__________________________________________________________________ Title: Warranted Christian Belief Creator(s): Plantinga, Alvin (1932-) Print Basis: Oxford University Press, 2000 Rights: Copyright 2000 by Alvin Plantinga. Used by permission of Oxford University Press. CCEL Subjects: All; Philosophy; Apologetics LC Call no: BT1102.P57 1999 LC Subjects: Doctrinal theology Apologetics. Evidences of Christianity __________________________________________________________________ WARRANTED CHRISTIAN BELIEF Alvin Plantinga New York Oxford Oxford University Press 2000 Copyright © 2000 by Alvin Plantinga Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York, 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Plantinga, Alvin. Warranted Christian belief / Alvin Plantinga. p. cm. Includes index. ISBN 0-19-513193-2 (pbk.)--ISBN 0-19-513192-4 1. Apologetics. 2. Christianity--Philosophy. 3. Faith and reason--Christianity. I. Title. BT1102.P57 1999 230'.01-dc21 98.054362 __________________________________________________________________ To WILLIAM P. ALSTON Mentor, Model, Friend __________________________________________________________________ Preface This book is about the intellectual or rational acceptability of Christian belief. When I speak here of Christian belief, I mean what is common to the great creeds of the main branches of the Christian church, what unites Calvin and Aquinas, Luther and Augustine, Menno Simons and Karl Barth, Mother Teresa and St. Maximus the Confessor, Billy Graham and St. Gregory Palamas--classical Christian belief, as we might call it. Classical Christian belief includes, in the first place, the belief that there is such a person as God. God is a person: that is, a being with intellect and will. A person has (or can have) knowledge and belief, but also affections, loves, and hates; a person, furthermore, also has or can have intentions, and can act so as to fulfill them. God has all of these qualities and has some (knowledge, power, and love, for example) to the maximal degree. God is thus all-knowing and all-powerful; he is also perfectly good and wholly loving. Still further, he has created the universe and constantly upholds and providentially guides it. This is the theistic component of Christian belief. But there is also the uniquely Christian component: that we human beings are somehow mired in rebellion and sin, that we consequently require deliverance and salvation, and that God has arranged for that deliverance through the sacrificial suffering, death, and resurrection of Jesus Christ, who was both a man and also the second member of the Trinity, the uniquely divine son of God. I shall use the term Christian belief' to designate these two components taken together. Of course, I realize that others may use that term more narrowly or more broadly. There is no need to argue about words here: the beliefs I mentioned are the ones I shall discuss, however exactly we propose to use the term Christian'. I also recognize that there are partial approximations to Christian belief so understood, as well as borderline cases, beliefs such that it simply isn't clear whether they qualify as Christian belief. All of this is true, but as far as I can see, none of it compromises my project. Accordingly, our question is this: is belief of this sort intellectually acceptable? In particular, is it intellectually acceptable for us, now? For educated and intelligent people living in the twenty-first century, with all that has happened over the last four or five hundred years? Some will concede that Christian belief was acceptable and even appropriate for our ancestors, [1] people who knew little of other religions, who knew nothing of evolution and our animal ancestry, nothing of contemporary subatomic physics and the strange, eerie, disquieting world it postulates, nothing of those great masters of suspicion, Nietzsche, Marx, and Freud, nothing of the acids of modern historical biblical criticism. But for us enlightened contemporary intellectuals (so the claim continues) things are wholly different; for people who know about those things (people of our rather impressive intellectual attainments), there is something naive and foolish, or perhaps bullheaded and irresponsible, or even vaguely pathological in holding onto such belief. But can't we be a little more precise about the objection? What, exactly, is the problem? The answer, I think, is that there are alleged to be two main problems. Western thought since the eighteenth-century Enlightenment has displayed at least two distinct styles of objection. First, there have been de facto objections: objections to the truth of Christian belief. Perhaps the most important de facto objection would be the argument from suffering and evil. This objection goes all the way back to Democritus in the ancient world but is also the most prominent contemporary de facto objection (see chapter 14). It has often been stated philosophically, but has also received powerful literary expression (for example, in Dostoevski's The Brothers Karamazov). The objection goes as follows: according to Christian belief, we human beings have been created by an all-powerful, all-knowing God who loves us enough to send his son, the second person of the divine Trinity, to suffer and die on our account; but given the devastating amount and variety of human suffering and evil in our sad world, this simply can't be true. The argument from evil may be the most important de facto objection, but it isn't the only one. There are also the claims that crucial Christian doctrines--Trinity, Incarnation, or Atonement, for example--are incoherent or necessarily false. Many have argued that the Christian doctrine of three divine persons with one nature cannot be coherently stated; many have claimed that it is not logically possible that a human being, Jesus of Nazareth, should also be the second person of the divine Trinity, and many have thought it impossible that one person's suffering--even if that person is divine--should atone for someone else's sins. Indeed, there are claims that the advance of science has somehow shown that there really isn't any supernatural realm at all--no God who has created us and governs our world, let alone a Trinity of divine persons, one of whom became a human being, died, and rose from the dead, thereby redeeming human beings from sin and suffering. De facto objections, therefore, are many, and they enjoy a long and distinguished history in Western thought. Even more prevalent, however, have been de jure objections. These are arguments or claims to the effect that Christian belief, whether or not true, is at any rate unjustifiable, or rationally unjustified, or irrational, or not intellectually respectable, or contrary to sound morality, or without sufficient evidence, or in some other way rationally unacceptable, not up to snuff from an intellectual point of view. There is, for example, the Freudian claim that belief in God is really a result of wish fulfillment; there is the evidentialist claim that there isn't sufficient evidence for Christian belief; and there is the pluralist claim that there is something arbitrary and even arrogant in holding that Christian belief is true and anything incompatible with it false. De facto and de jure objections are separate species, but they sometimes coincide. Thus there is a de jure objection from suffering and evil as well as a de facto: it is often claimed that the existence of suffering and evil in the world makes it irrational to hold that Christian belief is, in fact, true. De facto objections are relatively straightforward and initially uncomplicated: the claim is that Christian belief must be false (or at any rate improbable), given something or other we are alleged to know. Quite often the claim is that it is something we now know, something our ancestors allegedly did not know, as in Rudolf Bultmann's widely quoted remark that ?it is impossible to use electrical light and the wireless and to avail ourselves of modern medical and surgical discoveries, and at the same time to believe in the New Testament world of spirits and miracles.? [2] De jure objections, by contrast, while perhaps more widely urged than their de facto counterparts, are also much less straightforward. The conclusion of such an objection will be that there is something wrong with Christian belief--something other than falsehood--or else something wrong with the Christian believer: it or she is unjustified, or irrational, or rationally unacceptable, in some way wanting. But what way, exactly? Just what is it to be unjustified or irrational? No doubt it is a bad thing to hold beliefs that are rationally unjustified: but what precisely is the problem? Wherein lies the badness? This is ordinarily not made clear. According to the evidentialist, for example, the evidence for Christian belief is insufficient: but insufficient for what? And suppose you believe something for which the evidence is insufficient: what, exactly, is the matter with you? Are you thereby subject to moral blame, or shown to be somehow incompetent, or unusually ignorant, or subject to some kind of pathology, or what? According to Freud and some of his followers, Christian and theistic belief is a product of wish fulfillment or some other projective mechanism. Well, suppose (contrary to fact, as I see it) that is true: just what is the problem? Is it that such a belief is likely to be false? Is it that if you accept a belief formed on the basis of wish fulfillment, you have done something that merits blame? Or are you, instead, a proper object of pity? What, precisely, is the problem? These questions are much harder to answer than one might think. One project of this book is to try to answer them: I try to find a serious and viable de jure objection to Christian belief. That is, I try to find a de jure objection that is both a real objection and also at least plausibly attached to Christian belief. But there is a prior question: is there actually any such thing as Christian belief, conceived as Christians conceive it? Some thinkers (often citing the authority of the great eighteenth-century philosopher Immanuel Kant) argue that we couldn't so much as think about such a being as the Christian God, infinite and transcendent as he is supposed to be. That is because our all-too-human concepts could not apply to such a being; our concepts can apply only to finite beings, beings who are not transcendent in the way Christians take God to be. But if it is really true that our concepts cannot apply to an infinite and transcendent being, if we cannot so much as think about such a being, then we human beings also have no beliefs about such a being. Indeed, we can't have beliefs about such a being. ANd then the fact is there isn't any such thing as Christian belief: Christians think they have beliefs about an infinite and transcendent being, but in fact they are mistaken. In part I, ?Is There a Question?? (chapters 1 and 2), I argue that there is no reason at all to accept this skeptical claim: Kant himself provides no reason, and those contemporaries who appeal to his authority certainly do no better. That conclusion clears the deck for the main question of the book: is there a viable de jure objection to Christian belief? One that is independent of de facto objections and does not presuppose that Christian belief is false? There are, I believe, fundamentally three main candidates: that Christian belief is unjustified, that it is irrational, and that it is unwarranted. These candidates will be introduced in due course; for the moment, note just that three of the main characters of this drama, therefore, are justification, rationality, and warrant. In part II, ?What Is the Question?? (chapters 3-4), I ask first whether a viable de jure criticism can be developed in terms of justification and rationality; I conclude that it cannot. Then I turn (chapter 5) to the objections offered by Freud, Marx, and Nietzsche; and here we finally find an initially promising candidate for a de jure objection. This complaint is in the vicinity of warrant. To see what warrant is, note that not all true beliefs constitute knowledge. You are an ardent Detroit Tigers fan; out of sheer bravado and misplaced loyalty, you believe that they will win the pennant, despite the fact that last year they finished last and during the off season dealt away their best pitcher. As it happens, the Tigers unaccountably do win the pennant, by virtue of an improbable series of amazing flukes. Your belief that they will, obviously, wasn't knowledge; it was more like an incredibly lucky guess. To count as knowledge, a belief, obviously enough, must have more going for it than truth. That extra something is what I call warrant'. As I see it, if there are any real de jure objections to Christian belief, they lie in the neighborhood of warrant. That may not come as much of a surprise, given that this book is a sequel to Warrant: The Current Debate and Warrant and Proper Function. [3] In the first of those books I introduced the term warrant' as a name for that property--or better, quantity--enough of which is what makes the difference between knowledge and mere true belief. I went on to examine the various contemporary theories of warrant: what exactly is the property that distinguishes knowledge from mere true belief? I canvassed the contemporary theories on offer: is it justification? coherence? rationality? being produced by reliable belief-producing faculties or processes? The answer, I argued, is none of the above; none of these theories is right. In Warrant and Proper Function I went on to give what seems to me to be the correct answer: warrant is intimately connected with proper function. More fully, a belief has warrant just if it is produced by cognitive processes or faculties that are functioning properly, in a cognitive environment that is propitious for that exercise of cognitive powers, according to a design plan that is successfully aimed at the production of true belief. (For explanation of that perhaps baffling formula, see chapter 5.) According to Freud and Marx, therefore, the real problem with theistic (and hence Christian) belief is that it lacks warrant. In part III, ?Warranted Christian Belief? (chapters 6-10), I address this objection. As it turns out, this de jure objection is really dependent on a de facto objection. That is because (as I argue) if Christian belief is true, then it is also warranted; the claim that theistic (and hence Christian) belief is unwarranted really presupposes that Christian belief is false. Freud and Marx, therefore, do not give us a de jure objection that is independent of the truth of Christian belief; their objection presupposes its falsehood. I go on (in chapter 6) to offer a model, the Aquinas/Calvin (A/C) model, for theistic beliefs having warrant and argue that if theistic belief is in fact true, then something like this model is in fact correct. In chapters 7 through 10, I extend the A/C model to cover full-blown Christian belief (as opposed to theistic belief simpliciter), beginning (in chapter 7) with an account of the place of sin in the model. Next (in chapters 8 and 9), I propose the extended A/C model; according to this model, Christian belief is warranted because it meets the conditions of warrant spelled out in Warrant and Proper Function. That is, Christian belief is produced by a cognitive process (the ?internal instigation of the Holy Spirit? [Aquinas] or the ?internal testimony of the Holy Spirit? [Calvin]) functioning properly in an appropriate epistemic environment according to a design plan successfully aimed at truth. Chapter 8, ?The Extended A/C Model: Revealed to Our Minds,? sets out the cognitive side of this process. The process involves the affections as well as reason, however (i.e., it involves will as well as intellect), and chapter 9, ?The Testimonial Model: Sealed upon Our Hearts,? explains some of the connections between reason and the affections. Chapter 10 concludes part III by considering various actual and possible objections to the model; none is successful. What I officially claim for the extended A/C model is not that it is true but, rather, that it is epistemically possible (i.e., nothing we know commits us to its falsehood); I add that if Christian belief is true, then very likely this model or something like it is also true. If I am right in these claims, there aren't any viable de jure criticisms that are compatible with the truth of Christian belief; that is, there aren't any viable de jure objections independent of de facto objections. And if that is so, then the attitude expressed in ?Well, I don't know whether Christian belief is true (after all, who could know a thing like that?), but I do know that it is irrational (or intellectually unjustified or unreasonable or intellectually questionable)?--that attitude, if I am right, is indefensible. Finally, in part IV, ?Defeaters?? (chapters 11-14), I confront the following claim. Someone might concede that Christian belief could in principle be warranted in the way the model suggests; she might go on to insist, however, that in fact there are various defeaters for the warrant Christian belief might otherwise enjoy. A defeater for a belief A is another belief B such that once you come to accept B, you can no longer continue to accept A without falling into irrationality. In the present case, then, these alleged defeaters would be beliefs a knowledgeable Christian can be expected to have; they would also be beliefs such that one who accepts them cannot rationally continue in firmly accepting Christian belief. After exploring the nature of defeaters, I examine the chief candidates: first, the alleged abrasive results of historical biblical criticism; second, a recognition of the variety and importance of religions incompatible with Christian belief, together with certain related postmodern claims; and third, a deep recognition of the facts of suffering and evil. I argue that none of these succeeds as a defeater for classical Christian belief. This book can be thought of in at least two quite different ways. On the one hand, it is an exercise in apologetics and philosophy of religion, an attempt to demonstrate the failure of a range of objections to Christian belief. De jure objections, so the argument goes, are either obviously implausible, like those based on the claim that Christian belief is not or cannot be justified, or else they presuppose that Christian belief is not true, as with those based on the claim that Christian belief lacks external rationality or lacks warrant. Hence there aren't any decent de jure objections that do not depend on de facto objections. Everything really depends on the truth of Christian belief; but that refutes the common suggestion that Christian belief, whether true or not, is intellectually unacceptable. On the other hand, however, the book is an exercise in Christian philosophy: in the effort to consider and answer philosophical questions--the sorts of questions philosophers ask and answer--from a Christian perspective. What I claim for the extended A/C model of chapters 8 and 9 is twofold: first, it shows that and how Christian belief can perfectly well have warrant, thus refuting a range of de jure objections to Christian belief. But I also claim that the model provides a good way for Christians to think about the epistemology of Christian belief, in particular the question whether and how Christian belief has warrant. So there are two projects, or two arguments, going on simultaneously. The first is addressed to everyone, believer and nonbeliever alike; it is intended as a contribution to an ongoing public discussion of the epistemology of Christian belief; it does not appeal to specifically Christian premises or presuppositions. I shall argue that, from this public point of view, there isn't the faintest reason to think that Christian belief lacks justification, rationality, or warrant--at least no reason that does not presuppose the falsehood of Christian belief. The other project, however--the project of proposing an epistemological account of Christian belief from a Christian perspective--will be of special interest to Christians. Here the project is that of starting from an assumption of the truth of Christian belief and from that standpoint investigating its epistemology, asking whether and how such belief has warrant. We might think of this project as a mirror image of the philosophical naturalist's project, when he or she assumes the truth of naturalism and then tries to develop an epistemology that fits well with that naturalistic standpoint. I hope Christians will find this second project appealing; I hope others as well may be interested--just as those who don't accept philosophical naturalism might nevertheless be interested to see what kind of epistemology would best go with naturalism. The centerpiece of each of the two projects--the apologetic project and the project in Christian philosophy--is the extended A/C model. Taken the first way, that model is a defense of the idea that Christian belief has warrant and an effort to show that if Christian belief is true, then (very likely) it does have warrant; taken the second, it is a recommendation as to how Christians can profitably understand and conceive of the warrant they take Christian belief to have. The reader is owed an apology for the inordinate length of this book. All I can say by way of self-exculpation is that its length is due to a determination not to commit a tetralogy: a trilogy is perhaps unduly self-indulgent, but a tetralogy is unforgivable. (I suppose a cynic might question the difference between a tetralogy and a trilogy, the last member of which is twice as long as the preceding members.) In any event, not every reader need read every page. For example, those readers who are not tempted to think that our concepts could not apply to God can safely skip part I, and those readers who want to read only the central part of the story line can confine themselves to chapters 6 through 9. Furthermore, even though the book is long, I am aware that it covers a shameful amount of ground, and that in nearly every chapter (but in particular chapters 8 and 12) a really proper job would go into considerably more detail. My excuse is that it is important to see the forest as well as the trees; God may be in the details, but God is also in the whole sweeping vista of the epistemology of Christian belief. Trained observers may note two styles of print: large and small. The main argument of the book goes on in the large print; the small print adds further analysis, argument, or other points the specialist may find of interest. This book is not written mainly for the specialist in philosophy. I hope and intend that it will be intelligible and useful to, for example, students who have taken a course in philosophy or apologetics, as well as the fabled general reader with an interest in its subject. Although it is the third member of a trilogy, the book is designed to be relatively independent of Warrant: The Current Debate and Warrant and Proper Function (and it therefore sometimes contains brief accounts of what goes on in them). Most books, of course, are to one degree or another cooperative enterprises; every author is heavily indebted to others in a thousand profound ways. (This is, if anything, especially evident in those authors--one thinks of a long line of modern and contemporary philosophers, beginning perhaps with Descartes--who apparently believe that they have jettisoned all that has already been thought and written, starting the whole subject anew.) The present book is no exception; it is very much a cooperative enterprise. This is so for the usual reasons, but also for a special reason. At several junctures, I have simply appealed to the work of others--most often William P. Alston and Nicholas Wolterstorff--for a particular building block of the argument. This is especially so when I have little or nothing to add to what they have already said on the topic in question, but sometimes also when I might myself have a mildly different slant on the topic in question. The book is also something of a cooperative enterprise by virtue of the advice, instruction, and criticism I have received from others--an embarrassingly large number of others (I am entirely sensible of the fact that with so much help, I should have done better). I'm grateful to all those who helped me with the first two volumes and also to Jonathan Kvanvig and the authors of the essays in his Warrant in Contemporary Epistemology: Essays in Honor of Plantinga's Theory of Knowledge. Among those to whom I am particularly grateful for help with this third volume are Karl Ameriks, Jim Beilby (who also kindly prepared the index), David Burrell, Kelly Clark, John Cooper, Kevin Corcoran, Andrew Cortens, Fred Crosson, Paul Draper, Steve Evans, Ronald Feenstra, Fred Freddoso, Richard Gale, Lee Hardy, John Hare, Van Harvey, David Hunt, Hugh McCann, Greg Mellema, Ric Otte, Neal Plantinga, Bill Prior, Tapio Puolimatka, Philip Quinn, Del Ratzsch, Dan Rieger, Robert Roberts, Bill Rowe, John Sanders, Henry Schuurman, James Sennett, Ernie Sosa, Michael Sudduth, Richard Swinburne, Bill Talbott, James VanderKam, Bas van Fraassen, Calvin Van Reken, Rene van Woudenberg, Steve Wykstra, and Henry Zwaanstra. (I've undoubtedly omitted people who belong on this list; to them, I express both my gratitude and my apologies.) William Alston, Dewey Hoitenga, Eleonore Stump, and Nicholas Wolterstorff read and commented on the entire manuscript; to them, I am especially grateful. One of my most significant debts is to a rotating cadre of Notre Dame graduate students who, as a group and over a period of several years, read the entire manuscript and submitted it to the sort of searching and detailed criticism that only aroused and contentious graduate students can muster. This group includes, among others, Mike Bergmann, Tom Crisp, Pat Kain, Andy Koehl, Kevin Meeker, Trenton Merricks, Marie Pannier, Mike Rea, Ray Van Arragon and David VanderLaan. I am similarly indebted to Nicholas Wolterstorff's fall 1997 graduate seminar at Yale (especially Andrews Chignell and Dole), who (with their mentor) read the manuscript and provided illuminating and valuable comments. And once again I thank Martha Detlefsen, whose valiant efforts to keep me and this manuscript properly organized have been ingenious and untiring. These three volumes began life as Gifford Lectures at the University of Aberdeen in 1986 and 1987 and Wilde Lectures at Oxford in 1988. I am grateful to both sets of electors; I am equally grateful for the hospitality my wife and I enjoyed while visiting Aberdeen and Oxford. Thanks are also due to the University of Notre Dame for a sabbatical in 1995 and 1996 and to the National Endowment for the Humanities for a fellowship that same year. A couple of bits of this volume have already seen the light of publication: chapter 13 contains a few pages from ?Pluralism: A Defense of Religious Exclusivism? in The Rationality of Belief and the Plurality of Faith, ed. Thomas Senor (Ithaca: Cornell University Press, 1995), and chapter 14 contains a few paragraphs from ?On Being Evidentially Challenged,? in The Evidential Argument from Evil, ed. Daniel Howard-Snyder (Bloomington: Indiana University Press, 1996). Finally, I make special mention of William P. Alston. Bill was my teacher when I began graduate school in 1954 (if I wasn't able to understand Alfred Whitehead's Process and Reality, the subject of that first seminar, the fault was mine [or maybe Whitehead's], not Bill's.) I learned much from him then and much more from him since. His generosity in reading this entire manuscript was characteristic; so was the trenchancy and penetration of his comments. Alston's contributions to contemporary philosophy and philosophy of religion (his leadership in establishing the Society of Christian Philosophers and the journal Faith and Philosophy, his splendid works in epistemology and philosophy of religion) are, of course, well and widely known; there is no need to enumerate them here (and in any case they are nearly indenumerable). It is to him that I dedicate this book. A. P. Notre Dame, Indiana September 1998 __________________________________________________________________ [1] And perhaps (as they may add) even for contemporaries who lead sheltered lives in cultural backwaters--for instance, the area between the east and west coasts of the United States. [2] Kerygma and Myth (New York: Harper and Row, 1961), p. 