H. A Narrow Original Position?

In any event, there would be something very peculiar about supposing that the original position includes the beliefs I actually have about the reliability of SP and CMP (as well as the beliefs I actually form on the basis of those practices). The whole question of the rationality or sensibleness of CMP and SP arises, after all, because we realize we can’t successfully argue that those sources of belief are reliable. (It is this realization that precipitates the “crisis of rationality.”) We need a term for those beliefs which are such that we can’t successfully argue that the sources that produce them are reliable: say that such beliefs are uncredentialed. Then the crisis of rationality, with respect to SP and CMP beliefs, arises because we realize that they are uncredentialed. What to do? Alston suggests that at any rate we can argue that it is practically rational to form belief in the CMP and SP ways. That will give us something, even if it is settling for second best. So the idea is to show that there is something rational or reasonable about beliefs formed the SP and CMP ways, by showing that it would be rational to choose to form beliefs that way in the original position. But then presumably there would be something at best very peculiar about relying on the belief that CMP and SP are reliable; that belief itself, of course, is uncredentialed.

And, in fact, Alston’s idea, in Perceiving God, is that these beliefs are not to be included in the original position:

But I was also thinking of this subject [the person in the original position] as realizing that s/he is unable to show that any of these practices are reliable, and believing that this implies that s/he is unable to use beliefs in that reliability, or beliefs that presuppose that reliability, to determine the most rational course to take vis-à-vis belief formation.¹⁴⁰¹⁴⁰   Alston’s reply to comments on Perceiving God at a meeting of the Society of Christian Philosophers (concurrent with the meeting of the Eastern Division of the American Philosophical Association) in Atlanta, December, 1993 (a published version can be found in The Journal of Philosophical Research 20 (1995), pp. 67ff.

The suggestion is, I think, that in the original position we bracket our confidence in the practices in question; better, we simply don’t have any beliefs of this sort in that position. In making this decision, we languish (or flourish) behind a veil of ignorance. This is a narrow original position. We are to engage in the following thought experiment: try to see what it would be rational to do if you didn’t already believe in the reliability of SP or CMP, knew that there are no good noncircular arguments for their reliability, and (correctly) believed that it is up to you whether you engage in those practices: under those conditions, would it be rational to continue in forming beliefs the SP or CMP way? The idea here seems to be that the original position wouldn’t include the belief that SP or CMP is reliable, or even any of beliefs formed on the basis of SP and CMP; for presumably those beliefs presuppose the reliability of SP and CMP (at least if I understand what Alston means here by ‘presuppose’).¹⁴¹¹⁴¹   Another reason for supposing SP and CMP beliefs are not to be included in the narrow original position: if they were, then, of course, in that position we would think them true, and we would know that they were delivered by SP and CMP. But then, obviously, we would have excellent reason to think SP and CMP are reliable, and it would be obvious that the rational thing to do would be to continue forming beliefs the SP and CMP way; we should be back at the previous condition of triviality.

Well then, what beliefs are included in this narrow original position? Which of my beliefs could I sensibly use, in coming to a decision as to whether to continue with CMP, SP, or both? Alston holds, of course, that I can properly use the premises of his arguments for the practical rationality of CMP and SP; these are included in the original position. As you recall, the premises of the first argument include something like

(1) SP and CMP do not lead to massive inconsistencies; there is no reason to think them unreliable; we know of no alternative doxastic practices whose reliability we could demonstrate in an epistemically noncircular fashion; and it would be disruptive to stop forming belief in these ways;

the premises of the second include

(2) SP and CMP are socially established practices that are not demonstrably unreliable or otherwise disqualified; and it would be disruptive to stop forming belief in these ways.

Here (1) would be teamed with another premise according to which it is practically rational to decide to continue a practice that meets the conditions (1) says SP and CMP meet; there would be a similar premise to go with (2). And Alston’s idea is that at any rate we can use these premises in coming to a decision as to whether to continue in SP and CMP. So even though the original position is narrow, it would still include the premises of his arguments.

