whether you perceive a horse and determining that you did see a horse: it is also logically possible that you have all those experiences when, in fact, there is no horse there. So it could be that on some occasion I really can’t tell: still, this hardly shows that I can never tell. And once more: the way I tell whether it is God, that very person, whom I perceive or experience on two different occasions, is by forming (under warrant-conferring conditions) the true belief that indeed I encountered God on both occasions.

I think it is clear, therefore, that these arguments don’t even begin to show that perception of God is impossible, or that religious experience is never cognitive, or that there couldn’t be knowledge of God by way of the sensus divinitatis and IIHS. Gale’s arguments depend upon a lot of assumptions that have little or no claim to assent: that if you sometimes can’t tell whether p, for example, then you can never tell whether p, or that if God is not in space and time, then you could never tell that it was he with whom you were in contact on successive occasions; or that if God is not in space and time, it couldn’t be that you should experience him on one occasion and not on another. All of these assumptions seem monumentally dubious at best.⁴³⁰⁴³⁰   And, of course, there remains the point that the extended A/C model does not, strictly speaking, require that one perceive God at all.