5. [3] Both published by Oxford University Press (New York, 1993). This book is also, and in a slightly different direction, a sequel to God and Other Minds (Ithaca: Cornell University Press, 1967) and ?Reason and Belief in God,? in Faith and Rationality, ed. A. Plantinga and N. Wolterstorff (Notre Dame: University of Notre Dame Press, 1983). __________________________________________________________________ PART I IS THERE A QUESTION? __________________________________________________________________ 1 Kant ?To whom will you liken me? Who is my equal? With what can you compare me? Where is my like?? Isaiah 46:5 __________________________________________________________________ I. The Problem Our interest, in this book, is the de jure question: [4] is it rational, reasonable, justifiable, warranted to accept Christian belief--Christian belief as outlined in the preface? Or is there something epistemically unacceptable in so doing, something foolish, or silly, or foolhardy, or stupid, or unjustified, or unreasonable, or in some other way epistemically deplorable? But there is a prior question: is the very idea of Christian belief coherent? Can there really be such a thing as Christian belief? Well, why should that be a question? Isn't it obvious that many people hold just those beliefs mentioned in the preface? Here is the problem. To accept Christian belief, I say, is to believe that there is an all-powerful, all-knowing, wholly good person (a person without a body) who has created us and our world, who loves us and was willing to send his son into the world to undergo suffering, humiliation, and death in order to redeem us. It is also to believe, of course, that no more than one being has these properties. And Christian belief involves not only that there is such a being but also that we are able to address him in prayer, refer to him, think and talk about him, and predicate properties of him. We have some kind of cognitive access to and grasp of him. We can refer to him, for example, as the all-powerful, all-knowing person who has created and upholds the world, and we can predicate of him such properties as being all-powerful, being all-knowing, and having created the world. We can use a definite description like this to refer to this being, to pick him out, to single him out for thought; and we can give a proper name to the being thus singled out. For example, we can use the term God' as his name. Accordingly, Christians ordinarily take it for granted that it is possible to refer to God by such descriptions as the all-powerful, all-knowing creator of the universe', and possible, furthermore, to predicate properties (wisdom, goodness) of the being thus referred to. Of course, such a description succeeds in actually naming something only if there really is a being who is all-powerful and all-knowing and created the universe. Furthermore, it must be possible, if I can think about God and predicate properties of him, not only that there be such a being but also that my concepts apply to it. If not, then I am not in a position to assert or believe or even entertain any of the propositions mentioned above, if indeed there are any such propositions. Now Christians also take it for granted that God is infinite, transcendent, and ultimate (however, precisely, we gloss those terms). And just here is the alleged problem. It seems many theologians and others believe that there is real difficulty with the idea that our concepts could apply to God--that is, could apply to a being with the properties of being infinite, transcendent, and ultimate. The idea is that if there is such a being, we couldn't speak about it, couldn't think and talk about it, couldn't ascribe properties to it. If that is true, however, then, strictly speaking, Christian belief, at least as the Christian understands it, is impossible. For Christians believe that there is an infinite, transcendent, ultimate being about whom they hold beliefs; but if our concepts cannot apply to a being of that sort, then there cannot be beliefs about a being of that sort. This idea often sees the light of publication; it is even more heavily present in the oral tradition. In the spirit of interdisciplinary ecumenism, therefore, I want to begin by looking into this question. Consider, for example, the theologian Gordon Kaufman: The central problem of theological discourse, not shared with any other ?language game? is the meaning of the term ?God.? ?God? raises special problems of meaning because it is a noun which by definition refers to a reality transcendent of, and thus not locatable within, experience. [5] In particular, it seems to be widely accepted, among theologians, that Kant showed that reference to or thought about such a being (even if there is one) is impossible or at least deeply problematic, [6] or at any rate much more problematic than the idea that we can refer to and think about ourselves and other people, trees and mountains, planets and stars, and so on. Those theologians who think or suspect Kant showed this do not ordinarily develop the point in detail; [7] they ordinarily content themselves with a ritual bow in his direction. They do not explain how they think these things were shown or what the arguments establishing them are; perhaps they think (quite properly) that that is the job of philosophers. Some of these theologians then go on to suggest that language ostensibly about a transcendent God isn't what it looks like at all; it really serves some quite different purpose. Alternatively, perhaps, it really serves no useful purpose as it stands; what we have to do is find a useful purpose for it to serve. Perhaps it can be used, somehow, to further or promote human flourishing and humaneness, [8] or religious tolerance, [9] or liberating praxis, or the rights of women, [10] or the fight against oppression. But what is important for my present purposes is not an exploration of the ways in which religious language might be reconstrued or restructured, once we see (as we think) that it cannot function the way ordinary believers think it does; I want, instead, to examine the prior claim that, indeed, it cannot function as ordinary believers assume it does. Is there really something especially problematic about referring to or thinking about God? Did Kant show that if there were such a person as God, we couldn't refer to or think about him? Or if show' is too strong a word, did he give us powerful or even decent reason to believe that our concepts couldn't apply to God, if there is such a being? Or if he didn't do that, do some of his contemporary followers--Gordon Kaufman, for example, or John Hick--give us a reason to think this is indeed true? And is the claim in question--that our concepts do not apply to God--a coherent one? (Or rather, is there a coherent claim somewhere in the nearby bushes, since clearly there are several different claims lurking in these bushes?) Initially, the answer seems to be no; one who makes the claim seems to set up a certain subject for predication--God--and then declare that our concepts do not apply to this being. But if this is so, then, presumably, at least one of our concepts--being such that our concepts don't apply to it--does apply to this being. Either those who attempt to make this claim succeed in making an assertion or not. If they don't succeed, we have nothing to consider; if they do, however, they appear to be predicating a property of a being they have referred to, in which case at least some of our concepts do apply to it, contrary to the claim they make. So if they succeed in making a claim, they make a false claim. Note how difficult it is, initially, to state the claim in question, the claim that if there is a being with the properties Christians ascribe to God, our concepts would not apply to that being. Consider the proposition (1) If there were an infinite, transcendent, and ultimate being, our concepts could not apply to it. But now suppose (1) were true. The idea, one takes it, is that we do have at least some grasp of the properties of being infinite, transcendent, and ultimate (else we shouldn't be able to understand the sentence or grasp the proposition it expresses). An infinite being, we might say, is an unlimited being--unlimited, that is, with respect to certain properties. Among these properties might be power, knowledge, goodness, love, and the like. (A being is unlimited with respect to power and (propositional) knowledge, for example, if there is a maximal degree of power and knowledge, and the being in question enjoys that maximal degree of those properties. It might be hard to say precisely what the maximal degree of these properties is; with respect to knowledge, we might begin by saying that a being displays that maximal degree if it knows all true propositions and believes no false proposition.) Perhaps we can also give an explanation of what it is for a being to be transcendent: such a being transcends the created universe; and a being transcends the created universe if it is not identical with any being in that universe (if it is not created) and if it depends on nothing at all for its existence. So we do have the ideas of transcendence and being infinite (and if not, then (1) makes no sense). And the idea behind (1) is that if there is such a being (i.e., if there is an infinite and transcendent being), then none of our concepts could apply to it. In particular, then, the concepts being infinite and being transcendent could not apply to it. But how could that be? How could it be that there is a being that is infinite and transcendent (i.e., falls under our concepts infinity and transcendence) but is nevertheless such that the concepts infinity and transcendence do not apply to it? Is the idea, perhaps, that these concepts are impossible, incoherent, like the concept of a round square, a concept such that we can just see a priori that it couldn't apply to anything, that there couldn't be a thing to which it applied? [11] That would make (1) trivially true, at least if a conditional with an impossible antecedent is thereby true. Of course, it would also make (1*) true: (1*) If there were an all-powerful, all-knowing being, our concepts would apply to it. So presumably that is not the idea here. What, then, is the idea? I think the best we can do in trying to state such a view coherently is to say with John Hick (see below, pp. 47ff.) that there is a being to which none of our positive, nonformal concepts apply (a being that has none of the positive, nonformal properties of which we have concepts) and that this being, somehow, is the one with which Christians and others are in touch in religious practice. This is perhaps the best we can do; I shall argue below (pp. 59ff.), however, that it isn't good enough; it suffers from serious, indeed fatal difficulties. So the suggestion is that Kant showed us, somehow, that there are real, perhaps insurmountable problems in the idea that there is a being like that acknowledged in traditional Christianity, to whom we can refer and to whom our concepts apply. This is a question of considerable import for our present project, for if this suggestion is right, then there really isn't any such question as the one I say I propose to discuss; then, the sentences Christians use to express (as they think) their beliefs, do not really express the kinds of propositions or thoughts Christians think they express. Indeed, perhaps, they don't express any propositions or thoughts at all but are a sort of disguised nonsense: they look as if they express propositions but in fact do not. Before we explicitly turn to Kant, however, it is worth reminding ourselves that the claim in question is by no means a new claim in the present historical context. Beginning in the 1930s, the logical positivists were fond of insisting that the sentences Christians typically use--God loves us' or The universe was created by God' or God was in Christ, reconciling the world to himself'--do not, as they are ordinarily used, say anything at all; they express no propositions at all; they are really disguised nonsense. [12] They look like they say something, and Christians and others think they say something; in fact, however, they altogether fail to express a proposition, just as does an obvious nonsense sentence like ?'Twas brillig, and the slithy toves / did gyre and gimbol in the wabe.? The positivists appealed to the dreaded ?Verifiability Criterion of Meaning,? according to which a sentence makes sense, is literally significant, or is cognitively meaningful only if it is empirically verifiable' (or falsifiable)--only if, that is, its truth (or falsehood) can be established by something like the methods of natural and empirical science. Beginning in the 1940s or so, the main questions asked and answered by philosophers of religion in the English-speaking world were whether it is possible to refer to God at all and whether the sentences typically uttered by Christians and other believers in God really make sense or are, instead, nonsense, cognitively insignificant. [13] Of course it doesn't follow that such meaningless sentences are altogether useless; perhaps they serve some other function. Rudolf Carnap, for example, wondered whether the meaningless sentences of metaphysics and theology might not really be a form of music. [14] (It isn't known whether he expected them to supplant Mozart and Bach, or even Wagner. I myself doubt that metaphysics will ever replace Mozart, but perhaps we could see it as a peculiarly avant-garde form of rock.) By now, logical positivism has retreated into the obscurity it so richly deserves. [15] There still persists, however, the widespread impression that reference to God is problematic; it is time to turn explicitly to Kant, the main source of this idea. Does his work offer cause for concern to those who propose to think about, refer to, pray to, or worship a being described the way Christians describe God--as a personal being who is transcendent and infinite? __________________________________________________________________ [4] For the contrast between de jure and de facto questions, see Preface, pp. viff. [5] God the Problem (Cambridge: Harvard University Press, 1972), p. 8. [6] The whole medieval tradition of negative theology also finds reference to God problematic. The difference is that the medievals took it for granted that, of course, we can refer to God; the problem is to explain just how this can be accomplished. For the contemporaries I am thinking of, however, the difficulties (whether apparent or real) lead them to doubt that we can, in fact, refer to and talk about a being that is ultimate and transcendent. [7] As we shall see in chapter 2, however, John Hick constitutes an exception. [8] As in Gordon Kaufman: see chapter 2, p. 41. [9] As in John Hick: see chapter 2, p. 60. [10] See Sallie McFague, Models of God (Philadelphia: Fortress Press, 1987). [11] Thus some philosophers have claimed that the notion of omnipotence is incoherent; others have paid the same compliment to the notion of omniscience (see Patrick Grim and Alvin Plantinga, ?Truth, Omniscience, and Cantorian Arguments: An Exchange,? Philosophical Studies 70 [August 1993]); still others have argued the same point with respect to the idea that God is a person without a body. [12] See, e.g., A. J. Ayer, Language, Truth and Logic (New York: Dover Publications, 1946), pp. 115ff. [13] See, e.g., New Essays in Philosophical Theology, ed. Antony Flew and Alasdair MacIntyre (London: SCM Press, 1955). [14] Perhaps metaphysics can have other aesthetic functions as well, as can Carnap's own work. Although, as far as I know, no one has ever used Carnap's writings as music (or even set them to music), in 1976 the Museum of Modern Art in Oxford displayed a page of Carnap's Logical Syntax of Language magnified about 20x and posted on the wall. No doubt a piece of metaphysics could serve the same purpose. [15] For an account of the harrowing vicissitudes of the Verifiability Criterion, see Carl Hempel, ?Problems and Changes in the Empiricist Criterion of Meaning,? in Semantics and the Philosophy of Language, ed. Leonard Linsky (Urbana: University of Illinois Press, 1952), and my God and Other Minds (Ithaca: Cornell University Press, 1967), chapter 7. Something like it lingers on, not only among some theologians who propose to reconstrue religious language in such a way that it no longer refers to God but also in the Wittgensteinian fideism of D. Z. Phillips and others, which is a sort of continuation of positivism by other means. Although some of this work is eminently worth discussing, I will not discuss it here, referring the reader instead to Nicholas Wolterstorff's perceptive ?Philosophy of Religion after Foundationalism I: Wittgensteinian Fideism? (presently unpublished), to which I have little to add. __________________________________________________________________ II. KANT Immanuel Kant was a virtual titan of philosophy, with an absolutely enormous influence upon subsequent philosophy and theology. This is no doubt due to his great insight and raw philosophical power; it is perhaps also due to the grave hermeneutical difficulties that attend study of his work. The British philosopher David Hume writes with a certain surface clarity that disappointingly disappears on closer inspection. With Kant, there is good news and bad news: the good news is that we don't suffer that disappointment; the bad news is that it's because there isn't any surface clarity to begin with. We can't turn to a settled interpretation of Kant to see whether he showed or even held that our concepts don't apply to God; there is no settled interpretation. The first thing to note, however, is that Kant often writes as if we can perfectly well refer to God. In the Critique of Practical Reason and elsewhere (Religion within the Boundary of Pure Reason; Lectures on Philosophical Theology), Kant regularly seems to refer to God and clearly takes himself to be doing exactly that. Even in the Critique of Pure Reason, his work most heavily influential in this skeptical direction, Kant often seems to suggest that we can indeed refer to and think about God. He often seems to suggest that the problem is not that we can't think about God but that we can't come to speculative or metaphysical knowledge of God. His aim in this Critique, he says, is to curb knowledge in order to make room for faith. [16] The faith in question, presumably, is like that expressed in the Critique of Practical Reason and elsewhere; it would certainly involve referring to God and taking his existence and attributes as a postulate of practical reason, a presupposition of the reality and seriousness of the moral life. Indeed, some who understand him this way believe that Kant was himself a theist, holding that the things in themselves are just things as they appear to God, that is, things as they really are. [17] Of course if this way of thinking about Kant is correct, then on his view it is perfectly possible to refer to God; if that is possible, it is also possible to ascribe properties and attributes to him; and if that is possible, then our concepts do, indeed, apply to him. For example, the negative concepts not being in space and time and not being dependent on human beings for his existence would thus apply to him. Further, on this understanding of Kant, such positive concepts as having knowledge and having power would apply to God, as would having created the world. On this understanding, it would be an error to suppose that Kant showed that our concepts can't apply to God--unless one were prepared to hold that Kant showed this but failed to notice that he did, thus mistakenly taking himself to be referring to that to which he himself showed it was not possible to refer. This latter is, of course, a possibility, although it would require an unusually high level of absentmindedness. Still, the idea that according to Kant our concepts couldn't apply to God is no mere fabrication, no merely thoughtless misunderstanding--or, more exactly, if it is a misunderstanding, it is one with considerable basis in the Kantian text. There is much in the Critique of Pure Reason to suggest this or something like it; at any rate, there is much to suggest that the categories of the understanding, which are concepts of the first importance, do not apply to the things in themselves (and thus not to God). For example: If, therefore, we should attempt to apply the categories to objects which are not viewed as being appearances, we should have to postulate an intuition other than the sensible, and the object would thus be a noumenon in the positive sense. Since, however, such a type of intuition, intellectual intuition, forms no part whatsoever of our faculty of knowledge, it follows that the employment of the categories can never extend further than to the objects of experience. (A353, B309, Kant's emphasis) Here and elsewhere, Kant suggests that the categories of the understanding do not apply beyond the realm of appearance, the world of phenomena. (?Suggests,? I say; these passages, like all the others, contain more than a hint of possible ambiguity.) But if those categories do not apply to the noumena, the Dinge an sich, then perhaps the same goes for the rest of our concepts. And if our concepts do not apply beyond the world of experience, the world of appearance, then they do not apply to God, who, of course, would be a noumenon in excelsis. So the claim would be that Kant shows or believes (at any rate in the Critique of Pure Reason) that our concepts do not apply to God, in which case we cannot refer to or think about him. __________________________________________________________________ A. Two Worlds or One? What is to be said for this understanding of Kant? Hermeneutical obstacles of formidable proportions loom. First, how are we to think about this distinction between the noumena and the phenomena, the things in themselves and the things for us? Unfortunately, the commentators are not of one mind. There is a huge interpretative watershed, a continental divide, between two fundamentally different interpretations or basic pictures of what Kant had in mind, each with several variations when it comes to detail. According to the first and more traditional picture, Kant held that there are two realms of objects, two fundamentally different kinds of things. These are the phenomena, on the one hand, and the noumena, on the other; the things in themselves and the things für uns. (These two distinctions don't exactly coincide in Kant; the ways in which they don't aren't relevant to our present inquiry.) On the one hand, on this picture, there are tables and chairs, horses and cows, stars and planets, the oak tree in your backyard, just as we ordinarily think. These things really exist and are really there. They are phenomenally real, real parts of the world of experience. But they are also transcendentally ideal: that is, they are not part of the world as it is independent of human experience. On the other hand, there are the noumena, which are transcendentally real. These are the things as they are in themselves; they do not depend for their existence or character upon human beings or human experience. These two realms are disjoint: none of the phenomenal objects is a noumenon, and none of the noumenal objects is a phenomenon. Here are a couple of passages supporting this interpretation: Now we must bear in mind that the concept of appearances, as limited by the Transcendental Aesthetic, already of itself establishes the objective reality of noumena and justifies the division of objects into phaenomena and noumena and so of the world into a world of the sense and a world of the understanding (mundus sensibilis et intelligibilis) and indeed in such manner that the distinction does not refer merely to the logical form of our knowledge of one and the same thing, according as it is indistinct or distinct, but to the difference in the manner in which the two worlds can be first given to our knowledge, and in conformity with this difference, to the manner in which they are in themselves generically distinct from one another. (A249, Kant's emphasis) Appearances are the sole objects which can be given to us immediately, and that in them which relates immediately to the object is called intuition. Appearances are not things in themselves; they are only representations, which in turn have their object--an object which cannot be intuited by us, and which may, therefore, be named the non-empirical, that is, transcendental object = x. (A109) These phenomena are objects, objects that exist in space and time. The noumena, by contrast, are neither temporal nor spatial; space and time are forms of our intuition rather than realities that characterize the things in themselves. Noumena and phenomena, therefore, are distinct. Still further, we have experience only of the phenomena, not of the noumena: We have sufficiently proved in the Transcendental Aesthetic that everything intuited in space or time, and therefore all objects of any experience possible to us, are nothing but appearances, that is, mere representations, Further still, the phenomena, the world of stars and planets, trees and animals, depends on us for existence. The above passage continues: which in the manner in which they are represented, as extended beings, or as series of alternations have no independent existence outside our thoughts. (A491, B519) Elsewhere: That nature should direct itself according to our subject ground of apperception, and should indeed depend upon it in respect of its conformity to law, sounds very strange and absurd. But when we consider that this nature is not a thing in itself but is merely an aggregate of appearances, so many representations of the mind, . . . (A114) Now to assert in this manner, that all these appearances, and consequently all objects with which we can occupy ourselves, are one and all in me, that is are determinations of my identical self, is only another way of saying that there must be a complete unity of them in one and the same apperception. (A129) This is the more traditional way of understanding Kant, the way Kant was taken by his great successors. To put it briefly and all too baldly, there are two realms of objects; our experience is only of one realm, the realm of phenomena, which themselves depend on us for their existence; if we should go out of existence, so would they. That is because the phenomenal realm is somehow constructed by us out of the given, the data, the raw material of experience. The noumenal realm, however, is not thus dependent on us but is also such that we have no intuition, no direct experience of it. Finally, there is nevertheless a connection between the two worlds in that something like a causal transaction between the noumena and the transcendental ego (itself a noumenon) produces in us the given out of which we construct the phenomenal world. Call the above the two-world picture; this has been the dominant interpretation. There has always been another basic interpretation of Kant, however, one that more recently has perhaps achieved majority status. According to this other picture, there really aren't two worlds after all, a world of phenomena and underlying it another world of noumena. There is only one world and only one kind of object, but there are (at least) two ways of thinking about or considering this one world. All objects are really noumenal objects, and talk about the phenomena is just a picturesque way of talking about how the noumena, the only things there are, appear to us. The phenomena-noumena distinction is not between two kinds of objects but, rather, between how the things are in themselves and how they appear to us. So, for example, Graham Bird: Such phrases [e.g., transcendental objects and empirical objects'] should be understood to refer not to two different kinds of entity, but instead to two different ways of talking about one and the same thing. [18] And Michael Devitt: It is tempting to equate an appearance with the foundationalist's sense datum, taking the thing-in-itself as the unknowable external cause of this mental entity. Kant's writing often encourages this temptation. Nevertheless, scholars seem generally agreed--and have convinced me--that this two-worlds interpretation is wrong. What Kant intends is the following influential, but rather mysterious, one world view. An Appearance is not a mental sense datum, but an external object as we know it. In contrast the thing-in-itself is the object independent of our knowledge of it; it is not a second object, and does not, indeed could not, cause an appearance. . . . [19] Although this second picture is perhaps now the majority opinion, it seems a bit difficult to reconcile it with Kant's own view that his thought constituted a revolution--his famous second Copernican revolution. [20] After all, much of this second picture would be accepted even by such staunch prerevolutionaries as Aristotle and Aquinas. Both would agree that there is or can be a difference between the world (or any less impressive object) as it is in itself and the world as it appears to us; this is to admit no more than that we can be mistaken about the world or things in the world, and of course Aristotle and Aquinas would hardly deny that. Both would agree to something much stronger: that the world might have many properties of which we have no conception, so that our way of thinking about the world, the properties we ascribe to it, are not necessarily all and only the properties it has. For Aquinas or any other theist, this would be close to a truism: God, obviously enough, has many properties we don't know about, and presumably many of which we could not so much as form a conception. The essential elements of the one-world view seem perhaps a bit too uncontroversial, at least with respect to Kant's predecessors, to constitute a revolution, Copernican or otherwise. According to Merold Westphal: Finally, all twelve categories insofar as they constitute the world of human experience and are not merely formal features of judgment, are schematized with an essential reference to time. Thus the object and property that would disappear from the world in the absence of human knowers are not object and property per se, but substance and accident as defined by human temporality. Similarly, the truth and falsity that would disappear derive from the categories of reality and negation as essentially linked to our experience of time. Thus we are back to the tautology that in the absence of human cognition the world as apprehended by human minds would disappear. [21] That does, indeed, seem to be a tautology, or at least a trivially necessary truth; we could add that in the absence of bovine cognition, the world as apprehended by bovine minds would disappear. But how could Kant think of this as constituting a revolution, one according to which objects must conform to our minds (rather than, as previously thought, our minds to objects) if we are to have knowledge? Could a tautology constitute a revolution? 