But why so? Why would it be appropriate to rely on these premises, in the original position? The problem with relying on the beliefs that CMP and SP are reliable and on the beliefs that are formed by way of CMP and SP, of course, is that it is these very beliefs that are uncredentialed. Doesn’t the same go, however, for the doxastic practices that yield (1) and (2)? Can we do any better with respect to them? For example, both (1) and (2) include the belief that we have been engaging in CMP and SP, forming beliefs in the CMP and SP ways. But how do I know that we have been doing this for some time? Presumably, it is only by way, in part, of perception itself: I perceive other people (or, to be really finicky, their bodies), and that perception is necessary to my knowledge that they use CMP, SP, or both. But perceptual beliefs are also uncredentialed: so these beliefs of mine are uncredentialed.¹⁴²¹⁴²   Does the fact that the argument takes as a premise a belief that is a deliverance of the very practice under consideration show that the argument is epistemically circular? Not obviously: the conclusion of this argument is not that SP is reliable, but that it is practically rational to engage in it. And how do I know that we have been doing this for some time? Presumably by way of memory. 128Memory beliefs too, sadly enough, are uncredentialed; there is no way, as far as I can see anyway, in which one can show, in an epistemically noncircular way, that memory is reliable. And how do we know that it would be disruptive to stop forming beliefs in these ways? Presumably on the basis of our general knowledge of human beings and human nature, at least part of which comes by way of perception. And how do I know the truth of those additional premises, according to which it would be practically rational to continue with doxastic practices that meet the conditions laid down in (1) and (2)? Here presumably the idea is that these beliefs are self-evident, or obvious, or at any rate have a good deal of intuitive support. Therefore these premises would be among the deliverances of reason. Therefore they too are uncredentialed; we can’t give an epistemically noncircular argument for the reliability of reason, for in giving such an argument, obviously enough, we would be obliged to rely upon reason. So (1) and (2) are no better off than the beliefs that SP and CMP are reliable: if the latter can’t properly be used in the original position because they are uncredentialed, then the same is true of the former.

Indeed, as Alston himself points out (146–47), it is easy to see that none of our beliefs is credentialed. Even if we could give an argument to show that a given source of belief was, in fact, reliable, in making that argument we would be obliged to rely on other sources of beliefs. In particular, we would have to rely on reason; but clearly we can’t establish that reason is reliable without relying on reason itself; so beliefs that are produced by reason are uncredentialed. Hence, if we insist that the original position must include only credentialed beliefs, it won’t include any beliefs at all. And if it doesn’t contain any beliefs at all, then in the original position you wouldn’t have the faintest idea what to do, whether to continue with SP and CMP or not. You might as well flip a coin; more likely, the rational thing to do would be to withhold judgment altogether. But why would that matter with respect to the question of the rationality of forming beliefs in the SP or CMP way? Obviously, if you had no beliefs to go on, you couldn’t come to a sensible decision as to whether to continue with SP or CMP; why would that fact show that there is something irrational in forming belief in accord with SP and CMP? If you had no beliefs at all on the subject, you couldn’t come to a sensible decision as to whether to continue with SP and CMP: but that fact is quite irrelevant to the question whether there is something wrong with forming beliefs in the SP and CMP ways. If so, however, the de jure question would not be the question whether it would be rational to continue with CMP (or SP) if I were in the narrow original position.¹⁴³¹⁴³   Do I know, however (in the narrow original position), that I have been forming beliefs all along in the SP and CMP way, and that it would therefore be inconvenient to change my way of forming beliefs, whether or not other people are involved? I think this leads to a puzzle, illustrating the limitations of this kind of counterfactual thought experiment. I am to imagine myself in the narrow original position, one in which I don’t have any SP and CMP beliefs; but then, of course, I would have a different way of forming beliefs from the way in which I actually do form them. If I were in that position, it would not be true that if I were not to employ SP and CMP, then I would be changing my ways of forming beliefs; for in the narrow original position I don’t form beliefs in those ways! What this shows, I think, is that this counterfactual way of trying to get at the de jure question, either about SP or about CMP, suffers from substantial limitations. For example, perhaps you endorse conservatism: all else being equal, you say, the sensible thing to do is to continue with the way you’ve been doing things. But in the narrow original position, the conservative thing would be to continue in the agnosticism that is part of that position; so if, in that position, you accept conservatism, then the rational thing to do would be to remain agnostic.