1. The One-World Picture and Reference to the Noumena Our main interest here does not lie in trying to resolve the question of what Kant intended: that is perhaps necessarily beyond our powers. Instead, we are looking to see if there is good reason, either given by Kant or constructible from materials given by him, for the conclusion that our concepts do not apply to God. And how does the difference between these two interpretations of Kant bear on this question? Consider the second picture first, and note that on this picture, if our concepts apply to anything, they apply to the Dinge, those being the only things there are. Similarly, if we manage to refer to and think about anything at all, we succeed in referring to and thinking about the Dinge, because they are all that there is. So how could it be that the categories and our other concepts do not apply to them? Well, what is it for a concept to apply to something, for something to fall under a concept? Consider the concept being wise. That concept applies to something (a thing falls under that concept) only if that thing is wise, only if, that is, it has the property of being wise. Properties and concepts are thus correlative. I have the concept being wise only if I grasp, apprehend, understand the property being wise. I have the concept being a prime number if and only if I grasp or apprehend the property being a prime number. For each property or attribute of which I have a grasp, I have a concept. Of course there are properties of which I have no concept. Small children often lack the concept of being a philosopher; that is to say, they have no grasp of the property being a philosopher. Large philosophers often lack the concept of being a quark; that is to say, they have no grasp of the property being a quark. No doubt there are properties none of us human beings grasps. One further familiar fact about properties and concepts: they have negations or complements. There is the property being red; there is also its complement, which, naturally enough, is being unred, not being red. There is the property of being wise but also the property of being unwise, failing to be wise. So if one of my concepts (e.g., being wise) does not apply to a thing, then the complement of that concept (being nonwise, not being wise) does apply to it. Perhaps you want to point out that this way of putting the matter presupposes that there are negative properties, such properties as being nonred, being unwise, and the like; you might object that in fact there are only positive properties, not negative ones. (You might also object to disjunctive and conjunctive properties.) This is no place to try to settle that issue. Clearly, there is the concept of a thing's failing to be wise (I know what it is for a thing not to be wise), even if there is no negative property nonwisdom. So if you object to negative properties, say that a thing falls under the concept nonwisdom just in case it does not fall under the concept wisdom; more generally, for any property P, a thing falls under the concept P if and only if it has the property P; it falls under the concept not-P if and only if it does not fall under the concept P. Given this elementary lore about concepts and properties, how could it be that the categories and our other concepts do not apply to the Dinge? Take the categories first--the category of causality, for example. What would it mean to say that this category does not apply to the Dinge? So far as I can see, what this would mean is that the noumena do not stand in causal relations to each other or anything else. Consider the property stands in causal relation to something; if the category of causality does not apply to the noumena, then it must be that none of them has that property. So our concept standing in causal relation to something wouldn't apply to things as they are in themselves. It follows, however, that the complement of that category or concept would apply to things as they are in themselves: each of them would be such that it does not stand in a causal relation to anything else. The same would go for our other concepts. On this way of thinking of the matter, our positive' concepts, you might say, do not apply to things as they are in themselves, which is really to say that there is no positive property we grasp that characterizes a thing as it is in itself. As it stands, however, this needs more work: there are problems about this distinction between positive and negative properties. There are also problems of other sorts: what about such positive properties as being self-identical, for example? Are we to suppose the Dinge are not self-identical? Well, perhaps these matters can be straightened out. (See chapter 2, p. 48.) For present purposes, what we need to see is that on this way of thinking, it would not really be the case that our concepts fail to apply to God in such a way that we cannot refer to and think about him. What would follow, given that he is a noumenon (of course, in this way of thinking, everything is a noumenon), is that God would not have any of the positive properties of which we have a grasp. It would not be the case that we couldn't refer to God and predicate properties of him: we could perfectly well do so, but we would be mistaken if we predicated of him a positive property of which we have a grasp. Thus we would make a mistake if we said that God is wise, or good, or powerful, or loving. That would be because nothing is wise, good, powerful, loving, and the like. (On the one-world picture, the Dinge are all there is; so if positive properties can't be ascribed to the Dinge, they can't be ascribed to anything.) Here there would be nothing at all special about God; what holds for him also holds for everything else. But those theologians who suggest that Kant showed we cannot refer to and think about God presumably believe that Kant showed there was a special problem about God; they don't think that what Kant really showed is that we can't talk or think about anything. As Kaufman puts it in the passage I quoted above (p. 4), ?The central problem of theological discourse, not shared with any other language game,' is the meaning of the term God.'?? So Kant, taken this way, doesn't fill this particular bill; it doesn't give us a relevant way of seeing that our concepts do not apply to God. 2. The Two-World Picture and Reference to the Noumena Now suppose we consider the other main interpretation of Kant: the two-world picture. This is the more traditional way to understand Kant and still, perhaps, deserves the nod. (Here I am not interested in which picture most accurately represents Kant, but whether Kant, taken any plausible way, gives support to the idea that we cannot refer to and think about God. [22] ) On this picture, there are two disjoint realms: phenomena and noumena, the Dinge and the things of experience. To add another quotation: Accordingly, that which is in space and time is an appearance; it is not anything in itself, but consists merely of representations, which, if not given in us--that is to say, in perception--are nowhere to be met with. (A494, B522) Now when we think about the application of our concepts to the noumena, we see that this two-world picture divides into two subpictures. (a) The Moderate Subpicture. On the one way of thinking, (some of) our concepts apply to the things in themselves; we can think about them and refer to them, all right, but we can't have any knowledge of them. When we think about them, predicate properties of them, what we have is just speculation, mere transcendental schein, and we deceive ourselves if we think we have more. Our knowledge doesn't extend beyond experience; hence, it does not extend to the realm of the things in themselves. This would explain that bewildering variety and proliferation of metaphysical views Kant found so shocking. The reason, fundamentally, is really that all the metaphysicians have been just guessing, whatever their pretensions to apodictic conclusions and conclusive certainty. Our reason can't operate in the rarefied atmosphere of the noumena, and the result of trying to do so is a mere beating of wings against the void. Of course Kant also represents his own work in the Critique of Pure Reason as knowledge and as certain and conclusive. And in that Critique he seems to tell us a fair amount about the Dinge: that they are not in space and time, that the world of experience is (in part) a result of a causal transaction' [23] between the Dinge and the transcendental ego, and that the latter has no intellectual intuition into the former. So the picture isn't wholly coherent. Coherent or not, however, this picture doesn't even suggest that we cannot think about and predicate properties of God. What it suggests, instead, is that when we do, we are not on the sure path of knowledge but on some much more hazardous climber's trail of mere opinion. So the moderate subpicture, too, gives no aid and comfort to the claim that our concepts do not apply to God. (b) The Radical Subpicture. There is a more striking version of the two-world picture, however, on which we do get the result that we can neither refer to God nor predicate properties of him (call it the radical subpicture'). On both versions of the two-world picture, the appearances are distinct from the things in themselves. The appearances are objects; they exist; they are empirically real. But they are also transcendentally ideal. And what this means, in part, is that they depend for their existence on us (on the transcendental ego[s]) and our cognitive activity. We ourselves are both noumena and phenomena: there is both a noumenal self and an empirical self. The things in themselves somehow impinge on us (taken as transcendental ego), causing experience in us; there is a productive interaction between the transcendental ego and the Dinge (the other Dinge, since the transcendental ego is itself a noumenon), the result of which is experience, the manifold of experience. As it is initially given to us, this manifold of experience is a blooming, buzzing confusion with no structure. Perhaps it contains among other things what Kant calls representations' (Vorstellungen); these are of more than one kind, but among them might be phenomenal qualia, something like sense data, or Humean impressions and ideas. The manifold must be worked up' (Kant's term) and synthesized by the application of the categories and other concepts. Thus we impose structure and form on it, and in so doing we construct the phenomena, the appearances. So the phenomena, the things für uns, are constructed out of the manifold of experience. Well, how do we do a thing like that? How do we construct a phenomenon (a horse, let's say) from the manifold of experience? At this point, the radical subpicture diverges from the more pedestrian version of the two-world picture, for on the radical subpicture, we construct objects by applying concepts (representations, Vorstellungen) to the manifold. The world of appearance gets constructed by virtue of our synthesizing the manifold, which proceeds by way of our applying concepts--both the categories and other concepts--to the manifold. We can't perceive or in some other way witness this construction; Kant says we are largely unconscious of the activity whereby we structure the manifold and construct the phenomena. Still, it proceeds by way of the application of concepts to the blooming buzzing manifold of experience. This would require a way of thinking about concepts and their function that is very different from the way of thinking about them I outlined above (a way according to which a concept is fundamentally a grasp of a property). And Kant suggests a different way of thinking of concepts: he sometimes calls them rules. Kant says that the understanding is the faculty of concepts; it is the source of our concepts. But he also says of the understanding, ?We may now characterize it as the faculty of rules. . . . Sensibility gives us forms (of intuition) but understanding gives us rules? (A126, Kant's emphasis). And he goes on to say, Rules, so far as they are objective . . . are called laws. Although we learn many laws through experience, they are only special determinations of still higher laws, and the highest of these, under which the others all stand, issue a priori from the understanding itself. They are not borrowed from experience; on the contrary, they have to confer upon appearances their conformity to law, and so to make experience possible. Thus the understanding is something more than a power of formulating rules through comparison of appearances; it is itself the lawgiver of nature. (A127) I don't for a moment pretend that this passage or others that could be cited are easy to interpret. Still, the passage does seem to suggest that concepts are rules and rules are laws. What sort of rules and what sort of laws? Perhaps they are rules for synthesizing the manifold, rules for constructing the phenomena. This is the heart of the radical subpicture. Again, I don't mean to suggest that this is Kant's view, but some of what he says suggests it. (Some of what he says also suggests that it is false; that is part of his charm.) For example: ?What is first given to us is appearance. When combined with consciousness, it is called perception. . . . ? Interpretative difficulties abound; the basic idea, however, is that concepts are rules, rules for the synthesis of the manifold and the construction of phenomena. (They are also laws, laws whereby the phenomena are constructed from the manifold of experience.) These rules apply to portions or bits of experience and, by way of their application, the phenomena are constructed. A rule of this sort perhaps specifies that certain portions of the manifold are to be combined or thought together' as an object. So, for example, consider your concept of a horse: it instructs you to associate, think together a variety of representations, a variety of items of experience, thus unifying that bit of the manifold into an empirical object: a horse. It is a rule which would say something like: think that particular congeries of representations together as a unity. Now again, I don't mean to claim that this is a coherent picture or a coherent way of thinking about concepts; on the contrary, I believe that it is not. But note that if it is coherent, then (at least if all of our concepts have this function [24] and only this function) our concepts will not apply to the noumena. Consider the concept being a horse. Understood this way, this concept is a rule for constructing phenomenal objects out of the manifold of experience. Of course it does not apply to the noumena: it cannot be used to construct an object out of them; they are not given to us (experience, the manifold, is what is given to us), and in any event they aren't the sorts of things out of which phenomenal objects could be constructed. So it isn't just that the concept being a horse does not apply to the Dinge in the sense that none of them, as it happens, is a horse (all are nonhorses), for then the complement of that concept--being a nonhorse--would apply. But that concept doesn't apply either: it, too, is a rule for constructing objects from the manifold. It is another way of unifying, synthesizing the manifold. So thought of, a concept could no more apply to the Dinge than a horse could be a number. On the radical subpicture, therefore, our concepts surely wouldn't apply to God, if there were such a person. For God would be a noumenon. God would not be something we have constructed by applying concepts to the manifold of experience (God has created us; we have not constructed him.) So, on the radical subpicture, we can't refer to, think about, or predicate properties of God. This way of thinking clearly displays a deep incoherence: on this picture, Kant holds that the Dinge stand in a causal or interactive relationship with us, taken as transcendental ego(s); [25] and he also says that they are not in space and time. But on the radical subpicture, Kant (at least if his intellectual equipment is like that of the rest of us) should not be able to refer to the Dinge at all, or even speculate that there might be such things. He certainly shouldn't be able to refer to them and attribute to them the properties of being atemporal and aspatial, or the property of affecting the transcendental ego(s), thereby producing experience in them. He shouldn't be able to refer to us (i.e., us transcendental egos), claiming that we don't have the sort of godlike intellectual intuition into reality that would be required if we were to have synthetic a priori knowledge of the world as it is in itself. (On this picture, we might say, Kant's thought founders on the fact that the picture requires that he have knowledge the picture denies him.) If this picture were really correct, the noumena would have to drop out altogether, so that all that there is is what has been structured or made by us. The idea that there might be reality beyond what we ourselves have constructed out of experience would not be so much as thinkable. [26] __________________________________________________________________ [18] Kant's Theory of Knowledge (New York: Humanities Press, 1962), p. 37. [19] Realism and Truth (Princeton: Princeton University Press, 1984), p. 59. See also D. P. Dryer, Kant's Solution for Verification in Metaphysics (Toronto: University of Toronto Press, 1966), chapter 11, section vi; H. E. Matthews, ?Strawson on Transcendental Idealism,? Philosophical Quarterly 19 (1969), pp. 204-220; Henry Allison, Kant's Transcendental Idealism (New Haven: Yale University Press, 1983). I am indebted, for these references, to Karl Ameriks (?Recent Work on Kant's Theoretical Philosophy,? American Philosophical Quarterly 19 [1982], and ?Kantian Idealism Today,? History of Philosophy Quarterly 9 [1992]) and to James Van Cleve, Problems from Kant (New York: Oxford University Press, 1999). [20] ?Hitherto it has been assumed that all our knowledge must conform to objects. But also attempts to extend our knowledge of objects by establishing something in regard to them a priori, by means of concepts, have, on this assumption, ended in failure. We must therefore make trial whether we may not have more success in the tasks of metaphysics, if we suppose that objects must conform to our knowledge. We should then be proceeding precisely on the lines of Copernicus' primary hypothesis.? (Bxvii) [21] ?In Defense of the Thing in Itself,? p. 170. [22] Of course, I do not mean to suggest that the one- and two-world pictures as I present them are the only possible (or actual) interpretations of Kant; clearly, there are various complications and extensions of each. What I claim is that none of them offers aid and comfort to the claim that our concepts do not apply to God. [23] We need the scare quotes because Kant's official view is that the concept of causality doesn't apply to the Dinge. [24] As Karl Ameriks (private communication) reminded me, Kant's metaphysical deduction certainly seems intended to reveal concepts which are rules for judgments of any sort, whether limited to items of experience or not. [25] How many of those transcendental egos are there, anyway? Like many questions of Kantian exegesis, this question is vexed. Indeed, on the radical subpicture, it is more than vexed. If the category of number doesn't apply to the noumena, then there is presumably no number n, finite or infinite, such that the right answer to the question ?How many of those transcendental egos are there?? is n. [26] In addition, of course, there is the problem that it takes a great deal of effort to believe that we are really responsible for the existence of sun, moon, and stars, not to mention dinosaurs and other things, that (as we think) existed long before there were any human beings. __________________________________________________________________ B. Arguments or Reasons? Clearly, there are problems of coherence here. Suppose we ignore them for the moment: what kinds of reasons does Kant give for the contention that we can't think about, refer to, predicate properties of the Dinge? Or, if he gives no such reasons (perhaps because he thinks we can think about them), what sorts of reasons or arguments does his work suggest for that conclusion? This conclusion--that our concepts are really rules for synthesizing the manifold into phenomenal objects and that the only things we can think about are objects we ourselves have somehow constructed--is, to say the least, rather startling. Some pretty powerful arguments would be required. Argument for this view is distressingly scarce. It is extremely difficult to find much that could pass muster as an argument, or even as one of those ?considerations determining the intellect? John Stuart Mill sometimes gave when, as he conceded, he didn't have an argument. There is nothing here like the ontological or cosmological arguments for the existence of God, or Descartes' argument that a person is not identical with her body (but is, instead, an immaterial substance), or the argument for the conclusion that propositions, the things we believe and assert, are not contingent objects. [27] Perhaps one must think of the radical subpicture as a sort of hypothesis proposed as best explaining certain phenomena. More likely, those who urge it are simply overwhelmed by what they see as its sheer intellectual beauty and power; they don't feel the need of argument. Indeed, they find the picture so dazzling they are willing to put up with a strong dose of incoherence in addition to absence of argument. Well, if you find the radical subpicture overwhelmingly attractive, then (incoherence aside) I guess you'll have to go with it. Then again, that doesn't constitute much of a reason for the rest of us--those of us more impressed by the incoherence of the picture than its beauty--to accept it. There is, however, a set of Kantian considerations that some might see as taking us partway to the conclusion. These are to be found in what he says about the antinomies: allegedly powerful arguments on both sides of a given question. Thus there is an allegedly compelling antinomical argument for the thesis that the world had a beginning in time, but an equally compelling argument for the antithesis that it did not. In the same way, there are compelling arguments for the theses that the world is composed of simples, that there is such a thing as agent causation (where an agent cause is a being that freely originates a new causal series), and that there is an absolutely necessary being; sadly enough, however, there are equally compelling arguments for the antitheses that the world is not composed of simples, that there is no such thing as agent causation, and that there is no absolutely necessary being. Here we seem to be in a nasty fix; we can prove four (everything in the Critique comes in fours) important theses, and for each of these four, we can also prove its denial. Now Kant apparently intends these antinomies to constitute an essential part of the argument for his transcendental idealism, the doctrine that the things we deal with (stars and planets, trees, animals and other people) are transcendentally ideal (depend upon us for their reality and structure), even if empirically real. We fall into the problem posed by the antinomies, says Kant, only because we take ourselves to be thinking about things in themselves as opposed to the things for us, noumena as opposed to mere appearances: If in employing the principles of understanding we do not merely apply our reason to objects of experience, but venture to extend these principles beyond the limits of experience, there arise pseudo-rational doctrines which can neither hope for confirmation in experience nor fear refutation by it. Each of them is not only in itself free from contradiction, but finds conditions of its necessity in the very nature of reason--only that, unfortunately, the assertion of the opposite has, on its side, grounds that are just as valid and necessary. (A421, B449) We solve the problem by recognizing our limitations, realizing that we can't think, or can't think to any good purpose, about the Dinge. In presenting the antinomies, Kant does not explicitly argue for the radical subpicture. But suppose we try to find something like an argument there, either for the radical subpicture or for the conclusion we have been deriving from the radical subpicture, the conclusion that our concepts do not apply to the noumena, so that we cannot refer to and think about them. Perhaps the premises would be: (2) If we are able to think about and refer to the Dinge, then the premises of the antinomical arguments (the premises of the arguments for the theses and for the antitheses) are about the Dinge and are all true, and (3) If those premises are all true, then the theses and antitheses would all be true, so that contradictions would be true. Naturally enough, however, (4) No contradictions are true. Therefore: (5) We cannot think about or refer to the Dinge. We could perhaps weaken the first premise (2) to make it a bit more plausible: (2*) If we can refer to and think about the Dinge, then each of the premises of the antinomical arguments will be about the Dinge and have overwhelming intuitive support. (This is weaker, of course, because it says, not that the antinomical premises are true, if we can think about the Dinge, but that they strongly seem true to us.) The second premise would then be: (3*) If each of the premises has overwhelming intuitive support, we will have overwhelming reason to accept each of the theses and antitheses, and we see that each thesis is contradicted by its antithesis. If, however, we weaken the first premise, we must strengthen one of the other two. Perhaps we could strengthen the third as follows: (4*) It couldn't be that we should have overwhelming reason to accept a proposition p and also its contradictory not-p. And the conclusion would be as before. Is it really true that (as (4*) claims) we couldn't have overwhelming reason to accept both a proposition p and also its denial not-p? [28] This would be an interesting inquiry but would take us too far afield; in any event, it isn't necessary for our present purposes, for there are at least two impressive problems with these arguments, one debilitating and the other fatal. I shall briefly outline the first and then look into the second in more detail. The first, the debilitating objection, is that even if we are not able to think of the noumena, we can think of the phenomena; and if the first premises of these arguments are true for the noumena, what is to prevent their being true for the phenomena as well? The two versions of the first premise ((2) and (2*)) of the argument claim the following: if it is true that we can think about the noumena, then the antinomical premises are about the noumena and either are true or have overwhelming intuitive support. Isn't it equally apparent that if we can think about the phenomena, then the antinomical premises are about the phenomena and are either true or have overwhelming support? If so, however, the argument would also prove that we can't refer to the appearances. What it would really prove, then, if it proved anything, is that we can't refer to or think about either noumena or phenomena. Because noumena and phenomena are all the things there are, the conclusion would be that we can't think about anything; and that seems a bit strong. Much more should be said about this objection to the argument, but I want to turn to the fatal objection. That is just that the antinomical arguments are not, to put the best face on it, at all compelling. Here I will argue this only for the premises of the first antinomy; exactly similar comments would apply to the others. In the first antinomy, there is an argument for the conclusion that ?The world had a beginning in time and is also limited as regards space? (A426, B454); this is the thesis. There is also an argument for the antithesis: ?The world has no beginning, and no limits in space; it is infinite as regards both time and space? (A426, B454). And the idea (in accordance with premises (2) and (2*)) is that if we can think about and refer to the Dinge, then both of these would be true or would have overwhelming intuitive support. Well, what is the argument? I am sorry to say it is hard to take seriously. The argument for the thesis goes as follows: If we assume that the world had no beginning in time, then up to every given moment an eternity has elapsed, and there has passed away in the world an infinite series of successive states of things. Now the infinity of a series consists in the fact that it can never be completed through successive synthesis. It thus follows that it is impossible for an infinite world-series to have passed away, and that a beginning of the world is therefore a necessary condition of the world's existence. (A426, B454) This argument proceeds by reductio ad absurdum: show that the denial of your conclusion leads to a contradiction, thereby proving your conclusion. The first premise is that if the world had no beginning in time, then at any point in time an infinite stretch of time would already have elapsed. This is dubious because it is at least abstractly possible that time and the world began together, some finitely many years (or seconds) ago. If so, then we should say that the world didn't have a beginning in time, although it did have a beginning with time. But let that pass. According to the second premise, ?the infinity of a series consists in the fact that it can never be completed through successive synthesis?; that is, it is characteristic of an infinite series that it can't be completed by starting from the beginning (or, more generally, some point only finitely far from the beginning) and adding things (events, say) one at a time (or more generally, finitely many at a time). This is true, provided the things (events) in question are added at a constant rate. If you start with the first event (or the nth, for some finite n) and add another event every second, you will never complete the series: at any subsequent time only a finite number of events will have occurred. According to current lore about the infinite, however, there is no bar of this kind to completing the infinite series in a finite time if the time taken for each event diminishes appropriately. For example, the first event takes one second to happen; the second event takes half a second; the third a quarter, the fourth an eighth of a second, and so on. At that rate, it won't take long at all for an infinite number of events to have elapsed--only a couple of seconds. But the real problem with the argument lies in a different direction. Kant points out that an infinite series can't be completed by starting from some point finitely far from the beginning and adding members finitely many at a time at a constant rate; fair enough. He then concludes, ?It thus follows that it is impossible for an infinite world-series to have passed away, and that a beginning of the world is therefore a necessary condition of the world's existence.? This doesn't follow at all. To claim that it does is to claim just what is to be proved: that the series in question had a beginning. The premise tells us that if you start from some finite point in a series--that is, some point finitely far from the beginning of the series--and add a finite number per unit time, then you will never complete the series. Fair enough; but if the world has existed for an infinite stretch of time, then there was no first moment, no first event, and no beginning either to the series of moments or the series of events; more generally, at any preceding moment an infinite time would already have elapsed. To conclude, as Kant does, that it is impossible that an infinite series of events has occurred is just to assume that the series in question had a beginning--that is, is finite--but that is precisely what was to be proved. So the argument really has no force at all. It is not as if it is an argument the premises of which have a certain limited amount of intuitive plausibility; it is rather that this transition to the conclusion completely begs the question by assuming what was to be proved: that the series in question has a beginning. The argument therefore fails to establish its conclusion; it merely assumes it. It therefore gives us no reason at all for accepting that conclusion. The argument for the antithesis is no more promising. Here is how Kant puts it: Let us assume that it [the world] had a beginning. Since the beginning is an existence which is preceded by a time in which the thing is not, there must have been a preceding time in which the world was not, i.e., an empty time. Now no coming to be of a thing is possible in an empty time, because no part of such a time possesses, as compared with any other, a distinguishing condition of existence rather than nonexistence. . . . (A427, B455) Again, the argument is by reductio: assume the denial of your conclusion and show that it is impossible, thereby establishing the conclusion. Here the two premises are (6) The beginning of an event or a thing is always preceded by a time in which the thing is not, that is, a time at which the thing in question does not exist. and (7) In an empty time (a time at which nothing exists) nothing could come to be, because there would be no more reason for it to come to be at one part of that empty time than at any other part of it. Neither premise is at all compelling. As to the first, this is true only if it is not possible that time and the world (the first event) should come into existence together, simultaneously. Is it known that this isn't possible? Certainly not. Indeed, some of the most popular theories of time (relational theories) would assume, not merely that this is possible, but that it is true. As for the second premise, it is equally unpromising. Suppose (in accord with the picture governing the argument) an infinity of time had elapsed before the first event of the world took place--before its creation, say. The objection is that there would have been no more reason for God to create the world at one moment than at any other; hence he wouldn't or couldn't have created it at any moment at all. Again, why believe this? If God proposed to create the world, and no time was more propitious than any other, why couldn't he just arbitrarily select a time? [29] This argument is like those arguments that start from the premise that God, if he created the world, would have created the best world he could have; they go on to add that for every world God could have created (weakly actualized, [30] say) there is an even better world he could have created or weakly actualized; therefore, they conclude, he wouldn't have weakly actualized any world at all, and the actual world has not been weakly actualized by God. Again, there seems no reason to believe the first premise. If there were only finitely many worlds among which God was obliged to choose, then perhaps he would have been obliged, somehow, to choose the best (although even this is at best dubious). [31] But if there is no best world at all among those he could have chosen (if for every world he could have chosen, there is a better world he could have chosen), why think a world's failing to be the best is sufficient for God's being unable to actualize it? Suppose a man had the benefit of immortality and had a bottle of wine that would improve every day, no matter how long he waits to drink it. Would he be rationally obliged never to drink it, on the grounds that for any time he might be tempted to, it would be better yet the next day? Suppose a donkey were stranded exactly midway between two bales of hay: would it be rationally obliged to stay there and starve to death because there is no more reason to move to the one bale than to the other? The arguments for the other antinomies don't fare any better. In no case is there anything like a conclusive argument (given the assumption that we are thinking about the Dinge) for either the thesis or the antithesis. In some cases, we may not know or be able to tell which (thesis or antithesis) is true: but that doesn't constitute much of an argument for the conclusion that we can't think about the noumena. What would be needed for the argument to work would be a really powerful argument for the thesis and an equally powerful argument for the antithesis. In none of these cases do we have something like that. Suppose we think a bit further about antinomies and paradoxes in connection with this question of concluding that we simply can't think about a given area or topic. Consider the Russell paradoxes, in their simple set-theoretical guise. Like Frege, we are all initially inclined to think that for every condition or property, there exists the set of just those things that meet the condition or have the property. It is pointed out that there is such a property as being nonselfmembered, the property a thing has just if it is not a member of itself; hence there must be a set S of nonselfmembered sets, but then S is a member of itself if and only if it is not a member of itself, which is a contradiction. Here it would be unduly enthusiastic to conclude that we can't really think and talk about sets as they are in themselves and can instead think only about sets that we have ourselves constructed, sets as they appear to us. One takes the argument as proving only that there is no set of nonselfmembered sets and that, contrary to appearances, it is not true that for every property or condition, there exists the set of just those things satisfying the condition or displaying the property. Take, instead, the Russell paradox as specified to properties, rather than sets; in some ways, this is a more serious paradox. One is initially inclined to think that there are properties, that some properties (for example, the property of being a property) exemplify themselves, so that there is such a property as self-exemplification, and that every property has a complement. These together lead to trouble: they imply that there is such a property as non-self-exemplification, which inconsiderately both does and doesn't exemplify itself. [32] Once again, however, it hardly seems to follow that we simply can't think and talk about properties an sich. We needn't hold that if we can think about properties an sich, then there is a property that both does and doesn't exemplify itself. We can quite properly conclude, instead, that one of the group of propositions we are initially inclined to accept must be false, and we look for the one with the least intuitive warrant or support, the one we are least strongly inclined to believe. (We might be inclined to think, for example, that there really isn't such a property as non-self-exemplification [even though it seems as if there is] so that either there is no such property as self-exemplification, or it is false that every property has a complement.) This is mildly disquieting, and gives us reason for a bit of humility with respect to the deliverances of reason, but we certainly aren't forced into the position of holding that we can't refer to and think about properties an sich. In what conditions would this drastic conclusion be right? Perhaps in none at all, and if in some, it is hard to say which. At the least, however, it would involve our being very strongly inclined to accept each member of a set of propositions about some subject matter, which set (by argument forms we are very strongly inclined to accept) entails a contradiction. It would also involve there being several such sets of propositions about the subject matter in question. Each of the premises and arguments involved would have to have very powerful, maximal or near maximal intuitive support; otherwise, we could more reasonably hold that a premise (or argument form) with only moderate intuitive support is false (or invalid). If there were several such sets of propositions--about properties, say--and each of these propositions and argument forms had the degree of intuitive support enjoyed by, say, 2 + 1 = 3 and modus ponens, then perhaps the right conclusion to draw would be that either there simply aren't any such things as the objects in the alleged realm, or that if there are, we are incapable of thinking about them. Even here, however, there would be reason to doubt the success of the argument. It would involve as a premise something like: (8) If there are several sets of premises about properties, each member of each set having maximal intuitive warrant, and the members of each set together entail a contradiction, then we cannot refer to and think about properties an sich. The next premise would be the antecedent of (8), and the conclusion would be the consequent of (8)--that is, the proposition that we cannot refer to and think about properties an sich. But if that conclusion were true, how could we grasp (8), the first premise? That premise seems to be, among other things, about properties an sich, and if we grasp it, we are able to think about properties an sich. The argument appears to be self-referentially self-refuting: if it is a successful argument, its first premise is both about noumena and such that we can grasp it, in which case that premise must be false. The sensible Kantian conclusion, so it seems to me, is that if, indeed, we can refer to and think about the Dinge, reason alone doesn't tell us such things as whether the world had a beginning in time or whether there are simple substances. It seems more likely than not, perhaps, that there are simple substances and that there are free agents who initiate new causal chains in the world, but the negations of these propositions are not demonstrably mistaken. Most certainly, it is not the case that both these propositions and their denials are demonstrable, so that each is both demonstrably true and, furthermore, demonstrably false. We must also recall that the whole scheme, the whole radical subpicture, seems incoherent in a familiar way. One who states and proposes this scheme makes several claims about the Dinge: that they are not in space and time, for example, and more poignantly, that our concepts don't apply to them (applying only to the phenomena), so that we cannot refer to or think about them. But if we really can't think the Dinge, then we can't think them (and can't whistle them either); if we can't think about them, we can't so much as entertain the thought that there are such things. The incoherence is patent. Would it be possible to induce coherence by refusing to make the distinction between phenomena and noumena, speaking only of what, if we did make that distinction, would be the phenomena, and claiming that whatever there is, is either a bit of experience or an object constructed by us from bits of experience by way of concepts (i.e., rules for constructing things from experience)? That is extremely hard to believe: are the stars, for example, which, as far as we can tell, existed long before we did, either bits of human experience or objects constructed by us from bits of human experience? How are we supposed to make sense of that? On this view, furthermore, the objection to Christian belief would not be that serious Christians improperly take it that they can refer to God; the objection would be that there is no God. If there were such a person, he certainly wouldn't be either a bit of human experience or something we have constructed from it. Still further, on this picture we ourselves (because we are among the things there are) would either have constructed ourselves from bits of experience or we would just be bits of experience; but of course we couldn't have constructed ourselves before we existed, so we must have started off, at least, as bits of experience with the power to construct things. Not a pretty picture. And even if we could somehow induce coherence here, why should we feel obliged to believe it? What possible claim could such a bizarre scheme have on us? By way of conclusion then: it doesn't look as if there is good reason in Kant or in the neighborhood of Kant for the conclusion that our concepts do not apply to God, so that we cannot think about him. Contemporary theologians and others sometimes complain that contemporary philosophers of religion often write as if they have never read their Kant. Perhaps the reason they write that way, however, is not that they have never read their Kant but rather that they have read him and remain unconvinced. They may be unconvinced that Kant actually claimed that our concepts do not apply to God. Alternatively, they may concede that Kant did claim this, but remain unconvinced that he was right; after all, it is not just a given of the intellectual life that Kant is right. Either way, they don't think Kant gives us reason to hold that we cannot think about God. __________________________________________________________________ [27] See my Warrant and Proper Function (WPF), pp. 117ff. [28] It seems we could have good reason to accept each member of a set S of beliefs such that there is no possible world in which all the members of S are true (the conjunction of the members of S is impossible), as is shown by the paradox of the preface. I write a book, of course believing every proposition asserted therein. Past experience and self-knowledge, however, lead me to think that very likely the book contains at least one false statement. (All of my previous books, as I've discovered to my sorrow, contain false statements.) In the preface, therefore, I sadly concede that at least one statement in the book is false. The total set of my beliefs, therefore--the statements in the book plus the statement that at least one statement in the book is false--is such that it must contain at least one falsehood; nevertheless, I have good reason to accept each member. [29] Compare Augustine's answer to those who wanted to know what God was doing before he created the world in The Confessions of St. Augustine, tr. Rex Warner (New York: New American Library, 1963), book 11, chapter 12, pp. 265-66. [30] For the notion of weak actualization, see my The Nature of Necessity (Oxford: Clarendon Press, 1974), p. 173, and Alvin Plantinga, ed. James Tomberlin and Peter van Inwagen (Dordrecht: D. Reidel, 1985), p. 49. [31] See Robert Adams's ?Must God Create the Best?? Philosophical Review 81 (1972), pp. 317-32. [32] If you balk at such properties as self-exemplification and non-self-exemplification, conduct the argument instead in terms of conditions; see Tomberlin and van Inwagen, Alvin Plantinga, p. 320. __________________________________________________________________ [16] Critique of Pure Reason, tr. Norman Kemp Smith (New York: St. Martin's Press, 1965), Preface to second edition, Bxxx, p. 29: ?I have therefore found it necessary to deny knowledge in order to make room for faith? (Kant's emphasis). [17] See Merold Westphal, ?In Defense of the Thing in Itself,? Kant-Studien 59/1 (1968), pp. 118ff. __________________________________________________________________ __________________________________________________________________ 2 Kaufman and Hick Our subject is the de jure question about Christian belief: the question whether it is rational, or reasonable, or rationally justifiable, or intellectually defensible to accept such belief. A previous question, as we saw in the last chapter, is the question whether there is any de jure question about Christian belief or, indeed, any de facto question either. Christian belief is belief, among other things, in the existence of God. And Christians believe that God is infinite: unlimited with respect to such important properties as knowledge, wisdom, goodness, and power. They also believe that God is transcendent: distinct from the created universe, in no way dependent on it, and such that it is dependent on him. Finally, they assume that it is possible to refer to God, talk and think about him, address him in prayer, and worship him. Many contemporary theologians, however, apparently believe that these ideas are excessively naive: they hold that there are profound problems in the very idea that we can refer to and think about a being characterized in the way Christians characterize God. In particular, they seem to believe that Immanuel Kant gave us excellent reason to be (at best) extremely suspicious of such naively realistic ways of thinking about God or religious language. As we saw in the last chapter, however, there is really nothing in Kant to suggest that in fact we can't think or talk about God. More generally, it is exceedingly hard to see how to construct an argument--an argument for the conclusion that we cannot refer to and think about God--from materials to be found in the work of Kant. Of course that doesn't show that no such argument can be found: but if one can be found, it is, I should say, up to those who think there is one to produce and develop it. In this chapter, I shall pursue this question into the present: if Kant gives us no reason to accept this conceptual agnosticism, do contemporary theologians (or writers in religious studies) do so? I choose two representatives: Gordon Kaufman and John Hick. __________________________________________________________________ I. Kaufman __________________________________________________________________ A. The Real Referent and the Available Referent According to Gordon Kaufman, The central problem of theological discourse, not shared with any other ?language game,? is the meaning of the term ?God.? ?God? raises special problems of meaning because it is a noun which by definition refers to a reality transcendent of, and thus not locatable within, experience. A new convert may wish to refer the ?warm feeling? in his heart to God, but God is hardly to be identified with this emotion; the biblicist may regard the Bible as God's Word; the moralist may believe God speaks through men's consciences; the churchman may believe God is present among his people--but each of these would agree that God himself transcends the locus referred to. As the Creator or Source of all that is, God is not to be identified with any particular finite reality; as the proper object of ultimate loyalty or faith, God is to be distinguished from every proximate or penultimate value or being. But if absolutely nothing within our experience can be directly identified as that to which the term ?God? properly refers, what meaning does or can the word have? [33] So the claim is that God is not to be identified with any particular finite reality--on the grounds, presumably, that God is not in fact identical with any particular finite reality. From the Christian perspective, this is, of course, no more than the sober truth: God is infinite and therefore not identical with any finite reality. So far, so good. Kaufman apparently infers from this, however, that ?absolutely nothing within our experience can be directly identified as that to which the term God' properly refers?; he adds that if this is so, then there is a real problem for the reference of our term God': if ?nothing within our experience can be directly identified as that to which the term God' properly refers, then what meaning does or can the word have?? I realize this last is a question, but it looks like a rhetorical question; the idea is that if nothing within our experience can be directly identified as that to which the term God' properly refers, then the term God' doesn't refer to anything, or at any rate there is a real problem about its referring to something. Here, therefore, we have two claims: (a) if God is not a finite reality, then absolutely nothing within our experience can be directly identified as that to which the term God' properly refers. and (b) if nothing within our experience can be directly identified as that to which the term God' properly refers, then the term God' doesn't refer to anything, or at least it is problematic that it does. These claims awaken Kantian echoes--echoes that get stronger as we move further into Kaufman's thought. And surely both are initially dubious. Consider (a). First, we must ask what it means to say that ?nothing within our experience can be directly identified as that to which the term God' properly applies.? What is it, as Kaufman is thinking of it, for something to be within our experience, and to be such that it can be directly identified as that to which a certain term properly applies? What about my friend's cat Maynard: is Maynard something within our experience which can be directly identified as that to which the term Maynard' properly applies? I should think so: else the problem is not merely with reference to God, but with reference to anything at all; Kaufman's suggestion, I think, is that the problem is specifically with respect to God. According to (a) it is because God is infinite that the term God' doesn't properly apply to anything within our experience. Now why, precisely, is that true? Maynard, I take it, is something within our experience, and this is because we can experience Maynard. We can perceive him: we can see, hear, touch, and sometimes smell him. The idea must be, then, that if God is not a finite reality, then we cannot experience him; we cannot perceive him (we cannot see, hear, or touch him) or in any other way experience him. An infinite being--one that is omnipotent and omniscient, for example--cannot be perceived or experienced in any way whatever. Is that really true? How does the fact that God is infinite mean that we cannot experience him? Many Christians and Jews believe that God spoke to Moses from the burning bush; Moses heard him. He spoke to Abraham in a dream. He spoke to several people when he said, ?This is my beloved Son in whom I am well pleased?; these people all heard him. Christians may also believe that the Holy Spirit works in their hearts, producing conviction and faith, as well as the religious affections of which Jonathan Edwards spoke; are they not then experiencing God? The term experience' (taken as either a noun or a verb) is notoriously slippery, but if these things do in fact happen, do not the people involved experience God? Christians may go still further and hold that in some circumstances some people perceive God, a theme that has received explicit and powerful treatment in William P. Alston's Perceiving God. If they are right, then in these cases too they experience God. Now Kaufman apparently thinks the fact that God is infinite--unlimited along several dimensions--means that these people are mistaken: whatever they think, they do not experience God. Again, why so? God is infinite