Should we perhaps consider a different possible narrow original position for SP and CMP? As for the first, take the original position to include the standard package minus perception: reason, memory, and introspection, the faculty (or means) whereby we know what our experience is (for example, how we are appeared to). Of course it would include only part of memory: in the original position thus conceived, I wouldn’t have any memory belief that depends upon perceptual belief. (For example, I wouldn’t have the memory belief that I saw a cat yesterday, but only the belief that it seems to me that I saw a cat.) What I would have to go on, therefore, would be just introspection, reason, and some fragment of memory. Then the original position with respect to SP includes (1) my knowing that it is within my power to form beliefs in the SP way and also within my power to withhold SP beliefs, (2) my knowing that it is not possible to give a good noncircular argument for the reliability of SP, (3) my having no views as to the reliability or unreliability of this practice, and (4) my not having SP beliefs or beliefs dependent on perceptual beliefs. My aim or purpose, of course, is to believe truth and avoid error. And now the question is: what would it be rational for me to do, if in fact I were in that position? Decide to continue to form beliefs the SP way? Or reject them?

Again, however, the real question, it seems to me, is this: why is that question relevant? That is, why would the answer to the question what it would be rational to do in that position have anything to do with whether it is rational, or whatever, for me to form beliefs in the SP way in the position I am actually in? I doubt that anything epistemically interesting hangs on the answer to it. What we have left isn’t much to go on, and I really can’t see where the probabilities would lie.¹⁴⁴¹⁴⁴   Another possibility is that in the narrow original position with respect to SP, I continue to form beliefs the CMP way (so that the narrow position with respect to SP includes the beliefs I actually form on the basis of CMP). In that case, I think the probabilities would be with SP, at least if one thing I know in that position is that I have an enormously powerful tendency or natural inclination to form beliefs the SP way. God, as Descartes insisted, is no deceiver. Well, suppose the answer is that those probabilities lie with agnosticism. All things considered, from the perspective of the narrow original position with respect to SP, it looks as if the course most likely to produce the most favorable position with respect to the truth is agnosticism about the deliverances 130of SP; the rational thing to do would be to withhold these beliefs. How would that be relevant to the question whether it is in fact (in the situation in which in fact I find myself) rational, in some interesting sense of ‘rational’, to form belief the SP way? If we decide this question by asking whether it would be practically rational to do so in this narrow original position, we are entirely ignoring perception as a source of warrant. We are treating it as if it had no authority or credentials of its own, even with respect to the very area to which it seems to be addressed. We are treating it in the way Thomas Reid thinks Hume treats it.

As Reid also asks, however, why should I trust reason (and that smidgin of memory) more than SP?¹⁴⁵¹⁴⁵   “The sceptic asks me, Why do you believe the existence of the external object which you perceive? This belief, sir, is none of my manufacture; it came from the mint of Nature; it bears her image and superscription; and, if it is not right, the fault is not mine; I ever took it upon trust, and without suspicion. Reason, says the sceptic, is the only judge of truth, and you ought to throw off every opinion and every belief that is not grounded on reason. Why, sir, should I believe the faculty of reason more than that of perception? They came both out of the same shop, and were made by the same artist; and if he puts one piece of false ware into my hands, what should hinder him from putting another?” (An Inquiry into the Human Mind, in Thomas Reid’s Inquiry and Essays, ed. Ronald Beanblossom and Keith Lehrer [Indianapolis: Hackett Publishing, 1983], pp. 84–85). Why should SP have to prove itself before the bar of reason?¹⁴⁶¹⁴⁶   See WCD, pp. 97ff. Of course I don’t mean to suggest that Alston thinks SP does have to prove itself before the bar of reason; I am exploring various answers to the question ‘what do we know or believe in the original position?’ To descend from the level of metaphor: why is it rational (in the relevant sense of ‘rational’, whatever precisely that is) for me to form belief in the SP way only if it is more likely than not from the perspective just of reason, that fragment of memory, and introspection that SP is reliable? Perhaps, from that impoverished point of view, it is not more likely than not that SP is reliable; does that show anything of interest? I doubt it. Suppose our battery of ways of forming beliefs, our belief-forming faculties, are in fact reliable; suppose, indeed, that we have been created by God, who intended that we be able to know the sorts of things we think we know by virtue of just such a battery of faculties: reason, memory, sense perception, introspection, sympathy, the sensus divinitatis and the internal instigation of the Holy Spirit (see below, chapter 8), if there are such sources of belief, and all the rest. What reason is there to think that if these faculties are reliable, then it would appear that they are from the perspective just of reason, that bit of memory, and introspection? Maybe those three simply aren’t able to give much of an answer: would that matter with respect to the rationality of forming perceptual beliefs? I can’t see that it would. So it isn’t clear to me that, in the case of SP, it matters much which answer we get here. The question was: would it be practically rational, in the narrow original position, to decide to engage in SP, to form beliefs in the SP way? The answer to that question, however, doesn’t really matter with respect to the question whether it is rational for us to engage in SP; we don’t have here a sensible de jure question about SP.