with respect to power, that is, omnipotent: how does that so much as slyly suggest that God cannot make himself heard or that he cannot be experienced? He is infinite with respect to knowledge, that is, omniscient; does that somehow show that he could not speak to Abraham or anyone else? Is it perhaps the combination of omnipotence and omniscience that shows this? It is certainly hard to see how. If God is omnipotent, infinitely powerful, won't he be able to manifest himself in our experience, bring it about that we experience him? He will be unable to do so, presumably, under those conditions, only if it is logically impossible (impossible in the broadly logical sense) that an omniscient and omnipotent being should be able to make himself heard. But so far as I can see, there isn't even the slightest reason to think that; certainly Kaufman gives us none. I will go into the question of the nature of experience of God in more detail in chapters 6, 8, and 9; here I only want to point out that it seems initially implausible to declare that God, if he is infinite and omnipotent, could not bring it about that we experience him. The second premise (b)--the claim that if nothing within our experience can be directly identified as that to which the term God' refers, then the term God' doesn't refer to anything (or it is at least problematic that it does)--also seems dubious. Cosmologists tell us of the Big Bang, an event that occurred several billion years ago in which an explosion of enormous energy caused an expansion from an initial configuration of enormous density. I suppose the Big Bang is not something within our experience, something that can be directly identified as that to which the term the Big Bang' correctly refers; does it follow that there is a profound problem with this term? Is the real problem with contemporary cosmology not just the speculative nature of those suggestions about many universes and what happened during Planck time, but rather the very idea that we can refer to and think about that initial Big Bang? It isn't easy to see why: at the least, a powerful argument would be required. And if there is no particular problem here, why is there a special problem in the case of God? [34] Well, then, someone might say, if there is no problem about referring to an infinite being, how do we refer to God? In chapter 1, I suggested that we could do so, first, by way of definite descriptions such as the creator of the heavens and the earth', the omnipotent and omniscient creator of the world', the divine father of our lord and savior Jesus Christ', the divine person who spoke to Abraham', the divine person I am presently experiencing,' and so on. Each of these descriptions will refer to something if there is exactly one thing exemplifying the properties mentioned in the description; if not, then the description will not refer. (If Christian belief is true, of course, then each of these terms does refer to something, indeed, to the same thing.) Furthermore, we can use the proper name God' to refer to the being denoted by those descriptions. That term can serve as a proper name, for me, of God, in several ways. For example, I might fix the reference' of the term God' by one of the above descriptions, such as the creator of heaven and earth'; if, indeed, just one person created the heavens and the earth, and if that person is also denoted by those other descriptions, then my name God' will be a proper name of the same being as that denoted by those descriptions. My name will be a proper name of a being who is omniscient, omnipotent, the creator of the world, the father of our lord and savior Jesus Christ, and the like. Under these conditions, my name God' will express an essence of that being. [35] Perhaps my name, introduced in that way, will not express the same essence of God as your name, introduced by way of a different description. Even so, however, they will express logically equivalent (even if epistemically inequivalent) essences of God. [36] Alternatively, I might not get my proper name of God by using a definite description to fix the reference and then officially baptize the thing to which the description refers: I might instead just catch the name, so to speak, from others. In fact, this is the more usual way. Proper names, like colds, are ordinarily caught from our associates. As a child, I hear talk of God, talk in which the name God' occurs; I pick up the name, tacitly or implicitly intending to use it to refer to the same being to which those from whom I get the name refer. If they do indeed succeed in referring to God by using that name, then so will I. (Here is another way in which the success of my noetic ventures depends on the success of similar ventures on the parts of those around me: see Warrant and Proper Function, pp. 77-78.) In any event, Kaufman holds that we can neither know nor experience what he calls the real referent' of the term God': The real referent for ?God? is never accessible to us or in any way open to our observation or experience. It must remain always an unknown X. . . . (GP 85) When Christians use the term God', therefore, they do not refer to the real referent of that term (but then why call it ?the real referent??). To whom or what (if anything) do they refer when they say such things as that God was in Christ, reconciling the world to himself, or that God created the heavens and the earth, or that God is our faithful and loving father? The answer, says Kaufman, is that when they say these things they are referring to the ?available referent? of the name God', and the available referent is an imaginative construct, something we have somehow created: For all practical purposes it is the available referent--a particular imaginative construct--that bears significantly on human life and thought. It is the ?available God? whom we have in mind when we worship or pray . . . it is the available God in terms of which we speak and think whenever we use the word ?God.? In this sense ?God? denotes for all practical purposes what is essentially a mental or imaginative construct. (GP 85-86) God is a symbol--an imaginative construct--that enables men to view the world and themselves in such a way as to make action and morality ultimately (metaphysically) meaningful. (GP 109) So the available God, the God whom we have in mind when we worship and pray, the being to which we refer when we use the term God'--this being is a human creation, an imaginative construct, something we ourselves have created. The view seems to be initially that there is this available referent, but also a real referent of the term God', a being with whom we have no noetic contact and about whom we cannot speak. Or rather, the view is, I think, that there might be a real referent, and that if there is, it is a being we cannot think about: This fact, that the God actually available to people is an imaginative construct, does not necessarily mean that God is ?unreal? or ?merely imaginary? or something of that sort. That question remains open for further investigation. (GP 86) Does this mean, then, that the conclusion is, after all, that God really does not exist, that He is only a figment of our imaginations? If those words are intended to put the speculative question about the ultimate nature of things, then, as we have seen, there is no possible way to give an answer. (GP 111) In essence, then, Kaufman's view in God the Problem appears to be the following. The term God' has an available referent: this is a human construction, something we have created; when we speak of God in worship or to him in prayer, it is this available referent about which (or to which) we are talking. Perhaps the term also has a real referent. If so, however, it transcends our experience and is hence something to which our concepts do not apply: a mere unknown X, to adopt Kaufman's Kantian terminology. Now I've already argued that there seems no good reason to hold this position. Here I must go on to add that there is excellent reason not to hold it. As it stands, the view is incoherent. First, the available referent': the suggestion is that when Christians pray and worship and speak about God, they are talking about the available referent. When they say such things, for example, as God created the heavens and the earth', they are really attributing this property--the property of having created the heavens and the earth--to the available referent. But the available referent is a human construct, and hence presumably did not exist before there were human beings. How then did it manage to create the heavens and the earth? Could it somehow do this before it existed? In any event, an imaginative construct, a symbol, a structure of meanings of some kind is just not the sort of thing that could create the heavens and the earth or, indeed, anything else. A symbol, an imaginative construct, may have properties: being a construct, for example, or being a symbol, or being appropriately used by human beings for such and such a purpose; it certainly won't have such properties as being omniscient or creating the world. I suppose it could be that Christians are confused: they think they are referring to and talking about something that created them, but the fact is they are referring to something they themselves have created. Is it really plausible to think they are as confused as all that, however? Those who believe there is no such person as God will see Christians as mistaken in thinking there is, and perhaps it is at least sensible to think them mistaken in that way. Is it really sensible to think them mistaken in such a way that they predicate the properties of God of a mere construct? Well, perhaps that could happen; but surely a strong argument would be required to make this even reasonably plausible. Say that a property P entails a property Q just if it is necessary in the broadly logical sense that everything that exemplifies P also exemplifies Q; and say that a concept C contains a property P if the property of which C is a grasp entails P. Then it is clear that a concept might contain such properties as being omniscient or having created the world (even if it couldn't exemplify them), and equally clear that the concept corresponding to the definite description the omniscient creator of the world' contains the properties being omniscient and being the creator of the world. Could it be that what Kaufman really means is not that Christians assert that the available referent--which is something like a concept containing salient properties of God--exemplifies those properties, but rather that it contains them? This too seems wrong. It is indeed true that certain concepts, including some associated with descriptions of God, contain those properties. When Christians make their characteristic claims, however, they are not merely saying such things as that the concept being the omniscient creator of the heavens and earth contains the properties being omniscient and being the creator of the heavens and earth. That would, of course, be true; it would also be wholly trivial. It wouldn't be at all distinctive of Christians or theists: even the most hardened atheist would agree that this concept contains those properties. What Christians claim entails rather that these properties are exemplified, that there really exists a being who has them. The above seems to be the literal construal of Kaufman's words; of course there are other possibilities in the neighborhood. Perhaps, for example, he thinks of the available referent not as a being with the properties Christians ascribe to God, but as something like a certain type with which those properties are associated. [37] This may seem a more sympathetic construal of Kaufman; I doubt that it really is. If Kaufman's claim is that Christians ordinarily worship that type, then his claim is outrageous in just the way I suggest. If his claim, however, is only that Christians believe they worship a being having the properties associated with the type but are possibly mistaken, then is his claim more than the uninteresting suggestion that Christians may be wrong about whether there is such a person as God? Now consider the real referent. The idea is that our concepts do not apply to the real referent, if indeed there is such a thing. It follows that this being is not wise, almighty, or the creator of the heavens and the earth. For consider our concept of wisdom. This concept applies to a thing just if that thing is wise. So a being to which this concept did not apply would not be wise, whatever else it might be. If, therefore, our concepts do not apply to the real referent of the term God', then our concepts of being loving, almighty, wise, creator, and redeemer do not apply to it, in which case it is not loving, almighty, wise, a creator, or a redeemer. It wouldn't have any of the properties Christians ascribe to God. And of course so far this is in accord with Kaufman's intentions. I suspect, however, that his official position has other consequences Kaufman does not intend. If this being, this real referent, is really such that none of our concepts applies to it, then it will also lack such properties as self-identity, existence, and being either a material object or an immaterial object, these being properties of which we have concepts. Indeed, it wouldn't have the property of being the real referent of the term God' or any other term; our concept being the referent of a term will not apply to it. The fact is this being won't have any properties at all because our concept of having at least one property does not apply to it. Kaufman's view seems to entail that there could be a being that had no properties, didn't exist, wasn't self-identical, wasn't either a material object or an immaterial object, and didn't have any properties. Taken strictly, therefore, Kaufman's position is incoherent. __________________________________________________________________ [33] God the Problem (Cambridge: Harvard University Press, 1972), p. 7. Henceforth GP. [34] One source of Kaufman's views here may be a sort of lingering allegiance to the ?Verifiability Criterion of Meaning? mentioned above (pp. 7-8): ?Since seemingly no clear experiential evidence can be cited for or against that to which the word God' allegedly refers, the question has been repeatedly raised whether all talk about him is not in the strict sense cognitively meaningless? (p. 8). As we saw in chapter 1, however, there is little to be said for the Verifiability Criterion. [35] See my The Nature of Necessity (Oxford: Oxford University Press, 1974), pp. 77ff. [36] See my ?The Boethian Compromise,? in American Philosophical Quarterly (1978). [37] See Nicholas Wolterstorff's so-far-unpublished From Presence to Practice; Mind, World, and Entitlement to Believe, chapter 1. __________________________________________________________________ B. The Function of Religious Language Perhaps it is for reasons like these that in more recent work, in particular The Theological Imagination, [38] Kaufman seems to have given up the real referent. Instead, he claims that ?it is an error to reify God into an independent being? (TI 38), that ?To regard God as some kind of describable or knowable object over against us would be at once a degradation of God and a serious category error? (TI 244), and that It is a mistake, therefore, to regard qualities attributed to God (e.g., aseity, holiness, omnipotence, omniscience, providence, love, self-revelation) as though they were features or activities of such a particular being. Rather, in the mind's construction of the image/concept of God, the ordinary relation of subject and predicate is reversed. Instead of the subject (God) being a given to which the various predicate adjectives are then assigned, here the descriptive terms themselves are the building blocks which the imagination uses in putting together its conception. . . . Contemporary theological construction needs to recognize that these terms and concepts do not refer directly to ?objects? or ?realities? or their qualities and relations, but function rather as the building blocks or reference points which articulate the theistic world-picture or vision of life. (TI 244) Why must we think these terms do not, in fact, denote an all-powerful, all-wise creator of the universe? As far as I can see, it is because Kaufman does not believe that there is any such thing: he thinks, so far as I can see, that the proper attitude toward this proposition is either disbelief or withholding, either atheism or agnosticism. Naturally enough, if there is no being of that sort, then none of our terms will denote a being of that sort. This is perhaps a surprising position for a theologian; a theologian who does not believe in God is like a mountaineer for whom it is an open question whether there are any mountains or a plumber agnostic about pipes: a beguiling spectacle, but hard to take seriously. [39] And why does he think there is no such person? Again, there is precious little by way of argument. He cites first ?the rise of a new consciousness of the significance of religious pluralism?; [40] second, he says in the same article, ?new theories about the ways in which cultural and linguistic symbolic or conceptual frames shape all our experiencing and thinking . . . have given rise in theologians to a new self-consciousness about the extraordinarily complex and problematic character of all so-called religious truth-claims,' including those that are made by Christian faith?; third, he refers to the traditional problem of evil but with a twist: Christians themselves are responsible for more of the evil the world displays than they would like to think. (This last, sadly enough, is true, and perhaps [to take just one example] part of the occasion for modern apostasy, in the West, was the unedifying spectacle of Christians at each other's throats in the sixteenth and seventeenth centuries.) It is an enormous leap, however, to the conclusion that probably there is no such person as God. In chapter 14, we'll examine the question whether evil constitutes a defeater for Christian belief, and in chapter 13 we'll do the same for the plurality of religions the world displays. As for the second suggestion--the claim that many now hold that ?cultural and linguistic symbolic or conceptual frames shape all our experiencing and thinking?--perhaps this claim is true: but if it casts doubt on all of our experiencing and thinking, thus including Christianity, doesn't it do the same for every other way of thinking, including the thought that it casts doubt on what we think? If so, it would seem to leave everything as it was, not functioning as a reason for being doubtful specifically about theism (or, indeed, anything else). One might expect someone who is atheist or agnostic about God to move away from religion altogether, viewing religious devotion and belief with something of a jaundiced or a pitying eye. This is not Kaufman's course. Instead, he argues that religious practice and devotion ?still has an important function to play in life.? This function, of course, is not that of putting us in touch with a being with the properties traditionally ascribed to God or that of enabling us to appropriate the salvation in Jesus Christ that God has promised us. Rather, this new function requires that theologians should construct or reconstruct the concept of God. Religious language is still important, but it should be recast so as no longer to involve a forlorn attempt to refer to a being who isn't there. Instead, it should be used to promote human flourishing, ?human fulfillment and meaning? (TI 34). The word God' is to be associated with a symbol or image or concept theologians construct; it is their job to reconstruct the concept or symbol God' in a way that is appropriate to our present historical situation. (Thus in Theology for a Nuclear Age, he suggests that in this modern nuclear age we should think of God as ?the historical evolutionary force that has brought us all into being.? [41] ) The word God', therefore, should no longer be thought of as referring to the all-powerful, all-knowing, all-loving person who has created the world; it is not to be thought of as referring to a person at all. Instead, this word is to be seen as a sort of symbol of certain states of affairs. For example, Christians have thought of transcendence as a property of God; Kaufman recommends that, in constructing the new symbol, we retain transcendence: [42] What seems to be at stake here is a claim that human individuals and communities need a center of orientation and devotion outside themselves and their perceived desires and needs if they are to find genuine fulfillment. (TI 35-36) God symbolizes that in the ongoing evolutionary historical process which grounds our being as distinctively human and which draws (or drives) us on toward authentic human fulfillment (salvation). . . . And ritualized devotion to God in religious cult as well as in the private disciplines of prayer and meditation still has an important function to play in life. (TI 41) More generally: ?God? is the personifying symbol of that cosmic activity which has created our humanity and continues to press for its full realization. Such a personification has a considerable advantage for some purposes over abstract concepts such as ?cosmic forces? or ?foundation for our humanity in the ultimate nature of things?: the symbol ?God? is concrete and definite, a sharply defined image, and as such it can readily become the central focus for devotion and service. . . . ?God? is a symbol that gathers up into itself and focuses for us all those cosmic forces working toward the fully humane existence for which we long. (TI 50) Speech about the Christian God as ?real? or ?existent? expresses symbolically this conviction that free and loving persons-in-community have a substantial metaphysical foundation, that there are cosmic forces working toward this sort of humanization. (TI 49) The Christian image/concept of God, as I have presented it here, is an imaginative construct which orients selves and communities so as to facilitate development toward loving and caring selfhood, and toward communities of openness, love, and freedom. (TI 48) The idea, so far as I can grasp it, seems to be this. Perhaps there is no such person as theists have traditionally believed in. Nevertheless, it is a good idea to continue to use the term God' and, in fact, to continue to utter many of the very same words and phrases and sentences as do those who believe in God; done properly, this will promote human flourishing. How, exactly? Perhaps as follows. We realize, first, that there is probably no such person as God. We are then free to select a concept/image God' and associate with it certain properties--existence and transcendence, perhaps--and use that symbol to symbolize such things as that the world is hospitable, to at least some degree, to distinctively human aspirations, goals, needs, and desires. We are to say such things as God is real', meaning that in fact there are forces in the world that contribute to human flourishing. (We should add, I suppose, that the devil is also real, thereby symbolizing that there are forces working against human flourishing.) We are to say God is independent of us,' meaning thereby that a community or person needs a focus of interest outside itself to flourish. (Perhaps we should add that We are justified by the suffering and death of Jesus Christ,' thereby symbolizing the fact that we do not always feel guilty, or God was in Christ, reconciling the world to himself,' thereby meaning that things are now more propitious for human flourishing than they have been at some times in the past.) And saying these things will itself promote human flourishing. Can we take any of this seriously? This is not a matter of pouring new wine into old wineskins: what we have here is nothing like the rich, powerful, fragrant wine of the great Christian truths; what we have is something wholly drab, trivial, and insipid. It is not even a matter of throwing out the baby with the bathwater; it is, instead, throwing out the baby and keeping the tepid bathwater, at best a bland, unappetizing potion that is neither hot nor cold and at worst a nauseating brew, fit for neither man nor beast. Furthermore, this rehashing of secularity under the guise of reconstructing' Christianity encourages dishonesty and hypocrisy; it results in a sort of private code whereby one utters the same phrases as those who accept Christian belief but means something wholly different by them. You thereby appear to concur with those who accept Christian belief; in fact, you wholly reject what they believe. You can thereby patronize the person in the pew (who has not reached your level of enlightenment) but without paying the cost of unduly disturbing her. The fact is such double-talk is at best confusing and deceptive, contributing only to misunderstanding, dishonesty, and hypocrisy. Wouldn't it be vastly more honest to follow the lead of, for example, Bertrand Russell, A. J. Ayer, Daniel Dennett, Richard Dawkins, or even Madalyn Murray O' Hair, declaring forthrightly that there is no God and that Christianity is an enormous mistake? __________________________________________________________________ [38] Subtitled Constructing the Concept of God (Philadelphia: Westminster Press, 1981; hereafter TI). See also his Essay on Theological Method (Missoula, Mont.: Scholars Press, 1975 and 1979). [39] Unhappily, this spectacle is not at present uncommon. Compare, for example, Don Cupitt, who has similar views sometimes expressed with a certain amiable dottiness: ?It is spiritual vulgarity and immaturity to demand an extra-religious reality of God? (p. 10 of his Taking Leave of God [New York: Crossroad, 1981]); he adds, ?The real external existence of God is of no religious interest? (p. 96). [40] ?Evidentialism: A Theologian's Response,? Faith and Philosophy (January 1989), p. 30. [41] Manchester: Manchester University Press, 1985, p. 43. [42] How do properties such as transcendence or aseity get related to those symbols we construct? Do we just draw up a list of properties and declare them associated with the term God'? It is far from easy to see how this is supposed to work, and Kaufman doesn't say. __________________________________________________________________ __________________________________________________________________ II. Hick John Hick's work is interesting both intrinsically and with respect to our topic; he too holds a view heavily indebted to Kant, and his view too can be put (with considerable qualifications) as the view that our concepts do not apply to God or The Real'. There are evocative echoes of Kant and also evocative echoes of some of the trials and tribulations dogging an effort to find a coherent interpretation or understanding of Kant. __________________________________________________________________ A. The Real The traditional doctrine of divine ineffability is to be found in Christianity, as well as other religious traditions. Hick believes that this doctrine is really the recognition of a quasi-Kantian distinction between God (the Real, the Ultimate [43] ) as it is in itself and as it is for us (as we know or experience it): In each of the great traditions a distinction has been drawn, though with varying degrees of emphasis, between the Real (thought of as God, Brahman, the Dharmakaya . . .) in itself and the Real as manifested within the intellectual and experiential purview of that tradition. (236) So far, so good; this claim--that there is a distinction between the Real as it is in itself and as it is for us--is relatively weak. It requires only that the way we think of God does not completely match what God actually is; it would be satisfied if, for example, there are things about God that we didn't know or, more strongly, if there were things about him we couldn't know. But Hick goes much further; the Real is such that we cannot say anything at all about it, in that none of our terms can be literally (and correctly) applied to it: Thus although we cannot speak of the Real an sich in literal terms, nevertheless we live inescapably in relation to it. (351) It is within the phenomenal or experienceable realm that language has developed and it is to this that it literally applies. Indeed the system of concepts embodied in human language has contributed reciprocally to the formation of the humanly perceived world. It is as much constructed as given. But our language can have no purchase on a postulated noumenal reality which is not even partly formed by human concepts. This lies outside the scope of our cognitive capacities. (350) This sounds like the two-world interpretation of Kant (above, pp. 10ff.). There is the phenomenal realm, to which our language literally applies; this ?humanly perceived world? is as much constructed as given, and it is constructed, in part, by virtue of our application of concepts. However, there is also a noumenal world (The Real'), ?which is not even partly formed by human concepts,? and as a result it is outside the scope of our cognitive capacities. And here some of the same questions arise as with the two-world interpretation of Kant: why think something is within the scope of our cognitive capacities only if it is partly formed by human concepts? Are horses and dinosaurs (partly) formed by our concepts? (Which parts?) And if the noumena lie outside the scope of our cognitive capacities, how is it that we know something about them, or even that there are any such things? More frequently, however, he adopts the one-world view: Kant distinguished between noumenon and phenomenon, or between a Ding an sich and that thing as it appears to human consciousness. . . . In this strand of Kant's thought--not the only strand, but the one which I am seeking to press into service in the epistemology of religion--the noumenal world exists independently of our perception of it and the phenomenal world is that same world as it appears to our human consciousness. . . . I want to say that the noumenal Real is experienced and thought by different human mentalities, forming and formed by different religious traditions, as the range of gods and absolutes which the phenomenology of religion reports. (241-42) Further, it is unclear whether Hick thinks we can or can't, do or don't perceive this being or in some other way experience it. On the negative side, we have If the Real in itself is not and cannot be humanly experienced, why postulate such an unknown and unknowable Ding as sich? The answer is that the divine noumenon is a necessary postulate of the pluralistic religious life of humanity. (249) On the positive side, we have, for example, Analogously, I want to say that the noumenal Real is experienced and thought by different human mentalities, forming and formed by different religious traditions, as the range of gods and absolutes which the phenomenology of religion reports. (242) and The noumenal Real is such as to be authentically experienced as a range of both theistic and nontheistic phenomena. (246-47) There are several more passages to quote on each side; clearly, Hick is ambivalent about the answer to this question. But perhaps this is not fatal; his answer, I should think, would be ?In a way, yes, and in a way, no.? The noumenal real makes a crucial causal contribution of some sort to our experience; perhaps it doesn't matter whether we say that we actually experience it or say, instead, only that it contributes to our experience. Another ambiguity, however, is not so easily dismissed. In chapter 19, Hick seems to say that our concepts do not apply to the noumenon, or, as he puts it there, none of our terms applies literally to it. He quotes the Buddha as saying, with respect to where or in what sphere a Tathagata (a fully enlightened being) arises after death, that none of the terms arises', does not arise', both arises and does not arise,' and neither arises nor does not arise' applies to the condition of the Tathagata (346). Hick apparently approves of this suggestion and adds, ?We have here the idea of realities and circumstances which transcend the categories available in our unillumined thought and language. Their total elusiveness is signaled by the Buddha's rejection not only of the straight positive and negative assertions but also of their combination and disjunction? (347). Hick also claims that ?we cannot speak of the Real an sich in literal terms? (351). If Hick really means that none of our terms applies literally to the Real, then it isn't possible to make sense of what he says. I take it the term tricycle' does not apply to the Real; the Real is not a tricycle. But if the Real is not a tricycle, then is not a tricycle' applies literally to it; it is a nontricycle. It could hardly be neither a tricycle nor a nontricycle, nor do I think that Hick would want to suggest that it could. In chapter 14, however, Hick makes a suggestion of quite a different kind. As he says, ?it would not indeed make sense to say of X that none of our concepts apply to it? (p. 239); for example, at least our concept being such that we can refer to it would have to apply to any X of which we were properly prepared to say anything at all, including that our concepts do not apply to it. The idea is rather, says Hick, that among our concepts, only formal concepts and negative concepts apply to the Real. That is to say, of the properties of which we have a grasp, only those that are formal, such as having some properties, being self-identical, and being such that 7 + 5 = 12, and those that are negative, such as not being a horse, not being a tricycle, and not being good, would apply to it. Hick adds that there is a substantial tradition within Christianity and other religions, according to which we should distinguish between what we might call substantial properties, such as being good', being powerful', having knowledge', and purely formal and logically generated properties such as being a referent of a term' and being such that our substantial concepts do not apply'. What they wanted to affirm was that the substantial characterizations do not apply to God in God's self-existent being, beyond the range of human experience. They often expressed this by saying that we can only make negative statements about the Ultimate. . . . This via negativa (or via remotionis) consists in applying negative concepts to the Ultimate--the concept of not being finite, and so on--as a way of saying that it lies beyond the range of all our positive substantial characterizations. It is in this qualified sense that it makes perfectly good sense to say that our substantial concepts do not apply to the Ultimate. (239) Here Hick is apparently endorsing what he sees these traditions as delivering. I am not sure there is any way of harmonizing chapter 14 with chapter 19; if not, I suggest we go with chapter 14. At some points in characterizing these traditions he is historically incorrect. For example, he claims that ?Calvin taught that we do not know God's essence but only God as revealed to us? (250), and he refers to Calvin's Institutes, I: xiii: 21. But Calvin doesn't teach that we can't know anything of God's essence. In this chapter, he begins by arguing that The scriptural teaching concerning God's infinite and spiritual essence ought to be enough, not only to banish popular delusions, but also to refute the subtleties of secular philosophy. [44] He goes on to point out that we can't measure' God by our own senses' as he puts it: But even if God to keep us sober speaks sparingly of his essence, yet by those two titles that I have used [infinite' and spiritual'], he both banishes stupid imaginings and restrains the boldness of the human mind. Surely his infinity ought to make us afraid to try to measure him by our own senses. (p. 121) Calvin's next point is that because God is a spirit, we can't properly attribute corporeal characteristics to him. He concedes that Scripture does seem to attribute such characteristics (a mouth, an arm, ears, eyes, hands) to him, but those who therefore take it that he has such bodily characteristics fail to understand that ?as nurses commonly do with infants, God is wont in a measure to lisp' in speaking to us? (p. 121). Here Calvin clearly thinks we know that God in himself' is infinite, spiritual, and incorporeal; his essence includes infinity and incorporeality. In the passage to which Hick refers, furthermore, Calvin's point is to caution us not to try to figure out God's essence by way just of the resources of reason; given its limitations, that is bound to prove futile: For how can the human mind measure off the measureless essence of God according to its own little measure, a mind as yet unable to establish for certain the nature of the sun's body, though men's eyes daily gaze upon it? . . . Let us then willingly leave to God the knowledge of himself. . . . But we shall be ?leaving it to him? if we conceive him to be as he reveals himself to us, without inquiring about him elsewhere than from his Word. (I, xiii, 1, p. 146) The point is that Scripture is a much better source of knowledge of God (including knowledge of his essence) than rational speculation. But Calvin didn't think for a moment that none of our positive substantial concepts applies to God; he clearly believed that God really is the creator of the heavens and the earth, that he really does love us, that he is incorporeal, wise, powerful, loving, and the like. On this view, Hick's claim about the Real is not that none of our concepts applies to it or that none of our terms literally applies to it; that is clearly incoherent. His claim, instead, is that only our formal concepts and terms and our negative concepts and terms apply to it. That is to say, the only properties it has of which we have a grasp are formal properties and negative properties. Consider first those formal concepts. Included here would be, first of all, concepts of properties which are such that everything has them and furthermore has them necessarily. [45] Hick is thinking (I take it) of properties such that it is necessary that everything has them: such properties as being self-identical, having properties, having essential properties, being either a horse or a nonhorse, and being such that 7 + 5 = 12. These properties are necessarily had by everything. We might add that they are essential to everything, where a property is essential to an object if it is not possible that the object exist but lack the property: the property of being self-identical would be an example. We could add still further that each of these properties is such that it is necessary that everything has it essentially. So take any of the properties under consideration: everything has it, it is necessary that everything has it, everything has it essentially, and it is necessary that everything has it essentially. Existence is another of those formal properties: everything exists, existence is an essential property of everything, and it is a necessary truth that existence is essential to everything. But these aren't the only properties that Hick means to include under the rubric formal'. Others are such properties as being referred to by human beings and being thought of by John. So the idea is not that we cannot talk or think about the being in question. On the contrary: we can think about it, refer to it, and say of it that it exists. We can say of it, furthermore, that we can refer to it. Second, in addition to formal properties, we can predicate negative properties of this being--that is, we can correctly predicate negative properties of it. This is implied by Hick's position as I have so far explained it. We can see this as follows. First, note that every property has a complement, where the complement of a property P is the property of not having P. Each of the properties of which we have a concept has a complement: the property of not having that property. Thus the complement of the property wisdom is the property of not being wise, a property enjoyed by everything that is not wise. And if we have a grasp, a conception, of the property in question (wisdom, for example) then we also have a grasp of its complement. Now consider any property P and its complement -P; the property P or -P is one of those formal properties every thing necessarily has. Of course, anything that has that property has either the property P or else has its complement -P. (For anything you pick, either it is wise or it is not wise.) According to Hick's position as so far explained, however, the Real doesn't have any positive nonformal property of which we have a grasp. It follows, then, that for all the positive properties P of which we have a grasp, the Real has -P. Here we are speaking of the properties the thing in question has, not about our abilities or lack thereof to know or warrantedly believe something or other about its properties. I say that every object has essentially the property of having P or -P for any property P; I say further that if a being has that property, then either it has P or it has -P. Still, a being might be known (or justifiably believed or warrantedly believed) to have P or -P without being known (justifiably believed, etc.) to have P and without being known to have -P. I do not know and have no view on the question whether Socrates ever owned a horse; nevertheless, either he did own a horse or he didn't. In developing Hick's view, then, I am taking for granted a sort of realism: the idea that things (some things) can be a certain way even if neither I nor any other human being knows whether they are that way or not. There is nothing in what Hick says to suggest that his position obliges him to take issue with this truism. So the idea is that we can predicate negative properties of this being. Furthermore, the idea is that we can correctly predicate negative properties of it: it has negative properties. Indeed, for each of the nonformal positive properties we grasp, the Real has the complement of that property (which is a negative property). Among our positive concepts, only the purely formal ones apply to this being; as for the rest of our concepts, only the negative ones apply to it. The Real doesn't have wisdom: therefore it has nonwisdom; it does not have love: therefore it has nonlove; and so on for all the other positive nonformal properties we grasp. But, you say: it is not possible that there be a being that has only formal and negative properties. No doubt that's true; still, it is neither here nor there, so far as Hick's claim goes. The being in question may very well have positive properties in addition to the formal properties; it is only required by Hick's position that those be positive properties of which we have no concept, properties of which we have no grasp. And we certainly don't know that there aren't any positive properties like that. We must ask two questions here: first, is this Hickian position coherent? And second, is there any reason to accept it? __________________________________________________________________ [43] An Interpretation of Religion (New Haven: Yale University Press, 1989), pp. 236-39. Unless otherwise noted, page references to Hick's work will be to this book. [44] Tr. Ford Lewis Battles and ed. John T. McNeill (Philadelphia: Westminster Press, 1960), I, xiii, 1, p. 120. (Page references to the Institutes are to this edition.) [45] A concept's meeting the first condition but not the second (i.e., being such that everything falls under it but not such that everything necessarily falls under it) is not sufficient for its being formal. For example, the concept either not living on the moon or else not being human applies to everything (there are no human beings who live on the moon), but it isn't a formal concept in the intended sense. __________________________________________________________________ B. Coherent? 1. Can There Be a Being with Only Formal and Negative Properties? This being, says Hick, has no positive, nonformal properties of which we have a concept; the only positive properties it has are those of which we have no grasp. This is not clearly incoherent. We can't just see, I think, that there couldn't be a being like that; that is because we have a very slim grasp of those properties we don't grasp. We just don't know enough about them to know that it isn't possible that there be a being like that. Of course, we also have no reason to think that there can be such a being: the fact that we can't see that there can't be a being like that is little or no reason for thinking we can see that there can be. (It is one thing to fail to see that something is impossible; it is quite another to see that it is possible.) In this case, so it seems, we don't know enough to be able to tell whether it is possible that there be such a being. So suppose we provisionally concede, at least for purposes of argument, that it is possible that there be such a being. But if there is such a being, how is it that we are able to refer to it, have some way of singling it out as a subject of predication? How can that be done? Not, to be sure, by way of the definite descriptions whereby Christians believe they can pick out God--such descriptions, say, as the all-powerful, all-knowing creator of the world'; such descriptions involve positive, nonformal properties of which we have a conception. Could we instead use the description: the being that has no positive, substantial properties of which we have a grasp'? No; for if there is one such being, then maybe there are several more, none with any positive, nonformal properties we can grasp but differing from each other in positive, nonformal properties we can't grasp. So we have no reason to think that that description will work either. (Of course once again we don't know that it doesn't work; for all we know or can tell, there is exactly one being with no positive nonformal properties of which we have a grasp.) Now Hick's idea, I think, is that those who practice the great religions really refer to this being (the Real, which has no positive nonformal properties of which we have a grasp) when (as it seems to them) they refer to God, Allah, Brahman, Shiva, Vishnu, the Dharmakaya, or whatever. So Christians think they refer to a being who is personal, loving, knowledgeable, and the like; the fact is, however, they do not refer to such a being, but to a being who doesn't have any of these properties or, indeed, any other positive properties of which we have a grasp. Is this really possible? Is it possible that we refer to a being, thinking it has properties P[1],. . .,P[n], when, in fact, it doesn't have any of those properties or any other positive properties of which we have a conception? This too is not at any rate clearly impossible. It can certainly happen that we refer to a being when we are very much mistaken about the properties it has. I have never met you; in a letter, I tell you that I am a world-class tennis player and an athlete of enviable talents; the fact is that I am a complete duffer at tennis and at every similar activity. I go on to claim that I have a tenor voice to rival Pavarotti's, have a Nobel prize in economics, am strikingly handsome, and write splendid poetry; all of this is whoppingly false. (In fact, I am unable even to appreciate any poetry above the level of William E. McGonagall, poet and tragedian, [46] know absolutely nothing about economics, can't sing a note, and am very plain.) Then I have few of the properties you ascribe to me; still, you can refer to me. Of course, there must be some kind of connection between us. You can't pick me out as that handsome tenor-cum-poet-cum-economist who lives in (say) Jamestown, North Dakota; that description doesn't apply to me. [47] However you can refer to me as the one who wrote you a letter claiming to be all these things (supposing you received only one such letter). And perhaps something similar would be so for the Real. How is the reference supposed to go? Well, presumably Hick's idea is that we can refer to the Real as the being that the practitioners of the great religions refer to when they think they are referring to beings with the properties ascribed to God, Allah, Brahman, Vishnu, and the like. Obviously, that just pushes the problem back a step: how do they refer to it? How does it happen that when Christians use the term God' they are, in fact, referring to this being that has no positive properties they grasp, despite the fact that they think they are referring to a being with a lot of positive properties of that sort? Again, there would have to be some connection between them and the Real. (It isn't the case, of course, that the practitioners of Christianity, say, hypothesize that there is a being with no positive properties of which we have a grasp to whom they refer when they think they are referring to an omnipotent, omniscient, and wholly good creator of the universe; that would make no sense.) So how do they refer to this being? Well, presumably this could happen only if they had some kind of experiential contact with it, experienced this being in one way or another (whatever precisely it means to say that one being experiences another). They think they are in contact with a being with the properties ascribed to God; they are mistaken, however--not in thinking they are in contact with something, but in thinking the something with which they are in contact has the properties they ascribe to God. Now perhaps this is possible: still, it does require a modification--and a significant modification--of Hick's position. If this is the way the wind blows, then the Real enjoys at least one positive nonformal property of which we have a conception: the property being experienced by us. It stands in at least one positive nonformal relation of which we have a conception: the relation being experienced by. (So the ambiguity we noted above, pp. 44-45, must be resolved in favor of the alternative according to which we do experience the Real.) And that may lead to more, for what is involved in something's being experienced by us or by the practitioners of the great religions? What is it for something to be experienced by us? Here there are several views. One is that the thing in question appears to us, in a way that defies further analysis. Another is that it causes us to be appeared to in a certain way, or causes some other kind of experience in us (and meets certain other conditions). What these have in common is at the least the idea that in order to experience the Real, we must be in causal contact with it, stand in a causal relation with it. There is perhaps one alternative to be found in the history of philosophy: that would be the idea that we could experience something, perhaps in an analogically extended sense of experience', if there were a preestablished harmony between experiential states of ours and states of the thing in question. [48] But that, too, would involve the thing in question's being in a causal relation to the thing or person (in Leibniz's thought, God) who arranges the preestablished harmony. Here, too, therefore, a causal relation would be required between the thing experienced and something else, so that here, too, that thing stands in causal relations and (perhaps at one or more removes) in causal relations to the experiencing subject. And this means that Hick or a Hickian (for perhaps we are going beyond Hick's position here) must also ascribe another positive nonformal property to the Real: the property of being causally connected with us human beings. This is not a merely formal property, and it is also not a negative property. Still further, it may involve additional properties: whatever properties are necessarily connected with standing in a causal relation to human beings. The thing in question could not, for example, be like numbers and propositions are ordinarily thought to be: abstract objects that are incapable of standing in causal relations. So the being in question must have the property of being a concrete object, as opposed to an abstract object. The property of being a concrete object is also a nonformal property (many things lack it); as we shall see below, it isn't easy to tell whether it is a positive or a negative property. And of course there may be still more properties necessarily connected with the property of standing in a causal relation with us human beings. 2. Positive versus Negative Properties Should we perhaps say that this last property--being a concrete object--is really a negative property? Why can't we think of the property being concrete as simply the complement of the property being abstract, with the latter positive and the former negative? Perhaps being concrete is really just not being abstract. But this leads to a real difficulty: why go that way? Can't we just as well take the property being abstract as the property not being concrete, so that being concrete is the positive property and being abstract the negative? How do we determine which of the two properties really is positive, and which is really negative? Indeed, are we guaranteed that this distinction between positive and negative properties really applies to properties at all? Is there really such a distinction for properties? Of course, there is a distinction between positive and negative predicates, linguistic items or phrases such as is a horse', is not a cat,' and the like. (Both not being abstract' and not being concrete' are negative predicates; both is concrete' and is abstract' are positive predicates.) But do we know that this distinction between positive and negative extends beyond predicates to properties? What makes a predicate (in English) negative is the presence in it of some negative particle, such as not' or non' or un' (as in unlimited') or a' (as in asymmetrical') or dis' or anti' (as in antidisestablishmentarianism'). But properties presumably don't contain particles or other bits of language. What would distinguish positive properties from negative properties? Is there really such a distinction? This is not an easy question to answer. Can the Hickian perhaps claim that he really owes us no further answer? There is a distinction between positivity and negativity for properties, he says, and there is no way of getting behind this distinction, to say what it consists in, what it is that makes a property positive, or anything of that sort. This distinction is rock bottom and cannot be explained in terms of anything else. There are clear examples: wisdom is a positive property, and its complement, unwisdom (enjoyed both by those things capable of but lacking wisdom and by those not capable of it), is clearly a negative property; the distinction itself is ultimate and can't be explained in other terms. Well, maybe so, but can we say anything general about which properties are positive and which negative? Presumably the idea is that (1) every property is either positive or negative, (2) every property has a complement, (3) the complement of a property P has the opposite sense from P (that is, the complement of a positive property is negative and the complement of a negative property is positive), (4) a property equivalent [49] to a given property has the same sense as that property, and (5) the Real has no positive properties of which we have a conception. (We have already seen that there must be exceptions to this last principle, but let that pass.) Furthermore, (6) no negative property of which we have a conception entails [50] a positive property of which we have a conception; else the Real would have those positive properties entailed by those negative properties of which we have a conception. What about conjunctive and disjunctive properties? A conjunctive property P & Q is negative if and only if both P and Q are negative. (If P & Q were negative and either P or Q were positive, the negative P & Q would entail a positive property, in which case the Real would have that positive property.) What about disjunctive properties? A disjunction P v Q of properties of which we have a conception could not be positive if either P or Q were negative: else the positive P v Q would be entailed by the negative P or the negative Q. So far as I can see, there is nothing problematic about (1)--(6) at the level of logic alone. Indeed, note that we can give a truth table' (actually, a positivity table') for the complement of a property and for disjunction and conjunction among properties, and note further that the positivity table for conjunction is the truth table for disjunction and the positivity table for disjunction the truth table for conjunction. Mapping disjunction for properties onto conjunction for propositions, and conjunction for properties onto disjunction for propositions, we can help ourselves to some results from propositional logic and see that the logic of properties generated by (1)-(6) is consistent, complete, decidable, and so on. Nevertheless, there still remains a problem with coherence. It appears, initially at least, that some nonformal positive properties are entailed by negative properties (given that there is such a distinction