The situation is a bit different with CMP. First, the narrow original position is different. It includes introspection, memory, and reason, as in the previous case, but it also includes perception and sympathy. So the narrow original position with respect to CMP includes my aiming at the truth, believing what I do, in fact, believe on the basis of the standard package; it also includes having no beliefs one way or the other about the reliability of CMP. I am to try to decide which among the courses open to me is the most likely to get me in the right relation to the truth. One option is to accept CMP. Another is to reject it in favor of some other systematic practice of forming beliefs on the questions to which CMP is addressed: for example, I could accept philosophical naturalism, or perhaps some non-Christian religious practice. Still another option, presumably, would be to continue in the agnosticism that is part of the original position, and yet another is to adopt a sort of ironic Rortian double-mindedness, a frame of mind as difficult to describe as it is intriguing, one in which at one level I believe these things; at another, I maintain a certain delicate distance, sheepishly conceding that I do in a way believe these things, but adding that officially I don’t take these beliefs at all seriously, instead adopting toward them an attitude of irony and condescension. (In my study, when I reflect on it, I can see things straight; but in church, with all that liturgy, those hymns, those people I love and admire, that Bible reading and powerful preaching. . . .) And the question is, if I were in this situation, what would be the rational thing for me to do: adopt CMP, adopt some alternative to it, or remain agnostic?

Here, it seems to me, agnosticism should probably get the nod. All things considered, the best road to avoiding error and believing truth on the topics of CMP as judged from this narrow original position is agnosticism. To establish this, of course, would require a lot of work—first, a canvass of all the rational arguments for and against the existence of God, and then an examination of the arguments for and against the thought that we human beings do, in fact, perceive God (given that there is such a person). From the point of view of the standard package,¹⁴⁷¹⁴⁷   Eliminating, of course, Calvin’s sensus divinitatis, even if, as Calvin thought, that belief-forming power or mechanism is part of the original epistemic equipment of humankind generally. I think, it is somewhat more likely than not that there is such a person as God. Although the standard arguments don’t have anything like the probative force some have claimed for them, they do have (I think) some force; there are, in addition, a great number of other theistic arguments, all with at least a bit of force.¹⁴⁸¹⁴⁸   As outlined in my so-far-unpublished “Two Dozen or So Good Theistic Arguments.” On the other side is the problem of evil, of course; on balance, however, it seems to me that the nod goes to theism. But what about the claim that we human beings do in fact perceive God? Here I think the appropriate attitude would be agnosticism: from the point of view of the resources included in the narrow original position, one simply can’t determine whether we human 132beings perceive God. But to discuss this matter in proper detail would take us too far afield—particularly in view of the fact that the question put this way is, in any event, the wrong question.

For why suppose that if CMP is sensible or rational (in some important sense of that multifarious term), then from the point of view of the standard package it must be more likely than not that CMP is reliable? Consider memory, and consider its credentials from the point of view of the rest of the standard package. Suppose you don’t know that there’s been a past; you know only what reason, perception, and introspection tell you. How likely is it, from that perspective, that the deliverances of memory are mostly true? Not very likely, I’d say. Would that be a reason for mistrusting it, regarding it as suspect, or believing that it was less than wholly rational to rely on it? Would it so much as slyly suggest that it isn’t rational to form beliefs in the memory way? I don’t see how. But then presumably the same thing goes for CMP. Suppose there is such a thing as perception of God; suppose that CMP is in fact reliable. Would it follow that it is more probable than not, just given the deliverances of the standard package, that CMP is reliable? I don’t think so.

So there is no reason to hold that it is rational to take part in CMP only if its reliability is more likely than not with respect to the standard package. To think otherwise is to arbitrarily assume in advance that if CMP is a source of warranted belief, it must be likely with respect to the standard package that it is reliable; but there is no reason to accept this assumption. Here things stand with CMP just as with SP. It seems entirely arbitrary to insist that it is rational to engage in SP only if the reliability of SP is more likely than not with respect to the deliverances of some group of epistemic powers that doesn’t include SP; in the same way, it is not sensible to conclude that CMP is rational only if its reliability is more likely than not from the perspective of the standard package. Suppose God has created us with a battery of faculties aimed at our being able to acquire truth in different areas: it doesn’t follow that the reliability of any of these faculties would be more probable than not with respect to the deliverances of some package of faculties that does not include the one in question.