for properties). For example, according to Hick, the Real is both ultimate and infinite: Unlimitedness, or infinity, is a negative concept, the denial of limitation. That this denial must be made of the Ultimate is a basic assumption of all the great traditions. It is a natural and reasonable assumption: for an ultimate that is limited in some mode would be limited by something other than itself; and this would entail its non-ultimacy. (237-38) But what about this property of being ultimate? First, what is it to be ultimate? Well, at the least it is to be independent of all other beings, not depending on any other beings for existence or for intrinsic properties. And that sounds appropriately negative. Nevertheless, it entails the property of being self-sufficient; and that sounds positive. So it looks as if the negative property being independent of all others entails the positive property being self-sufficient. Of course, it is perhaps possible to bite the bullet and maintain that the property of being self-sufficient, contrary to appearances, is really negative. Well, perhaps we can live with that. But what about infinity? According to Hick, the property of being unlimited, being infinite, is a negative property: it is the complement of the positive property of being limited or being finite. (Here is another case where it is far from obvious, initially, which of the pair in question is positive and which negative; let's just concede for purposes of argument that being limited is a positive property.) Here we run into a real problem. What is this property of being unlimited? It is the negative property not being limited. Well, what is it to be limited? In the spatial analogue from which the notion is taken, it is to have limits or borders. A country that is unlimited, therefore, would have no borders and occupy all of space. (Again, having no borders sounds negative while occupying all of space sounds positive.) Now of course the idea is not that the Real is spatially unlimited and occupies all of space. But then how does the analogy apply? In the spatial analogue, there are two features: an unlimited country is unlimited in a certain respect or along a certain dimension: space. It is also unlimited by any other country or space-occupying entity. In the same way, then, the Real, if it is unlimited, is unlimited along certain dimensions and is unlimited by any other being. Christians have traditionally thought of God as unlimited, infinite, in both these ways. To take the second first, God is unlimited by any other being: that is, he is unlimited in power or with respect to his being able to accomplish his will; no being can obstruct him, none can prevent him from doing what he wills. Clearly, this property, even if we think of it as negative, entails positive properties. If God is unlimited with respect to power, then he has power, which is certainly a positive property. If he is unlimited with respect to being able to accomplish his will, then he has the positive property of being able to accomplish his will. And if nothing can prevent him from doing what he wills, then he has the positive property of being able to do what he wills. Being unlimited along certain dimensions or in certain important respects is similar. God is not unlimited in every respect: that could presumably be so only if he had every property to the maximal degree, which is impossible. If, for example, he has the property of being a spirit, then he does not also have the property of being a material object--a tree, for instance. Rather, the traditional idea has been that God has every great-making property to the maximal degree. [51] So God is unlimited with respect, for example, to knowledge. And a being that is not limited with respect to knowledge has the maximal degree of knowledge, is omniscient, all-knowing. Such a being, of course, would have the positive property being a knower. [52] Is there a way out of this difficulty for the Hickian? Perhaps. He might try saying that this being is ultimate and unlimited, all right, but only with respect to properties of which we have no grasp. With respect to all the properties of which we do have a grasp, it is indeed limited, limited in the limiting sense of not exemplifying the property at all. It has the complement of every property we have a grasp of; it has other properties we have no grasp of; and the way in which it is infinite is that it has to the maximal degree some properties of which we have no grasp. Well, this sounds a little bizarre, but perhaps it avoids incoherence (and anyway, who promised us that reality would not be bizarre?). The idea is that there is a being that has no positive properties of which we have a conception, except for being involved in human experience and any properties that entails. This being is also unlimited in that it has to the maximal degree properties of which we have no conception. __________________________________________________________________ [46] See McGonagall's Poetic Gems (Dundee: David Winter and Son; London: Gerald Duckworth, first published in two parts in 1890 and first published as one volume in 1934). See also his More Poetic Gems, Still More Poetic Gems, Yet More Poetic Gems, and Poetic Gems Once Again. [47] But even this might be possible; if I have some other way of referring to you but also think that this description applies to you, then perhaps when I use that description, thinking it applies to you, I do, in fact, refer to you. Suppose God doesn't have some of the properties we think he has: suppose, for example, he isn't simple, in the classical sense, but composite, with a distinction to be made between him and his properties, between him and his existence, and so on. Even so, if a creed like the Belgic Confession refers to him as the spiritual, simple, creator of the universe, it still refers to him by that description, even if the description doesn't apply. [48] Kant's objection to this Leibnizian suggestion is that any such preestablished harmony would not be a cognitive relation; it couldn't support our having knowledge of the things in question. But it is very difficult to see why that should be so. Suppose God brings it about that our cognitive states appropriately match those of the world around us, and suppose the other conditions for warrant are met (as outlined, e.g., in Warrant and Proper Function): why wouldn't that be sufficient for our having knowledge of those things? [49] Where P is equivalent to Q if and only if it is necessary in the broadly logical sense that whatever exemplifies either P or Q exemplifies both P and Q. [50] Where a property A entails a property B if (and only if) it is necessary that any object that has A also has B. [51] This idea is what drives the ontological argument; see my The Nature of Necessity, chapter 10. [52] Indeed, the being in question would have at least uncountably many such properties: for each proposition P, the being in question would have the property of knowing whether P is true. Here I assume that there are at least uncountably many distinct propositions. This seems relatively uncontroversial in view of the fact that for each distinct real number r, there is the distinct proposition r is not identical with the Taj Mahal. __________________________________________________________________ C. Religiously Relevant? Even if the view is not incoherent, it pays another price. For suppose we believe there is a being of that sort. It is for us an empty idea; still, that is not to say that there couldn't be such a being. I don't know whether it is possible that there be a being like this, and have no grounds for making a judgment either way as to whether it is possible. Suppose we concede for the moment that it is. The basic question here is this: what reason is there for thinking such a being, if indeed there could be and is such a being, is in any special way connected with religion? According to Hick, ?we can say of the postulated Real an sich that it is the noumenal ground of the encountered gods and experienced absolutes witnessed to by the religious traditions? (246). Why should we think so? What reason is there for thinking this being is connected in some way with Christianity or with any other religion? Why say that Christians are in fact referring to or witnessing to this being? Maybe it is, instead, connected with warfare, prostitution, family violence, bigotry, or racism. And why think this being, or contact with it, has anything to do with ?transformation of human existence from self-centredness to Reality-centredness? (355)? Perhaps it is especially when human beings are in the grip of self-aggrandizement, hate, selfishness, and the like that they are most in contact with this being. If it has no positive properties of which we have a conception, why is the one assumption the least bit better than the other? The basic problem, here, is that if the Real has no positive properties of which we have a conception, then we have no reason at all to think that it is in religion that human beings get in experiential contact with this being, rather than in any other human activity: war or oppression, for example. This being has none of the properties ascribed by the practitioners of most of the great religions to the beings they worship: it is not good, or loving, or concerned with human beings, or wise, or powerful; it has not created the universe, does not uphold it, and does not pay attention to the universe or the creatures it contains. It is an unknown and unknowable X. But then why associate this unknowable X with religion, as opposed to warfare, violence, bigotry, and the horrifying things human beings often do to each other? We can put this question another way. Hick suggests that when Christians make their characteristic utterances, saying, for instance, ?In Christ, God was reconciling the world to himself,? what they say cannot be literally true but can be mythologically true. By literal' truth, he just means truth, ordinary truth: ?The literal truth or falsity of a factual assertion . . . consists in its conformity or lack of conformity to fact: it is raining here now' is literally true if and only if it is raining here now? (348). The mythological truth of a statement is a horse of quite another color: ?A statement or set of statements about X is mythologically true if it is not literally true but nevertheless tends to evoke an appropriate dispositional attitude to X? (348). So some dispositional attitudes toward the Real are appropriate and others (presumably) are not; some ways of responding to it are appropriate and others (presumably) not: Thus although we cannot speak of the Real an sich in literal terms, [53] nevertheless we live inescapably in relation to it, and in all that we do and undergo we are having to do with it as well as, and in terms of, our more proximate situations. Our actions are appropriate or inappropriate not only in relation to our physical and social environments but also in relation to our ultimate environment, the Real. True religious myths are accordingly those that evoke in us attitudes and modes of behaviour which are appropriate to our situation vis-ŕ-vis the Real. (351) And now our question can be put in terms of this suggestion. Why think that ?we live in relation to . . . the Real? at all? Either the relation living in relation to is merely formal or it is not. If it is, then it would have no connection with religion. So it isn't. But if it isn't, then (1) we have still another positive property had by the Real: the property of being such that we live in relation to it. And (2) (and more poignantly) why think we live in relation to the Real at all? After all, we don't live in relation to just any old thing--the highest mountain on Mars, for example, or the meanest shark in the Indian Ocean. If the Real has no positive properties of which we have a grasp, what is the reason for thinking we live in relation to it? Second, why think some ways of behavior are appropriate to the Real and others are not? Again, unless this is a purely formal relationship, not just anything is such that some ways of behaving are appropriate or, for that matter, inappropriate with respect to it. None of my behavior, I think, is either appropriate or inappropriate with respect to the highest mountain on Mars or the meanest shark in the Indian Ocean. If the Real has no positive properties of which we have a grasp, how could we possibly know or have grounds for believing that some ways of behaving with respect to it are more appropriate than others? Hick, of course, thinks that some ways of behaving with respect to the Real are more appropriate than others, and goes on to specify what those ways are: we are behaving appropriately with respect to the Real when we learn to turn away from self-centeredness and selfishness (?the transformation of human existence from self-centredness to Reality-centredness?). But why think that? This would, indeed, make sense if the Real had the attributes of God: if it were, in fact, a person who loves us, and wills that we turn from loving only ourselves to loving him above all and our neighbor as ourselves, and has so designed us that we attain our end and happiness when we do so. But the Real doesn't have any of those properties (they being positive properties of which we have a conception). And if it has no positive properties of which we have a grasp, why not suppose that hateful and selfish behavior are appropriate with respect to it? Or behaving in that weak, sniveling, envious way that Nietzsche thought characteristic of real Christians? We can't have it both ways. If this being is really such that we literally know nothing positive about it (if it has no positive properties of which we have a grasp), then there is no reason to think self-centered behavior is less appropriate with respect to it (granting, indeed, that some modes of behavior are more appropriate with respect to it than others) than living a life of love. We can put the question in still another way. Hick apparently thinks that some religious conceptions or ideas are authentic manifestations (?personae? or ?impersonae?) of the Real: And to the extent that the God and Father of our Lord Jesus Christ' is indeed an authentic persona of the Real, constituting the form in which the Real is validly thought and experienced from within the Christian strand of religious history, to that extent the dispositional response appropriate to this persona constitutes an appropriate response to the Real. . . . and of the eternal Buddha nature, he says And to the extent that this is an authentic impersona of the Real, validly thought and experienced from within the Buddhist tradition, life in accordance with the Dharma is likewise an appropriate response to the Real. (353) Again, the main question here is obvious: if the Real has no positive, nonformal properties of which we have a grasp, how could we possibly know or have reason to believe that any such personae or impersonae are authentic or, for that matter, inauthentic? Authenticity implies a certain fit between the (im)persona in question and the Real; but if we have no positive idea what the Real is like, we have no reason to think some personae or impersonae fit it better than others. Again, Hick thinks not only that some do fit better than others but also that we have some idea what they are: the God and Father of our Lord Jesus Christ, for example, and also others from other traditions. But how could we have reason to think a thing like that? If all we know about the Real is what Hick says we know, then for all we can tell, God, Vishnu and the Buddha are inauthentic and it is, for example, Ares, the god of war, Lucifer, and Stalin (or, for that matter, Uriah Heep or Beavis and Butthead) that are the authentic personae of the Real. Again, we can't have it both ways. If we know nothing about the Real, we have no reason to pick the personae Hick picks as authentic manifestations of it. The main point is that if the Real has no positive nonformal properties of which we have a grasp, then, for all we can see, any department of human life is as revelatory of the Real as any other. We don't have any way to pick and choose among them, thinking, for example, that the great world religions are where the Real manifests itself or where human beings experience it. For all we know, on this showing, it is in living like a thugee or like a member of the Ku Klux Klan that one is most authentically in touch with the Real. __________________________________________________________________ [53] Recall that we are amending this claim from chapter 19 in the light of chapter 14; we can speak literally of the Real, but only to predicate negative and purely formal properties of it, together with those positive properties entailed by its being such that we human beings are in experiential contact with it. __________________________________________________________________ D. Is There Such a Thing? So the view seems to be of dubious coherence. Stated carefully, it isn't initially incoherent. From Hick's point of view, however, the most important feature of this alleged being is that it is in some special way associated with religion; this being is what those who serve God, Brahman, and so on are really referring to; in the great world religions, people get into a special relationship to it. And that seems wholly gratuitous; perhaps the Real is really connected with those who serve themselves, or power, or white supremacy. But I come now, finally, to the question what reason Hick has for postulating such a being: why does he think there is a being with no positive properties of which we have a grasp? Hick's answer: The Real an sich is postulated by us as a pre-supposition, not of the moral life, but of religious experience and the religious life, whilst the gods, as also the mystically known Brahman, Sunyata and so on, are phenomenal manifestations of the Real occurring within the realm of religious experience. Conflating these two theses one can say that the Real is experienced by human beings, but experienced in a manner analogous to that in which, according to Kant, we experience the world: namely by informational input from external reality being interpreted by the mind in terms of its own categorial scheme and thus coming to consciousness as meaningful phenomenal experience. (243) Why should we want to postulate such a thing, a thing with no positive properties of which we have any grasp, but which is experienced by human beings in the great religions? More to the point, why does Hick postulate such a being? The answer, I think, must be explained dialectically. Hick began his spiritual odyssey as a traditional, orthodox Christian, accepting what I have been calling Christian belief'. He was then struck by the fact that there are other religions in which the claims of orthodox Christianity--trinity, incarnation, atonement--are rejected. Furthermore, so far as one can tell from the outside, so to speak, the claims of these other religions, taken literally, are as respectable, epistemically speaking, as the claims of Christianity. Still further, according to Jesus himself, ?By their fruits you shall know them.? The most important fruits, Hick thinks, are practical: turning away from a life of selfishness to a life of service; on this point, these other religions, he thinks, seem to do as well as Christianity. The conclusion he draws is that where Christianity differs from the others, we can't properly hold that it is literally true and the others literally false; that would be, he thinks, a sort of intellectual arrogance, a sort of spiritual imperialism, a matter of exalting ourselves and our beliefs at the expense of others. Instead, we must hold that the great religions are all equally valuable and equally true. How do we do this? Here is Hick's response: But if the Real in itself is not and cannot be humanly experienced, why postulate such an unknown and unknowable Ding an sich? The answer is that the divine noumenon is a necessary postulate of the pluralistic religious life of humanity. For within each tradition we regard as real the object of our worship or contemplation. If, as I have already argued, it is also proper to regard as real the objects of worship or contemplation within the other traditions, we are led to postulate the Real an sich as the presupposition of the veridical character of this range of forms of religious experience. Without this postulate we should be left with a plurality of personae and impersonae each of which is claimed to be the Ultimate, but no one of which alone can be. We should have either to regard all the reported experiences as illusory or else return to the confessional position in which we affirm the authenticity of our own stream of religious experience whilst dismissing as illusory those occurring within other traditions. But for those to whom neither of these options seems realistic the pluralistic affirmation becomes inevitable, and with it the postulation of the Real an sich, which is variously experienced and thought. . . . (249) Now this passage is apparently an argument of some kind, an argument for the conclusion that there is a being (the Real) of the sort Hick says there is. The argument seems to proceed from two premises: (1) all the great religions are ?veridical,? and (2) none of them is more veridical than the others. How does the argument go? The idea, I think, is that if we suppose there is a being of the sort Hick says there is, then, according to Hick, we can see how (1) and (2) could be true. I'm not sure I see just how that is supposed to go, but let that pass. What isn't as easily ignored, however, is a sort of incoherence. ?Within each tradition,? he says, ?we regard as real the object of our worship or contemplation.? So within the Christian tradition, we regard God as real; it is also ?proper to regard as real the objects of worship or contemplation within the other traditions.? This, of course, leads to a problem; for some of the personae and impersonae are such that if they are real, and have the properties ascribed to them, then other (im)personae are either unreal or do not have the properties ascribed to them. Well, perhaps the idea, as Hick seems to suggest elsewhere, is that we are to regard each of the (im)personae as empirically real, not transcendentally real (not really real); and perhaps we are to understand that as meaning that each being is such that by way of it, the practitioners of the religion in question somehow get in touch with the Real. The essential point, however, is that we are not to think of one or some of these as more valuable or closer to the truth than the others; that would be arbitrary and unwarranted. We should no longer regard as really real (or rather real simpliciter) the objects of worship of our own tradition. We must treat all traditions alike. Now the way Hick proposes to do this is to declare that all the traditions are actually mistaken; the beliefs in each tradition are mostly false. (?Literally? false, he says; but literal truth and falsehood, as Hick conceives them, are just truth and falsehood.) Still, there is something right or valid in religion--the recognition that there is something beyond the natural world, and the encouragement to live a life in which self-centeredness is overcome. So really, the bottom line is that Hick cannot find it in himself to think one religion--Christianity, say--is true and the others false, or that one is closer to the truth than the others. At bottom, there is a generous desire to avoid the self-aggrandizement and self-exaltation he sees as attaching to the declaration that one's own religious beliefs are true and those of others false. Here there are three comments or questions. First, is this posture in fact possible for a human being: can a person accept it, and accept it authentically, without bad faith or doublethink? I am to remain a Christian, to take part in Christian worship, to accept the splendid and powerful doctrines of traditional Christianity. However, I am also to take it that these doctrines are only mythologically true: they are literally false, although accepting them (i.e., accepting them as true, as literally true) puts or tends to put one into the right relation with the Real. And how can I possibly accept them, adopt that attitude toward them, if I think they are only mythologically true--that is, really false? I could, indeed, believe that they are mythologically true; believing that, however, doesn't move one toward the right kind of life; it is only believing the teachings themselves that allegedly has that salutary effect. Once I am sufficiently enlightened, once I see that those doctrines are not true, I can no longer take the stance with respect to them that leads to the hoped-for practical result. I am left, instead, in the position of a sad and disillusioned Gnostic. I no longer hold Christian belief; I recognize, as I think, that it is in fact false. I also see, of course, that those who do accept it as true are mistaken, deluded; but at any rate they are in the fortunate position of enjoying the comfort and strength and consolation these false beliefs bring; they are also being moved closer to the right kind of life by virtue of accepting them. Neither the comfort [54] and consolation nor the practical efficacy is available to me. Second, there is something wholly self-defeating, so it seems to me, in Hick's posture. If we take this position, then we can't say, for example, that Christianity is right and Buddhism wrong; as Christians, we don't disagree with the Buddhists; and we take this stance in an effort to avoid self-exultation and imperialism. But we do something from the point of view of intellectual imperialism and self-exaltation that is much worse: we now declare that everyone is mistaken here, everyone except for ourselves and a few other enlightened souls. We and our graduate students know the truth; everyone else is sadly mistaken. Isn't this to exalt ourselves at the expense of nearly everyone else? Those who think there really is such a person as God are benighted, unsophisticated, unaware of the real truth of the matter, which is that there isn't any such person (even if thinking there is can lead to practical fruits). We see Christians as deeply mistaken; of course we pay the same compliment to the practitioners of the other great religions; we are equal-opportunity animadverters. We benevolently regard the rest of humanity as misguided; no doubt their hearts are in the right place; still, they are sadly mistaken about what they take to be most important and precious. I find it hard to see how this attitude is a manifestation of tolerance or intellectual humility: it looks more like patronizing condescension. The basic problem is that, given our actual intellectual and spiritual situation, it simply isn't possible to avoid serious disagreement with others. If some people believe p and others believe something q incompatible with p, there is no way in which we can avoid serious disagreement. If we affirm p, we disagree with those who affirm q; if we affirm q, we disagree with those who affirm p; if we propose a higher resolution, saying that neither p nor q is true (though perhaps each is mythologically true'), then we disagree with both groups. [55] But if it is imperialistic or somehow out of order to affirm p, thus disrespecting the partisans of q, why is it better to disrespect them all by pronouncing them all wrong? Third, Hick doesn't, of course, produce an argument for the conclusion that no religion could be closer to the truth than others; it is more like a practical postulate, a benevolent and charitable resolution to avoid imperialism and self-aggrandizement. But is this the way to do it? Clearly, in most areas of life, some people are closer to the truth than others. If the nominalists are right, all of us realists are wrong; if the modal skeptics are right, we modal true believers are mistaken; if the white supremacists are right, many of the rest of us, bent as we are on toleration, are wrong, and seriously wrong. Why should it be different in religion? The idea that in religion we must all be equally right and all equally wrong seems no more compelling than the idea that in thinking about religion we must all be equally right and equally wrong. Hick's reason for thinking all religions equally right seems to be a desire to avoid self-aggrandizement; shouldn't the same desire lead him to hold that his views about religion--his view, for example, that they are all