By way of summary, then: either the original position with respect to CMP is wide or it is narrow. If it is wide, then it will include my belief that CMP is reliable; in that case, the rational decision, clearly enough, would be to continue with CMP. But this does nothing to relieve any anxieties someone might have about the rationality or reasonability of CMP. If the original position is narrow, however, then it really doesn’t matter whether from that position it would be rational to continue with CMP.

Now suppose we return to specifically Christian belief. Our quarry is the de jure question: what is this rationality or rational justification Christian belief is alleged by its detractors not to have? Our current suggestion is that perhaps it is practical rationality. Perhaps the de jure question is the question whether Christian belief is in fact practically rational and the de jure objection is that it is not. But the same dialectic applies here as in the case of CMP. If we are thinking of the original position with respect to Christian belief as wide, then it will include Christian belief 133itself. From that point of view, obviously, the rational decision would be to continue to form and maintain belief in the way in which I do, in fact, form and maintain it (i.e., to form and maintain Christian belief); but that does little to show that Christian belief is rational in any interesting sense. So suppose, by contrast, that the appropriate original position is narrow. Then, to be sure, it will include only the standard package and it won’t include Christian belief. Now perhaps from that perspective it isn’t at all clear that the rational decision would be to endorse Christian belief; perhaps the rational decision would be to give it up. So what? Why should the truth of Christian belief (or the reliability of the sources producing it) have to be more likely than not from that standpoint for it to be rational? Why think that the rationality, in some interesting sense, of Christian beliefs requires that it be more likely than not from the standpoint of the standard package that it is reliably produced? No reason; hence we still haven’t located the de jure question.

So what is the question? Surely there is a sensible de jure question lurking somewhere in this neighborhood: what might it be? Where shall we look for it? Perhaps in the following locality. Go back to the wide original position, and recall that if, in that position, I accept SP and CMP beliefs, then, trivially, the rational thing to do is to decide to continue to form beliefs in those ways. Of course this would be true for other beliefs as well. In fact it would be true even for beliefs that are in some clear sense irrational. René Descartes notes that there are people “whose cerebella are so troubled and clouded by the violent vapours of black bile, that they . . . imagine that they have an earthenware head or are nothing but pumpkins or are made of glass.”¹⁴⁹¹⁴⁹   Meditations, Meditation I. No doubt these people avoided bumps like the plague. Given that you do believe you are made of glass, the rational thing to do is to avoid bumps. In the same way, given that you do believe you are made of glass, the rational thing to do in the service of truth is (if you are given the choice) to continue in that belief. After all, you think the belief is true; so if your aim is to believe truth and avoid falsehood, you will continue to hold it.

Fair enough: given that you think your head is made of glass, it is rational to wear your football helmet wherever you go, and rational to decide, if presented with the choice, to continue in that belief. But is it rational to hold that belief in the first place? Given that you hold the beliefs produced by SP or CMP and you don’t know of any epistemically superior practice, it is indeed rational to continue to form beliefs in that way: is it or was it rational, reasonable, sensible to hold those beliefs in the first place?

It is in this neighborhood, I suggest, that we must look for the de jure question with respect to Christian belief. What is it that determines whether a given way of acting or believing, given that your circumstances are thus-and-so, is rational or reasonable, in the relevant sense? Here is my suggestion: what determines this is what a creature of our kind with properly functioning reason (ratio) would do or believe, given that she was in those circumstances. Or perhaps it is what someone with 134ideal ratio—ratio ideal for our kind of creature—would do or think in the circumstances. The question is really about the human design plan; it has to do with what that design plan, or perhaps a slightly idealized version of it, dictates for the situation in question. The question is about the sorts of beliefs a properly functioning human being would have in the relevant circumstances. What kind of question is this? It isn’t a question of practical rationality. The question is not: given that I am in circumstances C, have aims and beliefs A and B, and have raised the question whether or not to do X, how likely is it that doing X will contribute to my aims and goals? (How sensible would it be to do X?) It’s a different kind of question altogether. In the next chapter we shall have to try to specify this question and get a closer look at it.