equally right and equally wrong--really have no more claim to truth than any other view here (for example, the view that Christianity alone, say, is correct)? He doesn't do that, and rightly so. We can't properly do so in religious belief either. In religious belief as elsewhere, we must take our chances, recognizing that we could be wrong, dreadfully wrong. There are no guarantees; the religious life is a venture; foolish and debilitating error is a permanent possibility. (If we can be wrong, however, we can also be right.) Our topic, in this book, is the de jure question with respect to Christian belief--not the question whether Christian belief is true (although, of course, that is the more important question) but whether it is reasonable or rational or rationally justifiable to accept it. We have been examining a preliminary matter: is there really such a question? Is there really such a thing as Christian belief? Or is it rather the case that even if there were such a person as God, we couldn't refer to and think about him, or predicate positive, nonformal properties of him? Our results, so far, have been that there isn't the slightest reason to think so. There is no reason at all to think it isn't possible to think about God; there is no reason at all to think that we cannot predicate such positive, nonformal properties of him as wisdom, knowledge, love, and all the rest. Obviously, there is enormously more to be said; this topic deserves a book in itself. There isn't room for that in this book; so we shall have to content ourselves with what we have. At any rate, we can rest in the assurance that if there is reason to think our question ill-formed or somehow logically out of order, it is at the least exceedingly well concealed. We can therefore go in reasonable confidence to the next question: what, precisely, is this de jure question? __________________________________________________________________ [54] According to the Heidelberg Catechism (Q. 1), ?My only comfort in life and in death is that I am not my own, but belong--body and soul, in life and in death--to my faithful Savior Jesus Christ.? According to Hick, however, my only comfort in life and in death is that I know the sad truth: believing the great teachings of Christianity has beneficial effects, but those teachings are, in fact, false. [55] The same goes if we propose to remain agnostic about p and q; see below, chapter 13. __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ PART II WHAT IS THE QUESTION? __________________________________________________________________ 3 Justification and the Classical Picture In part I, I considered a certain kind of objection to the de jure question with respect to Christian belief--the question, that is, whether it is rational or reasonable or intellectually respectable to accept Christian belief. This objection was that the de jure question is, to say the least, premature: strictly speaking, it isn't really possible to hold a belief of the sort traditional Christians think they hold. That is because Christians think of God as ultimate and infinite, and there is something conceptually out of order with the very idea that it is possible to have a belief about a being that is ultimate and infinite. I concluded that this objection isn't cogent; there is here no obstacle to raising the de jure question. But the next thing to see is that it is far from obvious just what that de jure question or objection is supposed to be; precisely what question (or questions) is it that critics mean to press when they ask whether Christian and theistic belief is rational, or rationally defensible, or rationally justifiable, or whatever? Critics claim that Christian belief is not rationally justified or justifiable: what, precisely, is the infirmity or defect they are ascribing to the Christian believer? What, exactly, is the question? Call this question the metaquestion'. One problem with contemporary discussions of the justification of Christian belief is that the metaquestion is almost never asked. People ask whether Christian belief is rational or reasonable or rationally justifiable; they turn immediately to answering that question, without first considering just what the question is. What is it? That is not easy to say; nevertheless, it is our subject in part II. This chapter is devoted to examining a certain answer to the metaquestion: that the de jure question is whether Christian belief is justified. That question is one that originates in classical foundationalism, a way of thinking about these topics that has historically been extremely influential and is still very much with us now. According to the classical foundationalist, the de jure question is really the question whether Christian belief is justified; but how is this term to be understood? I shall examine the seventeenth-century roots of classical foundationalism, explore the connection between justification and evidentialism that the classical foundationalist sees, and briefly outline some of its contemporary descendants. Then in the second half of the chapter, I'll argue both that classical foundationalism faces insuperable problems, and that the notion of justification does not offer a satisfying version of the de jure question. We must begin with some history--first, a look back into the relatively recent past and then a deeper look into the more remote past. With respect to the first, I can make my point most easily by referring to some earlier work of my own; please forgive the personal reference. The present book, as I said in the preface, is a sequel: to Warrant: The Current Debate (hereafter WCD) and Warranted Proper Function (hereafter WPF); it is also and perhaps more important a sequel to God and Other Minds [56] and ?Reason and Belief in God.? [57] The chief topic of God and Other Minds, as I put it then, is the ?rational justification? of belief in God. I set out to address the de jure question; like everyone else, however, I didn't so much as raise the metaquestion. Following my elders and betters, I initially took it for granted that this question of the rational justification of theistic belief is identical with, or intimately connected with, the question whether there are proofs, or at least good arguments, for or against the existence of God. You discuss this question of the rationality of belief in God by consulting the evidence: does it on balance support theistic belief? (If it does [and does so strongly enough], such belief is rational; otherwise it is irrational.) And that question, in turn, was so taken that the way to answer it is by considering the arguments for and against the existence of God. On the pro side, there were the traditional theistic proofs, the cosmological, teleological, and ontological arguments, to follow Kant's classification. On the con side, there was, first of all, the problem of evil (construed as the claim that the existence of evil is logically inconsistent with the existence of a wholly good, all-powerful, and all-knowing God). Then there were also some rather opaque claims to the effect that the progress of modern science, or the attitudes necessary to its proper pursuits, or perhaps something similar lurking in the nearby bushes, or maybe something else that had been learned by ?man come of age?--the idea was that something in this general neighborhood also offers evidence against the existence of God. And it was also clearly assumed that belief in God was rational and proper only if on balance the evidence, so construed, favored it. So here is a possible answer to the metaquestion and a candidate for the post of being the de jure question: does the evidence support Christian belief? In this chapter I want to think about this answer to the metaquestion. Does it give us a serious question for Christian believers or a serious criticism of Christian belief? In God and Other Minds, I argued first that the theistic proofs or arguments do not succeed. In evaluating these arguments, I employed a traditional but wholly improper standard: I took it that these arguments are successful only if they start from propositions that compel assent from every honest and intelligent person and proceed majestically to their conclusion by way of forms of argument that can be rejected only on pain of insincerity or irrationality. Naturally enough, I joined the contemporary chorus in holding that none of the traditional arguments was successful. (I failed to note that no philosophical arguments of any consequence meet that standard; hence the fact that theistic arguments do not is of less significance than I thought.) I then argued that the objections to theistic belief are equally unimpressive; in particular, the deductive argument from evil (the argument that there is a contradiction between the existence of God and the existence of evil), I said, is entirely unsuccessful. So I saw, as I thought, that neither the arguments for the existence of God nor the arguments against it are conclusive; but then where does that leave us with respect to the question of the rationality or rational justifiability of belief in God? Does it follow, as seemed to be the prevailing opinion, that agnosticism was the right response and that belief in God, under these conditions, is irrational, contrary to reason, not rationally justifiable? That seemed to me wrong, but where could we go to pursue this question? How could we carry the inquiry further? Faced with this impasse, I decided to compare belief in God with other beliefs, in particular, our belief in other minds. There is allegedly a traditional philosophical problem of other minds: since we can't perceive the thoughts and feelings of other people, do we know and how do we know they really have thoughts and feelings? More poignantly, how do we know that what we take to be persons (beings with thoughts, feelings, and intentions) really are persons and not, for example, cunningly constructed robots? [58] I noted that the dialectical structure uncovered in the case of theistic arguments is recapitulated in the case of other minds: the objections to belief in other minds don't seem at all formidable, but unhappily there also aren't any good arguments for other minds--particularly if we employ the same high standards of goodness as were ordinarily applied to theistic arguments. I claimed that the strongest argument for the existence of God and the strongest argument for other minds are similar and that they fail in similar ways. Hence my ?tentative conclusion?: ?if my belief in other minds is rational, so is my belief in God. But obviously the former is rational; so, therefore, is the latter.? Here two things are noteworthy. First, I was somehow both accepting but also questioning what was then axiomatic: that belief in God, if it is to be rationally acceptable, must be such that there is good evidence for it. This evidence would be propositional evidence: evidence from other propositions you believe, and it would have to come in the form of arguments. This claim wasn't itself argued for: it was simply asserted, or better, just assumed as self-evident or at least utterly obvious. What was then taken for granted has now come to be called evidentialism' (a better title would be evidentialism with respect to belief in God', but that's a bit unwieldy). Evidentialism is the view that belief in God is rationally justifiable or acceptable only if there is good evidence for it, where good evidence would be arguments from other propositions one knows. If it is accepted apart from such evidence or arguments, then it is at best intellectually third-rate: irrational, or unreasonable, or contrary to one's intellectual obligations. Second, I failed to ask why this question of rational justifiability is important or, indeed, what the question is. I didn't give that question--namely, the question what is this rational justifiability of which I am speaking?--so much as a passing glance. Further, why would rational justification, whatever precisely it is, require evidence? What is the connection between evidence and justification? And if the latter does require evidence, why would that evidence have to take the form of arguments (deductive or probabilistic), evidence from other propositions one already believes? And what sorts of propositions could properly function as the premises of these arguments? I didn't raise these questions. It wasn't, however, because their answers were well-known, so that further inquiry would be carrying coals to Newcastle. On the contrary: no one else asked or answered these questions either; instead, people turned directly to the arguments for and against theistic belief, taking it utterly for granted that this was the way to investigate its rational justification. [59] Taking evidentialism for granted was de rigueur then and is still popular now. But what is this rational justification? And why does it require evidence, propositional evidence? And how does it happen that everyone just took for granted this connection between justification and propositional evidence? These are some of the questions we must ask. __________________________________________________________________ I. John Locke Here what we need is another bit of history, some more of that archaeology of which Foucault speaks (although again [see WCD, p. 11] I doubt that we will uncover a hidden political agenda or a subterranean bid for power). This question as to the rational justifiability of Christian belief goes back to the Enlightenment response to the spiritual and intellectual ferment generated (in part) by the Reformation; the characteristically modern response to this ferment can be seen as getting its start in the works of René Descartes and John Locke. Both Descartes and Locke were impressed by the enormous disagreement in religious and philosophical matters; this means, of course, that error pervades our belief in these areas. They were also impressed (along with their successors) with the meager progress made in philosophical matters. Philosophy, said Descartes, ?has been cultivated for many centuries by the best minds that have ever lived, and nevertheless no single thing is to be found in it which is not a subject of dispute, and in consequence which is not dubious.? [60] Descartes has his remedy (a characteristically modern remedy): start over. Discard anything that isn't certain, and rebuild your noetic structure on the basis of what is certain. Recall those famous words in the introduction to the Meditations: It is now some years since I detected how many are the false beliefs that I had from my earliest youth admitted as true, and how doubtful was everything I had since constructed on this basis, and from that time I was convinced that I must once for all seriously undertake to rid myself of all the opinions which I formerly accepted, and commence to build anew from the foundation. . . . [61] It is John Locke, however, not Descartes, who is probably most crucial for our understanding of the de jure question and the modern compulsion to ask it. [62] In the ?Epistle to the Reader? prefacing his long, rambling An Essay concerning Human Understanding, Locke recounts a meeting with ?five or six friends,? in which they discussed a certain subject that Locke doesn't there identify: [They] found themselves quickly at a stand, by the difficulties that rose on every side. After we had awhile puzzled ourselves, without coming any nearer a resolution of those doubts which perplexed us, it came into my thoughts that we took a wrong course; and that before we set ourselves upon inquiries of that nature, it was necessary to examine our own abilities, and see what objects our understandings were, or were not, fitted to deal with. This I proposed to the company, who all readily assented; and thereupon it was agreed that this should be our first inquiry. [63] That discussion was the genesis of the Essay; it probably took place in the winter of 1670-71, [64] and a momentous meeting it was. The book itself wasn't finished (or at least published) for another eighteen years or so, which accounts in part for its length and rather disorganized, repetitious character. Locke doesn't tell us what the topic of discussion was, but James Tyrell, one of the five or six friends at the gathering, noted in the margin of his copy of the Essay (now in the British Museum) that the topic of discussion was ?the principles of morality and revealed religion.? [65] And Locke's Essay has been immensely influential in modern thought on this topic; it is perhaps not too much to say that his seminal work is the single most important source of the way of thinking on these topics that has dominated Western thought for the last three centuries. This book ushers in epistemology in the West. It is not, of course, that previous philosophers had nothing to say about epistemology. After all, Plato's Theaetetus asks one of the main questions in the theory of knowledge: what is it that must be added to mere true belief to get knowledge? What is that quality or quantity, enough of which makes the difference between true belief and knowledge? Aristotle and Aquinas, furthermore, had much to say about scientia, scientific knowledge, and also much to say about how the process of intellection works, what goes on when someone knows or believes something. Still, the questions Locke asked and the answers he gave have a peculiarly modern ring; we resonate to them, because his way of thinking about them became the modern way of thinking about them; and despite postmodern proclamations of the death or end of epistemology, this is still, for the most part, our way of thinking about these matters. Locke lived through one of the most turbulent periods of British intellectual and spiritual history; it was, in particular, the religious ferment and diversity, the enormous variety of religious opinion, that caught his attention. Of course he knew that in parts of the world other than Europe there were religions quite different from Christianity, but he was particularly impressed by the diversity of religious opinion in his own country. There was the Catholic-Protestant debate, and within Protestantism there were countless sects, countless disagreements and controversies; it was a time when every man thought what was right in his own eyes. Locke proposes to inquire into the grounds of those persuasions which are to be found amongst men, so various, different, and wholly contradictory; and yet asserted somewhere or other with such assurance and confidence, that he that shall take a view of the opinions of mankind, observe their opposition, and at the same time consider the fondness and devotion wherewith they are embraced, the resolution and eagerness wherewith they are maintained, may perhaps have reason to suspect, that either there is no such thing as truth at all, or that mankind hath no sufficient means to attain a certain knowledge of it. (Locke's Introduction to the Essay, para. 2, p. 27) One problem here, says Locke, is fideism; many oppose faith to reason, declaring both that faith prescribes what reason proscribes, and that it is faith that is to be accepted and followed: For, to this crying up of faith in opposition [his emphasis] to reason, we may, I think, in good measure ascribe those absurdities that fill almost all the religions which possess and divide mankind. For men having been principled with an opinion, that they must not consult reason in the things of religion, however apparently contradictory to common sense and the very principles of all their knowledge, have let loose their fancies and natural superstition; and have been by them led into so strange opinions, and extravagant practices in religion, that a considerate man cannot but stand amazed at their follies, and judge them so far from being acceptable to the great and wise God, that he cannot avoid thinking them ridiculous and offensive to a sober good man. So that, in effect, religion, which should most distinguish us from beasts, and ought most peculiarly to elevate us, as rational creatures, above brutes, is that wherein men often appear most irrational, and more senseless than beasts themselves (bk. IV, chap. xviii, para. 11, p. 426) Another source of riotous error and confusion in religion is tradition, believing a proposition just because you have been taught it or because those around you believe it: The great obstinacy that is to be found in men firmly believing quite contrary opinions, though many times equally absurd, in the various religions of mankind, are as evident a proof as they are an unavoidable consequence of this way of reasoning from received traditional principles. So that men will disbelieve their own eyes, renounce the evidence of their senses, and give their own experience the lie, rather than admit of anything disagreeing with these sacred tenets. (IV, xx, 10, p. 450) Tradition, he says (in characteristic Enlightenment disparagement), keeps in ignorance or error more people than all the other [the other sources of error] together . . . I mean the giving up our assent to the common received opinions, either of our friends or party, neighbourhood or country. How many men have no other ground for their tenets, than the supposed honesty, or learning, or number of those of the same profession? (IV, xx, 17, pp. 456-57) Appeals to tradition to settle disagreement had become ineffective; there were just too many. One had to choose which of these many conflicting traditions to endorse. Locke thought this disorderly pluralism quite scandalous; it was even more scandalous that there seemed no rational way to put an end to the contentious disputes. The Essay was Locke's attempt to do what he could to put matters right. Book IV, ?Of Knowledge and Probability,? is the end of the book--both in comprising the last three hundred pages or so and in dealing with the question whose resolution is Locke's goal; and even in book IV he spends another two hundred pages before explicitly addressing it. That main question is: how should we regulate our opinion with respect to belief in general? In particular, how shall we regulate our opinion with respect to religious belief? As A. D. Woozley [66] says, this is the principal topic of the Essay. As he also says, readers often don't get to it, being a bit disheartened by having to wade through what amounts to a six-hundred-page preface. Still, it is what he says on this head that is most crucial to an understanding of Locke's enterprise, as well as to our metaquestion. __________________________________________________________________ A. Living by Reason The initial problem, of course, is that disorderly crowd of opinions: ?men, extending their inquiries beyond their capacities, and letting their thought wander into those depths where they can find no sure footing, it is no wonder that they raise questions and multiply disputes, which, never coming to any clear resolution, are proper only to continue and increase their doubts, and to confirm them at last in perfect skepticism? (Locke's Introduction to the Essay, para. 7, p. 31). Like Hume and Kant after him, Locke thinks the remedy requires that we first make a juster and more accurate appraisal of our intellectual capacities and capabilities: Whereas were the capacities of our understandings well considered, the extent of our knowledge once discovered, and the horizon found which sets the bounds between the enlightened and dark parts of things; between what is and what is not comprehensible by us, men would perhaps with less scruple acquiesce in the avowed ignorance of the one, and employ their thoughts and discourse with more advantage and satisfaction in the other. (Locke's Introduction to the Essay, para. 7, p. 31) The aim is not to achieve Cartesian certainty (about which he makes several disparaging remarks). Rather, ?If we can find out those measures, whereby a rational creature, put in that state in which man is in this world, may and ought to govern his opinions, and actions depending thereon, we need not to be troubled that some other things escape our knowledge? (Introduction, para. 6, p. 31). What we need to find out is how we may and ought to govern and regulate our opinion, or assent. And his answer, in prototypical Enlightenment fashion, is that we ought to govern our opinion by following reason. But what does that mean? What is opinion and what is reason, and how can we govern the former by following the latter? 1. Opinion If we are to have any hope of overcoming the contentious and disputatious horde of conflicting opinion with which we are beset, says Locke, we must all learn to govern opinion and assent properly. Following Plato, Locke thinks of opinion as contrasting with knowledge; to see what he thinks opinion is, we must therefore look to his views about knowledge. He thinks we have four kinds of knowledge, all of them involving certainty. First, there is what he regards as the paradigm of knowledge: perceiving the ?agreement or disagreement of our ideas.? It isn't easy to see precisely what he had in mind here, but the principal sort of knowledge involved here is the knowledge of self-evident propositions, such propositions as 2 + 1 = 3. [67] A properly functioning human being can simply see that these propositions are true (and further, that they couldn't possibly be false). There is no issue of regulating this kind of belief, says Locke, because a properly formed human being simply can't withhold belief from self-evident propositions: ?This part of knowledge is irresistible, and like bright sunshine, forces itself immediately to be perceived, as soon as ever the mind turns its view that way; and leaves no room for hesitation, doubt, or examination, but the mind is presently filled with the clear light of it? (IV, ii, 1, p. 177). Such knowledge is certain; it is ?beyond all doubt, and needs no probation, nor can have any; this being the highest of all human certainty? (IV, xvii, 14, p. 407). Second, there is knowledge of propositions about the contents of your own mind, that is, propositions about the ideas of which you are the subject. An example would be your knowledge that you have a mild pain in your left elbow, or that you seem to see something white (i.e., things look to you the way they look when you are in fact seeing something white). This knowledge, says Locke, is infallible (IV, i, 4, p. 169, and elsewhere). This means at least that you cannot mistakenly believe such a proposition; if you believe that you seem to see something white, it follows that you do seem to see something white (though, of course, you may be mistaken in thinking there really is something white there). Following later custom, let's say that propositions of this sort about my own mental states are incorrigible for me. Third, there is also a kind of knowledge of ?other things,? of external objects around you: And of this, the greatest assurance I can possibly have, and to which my faculties can attain, is the testimony of my eyes, which are the proper and sole judges of this thing; whose testimony I have reason to rely on as so certain, that I can no more doubt, whilst I write this, that I see white and black, and that something really exists that causes that sensation in me, than that I write or move my hand; which is a certainty as great as human nature is capable of, concerning the existence of anything, but a man's self alone, and of God. (IV, xi, 2, pp. 326-27; see also IV, ii, 14, p. 186) It isn't wholly clear just what it is I know here: do I know that the piece of paper is white, that my hand is moving, and that the ink is black? Locke vacillates. Sometimes (for example, when commenting on the relation between faith and reason) he speaks as if our knowledge of external objects includes the sort of everyday knowledge we get from perception: that my hand is moving, that the trees in the backyard are budding, and so on. Other times, and perhaps when he's being more careful or at least more official, he suggests that what we know of the external world is much sparser, more like My current ideas of treehood and green are caused by something external to me. I may not know what these external objects are like (I don't know that they include trees, or buds, or objects that are green), but I do know that there is something external causing me to have these ideas. And fourth, there is demonstrative knowledge. I can come to know a proposition by deducing it from or seeing that it is entailed by propositions of the above three sorts (where a proposition p entails a proposition q just if it is not possible, in the broadly logical sense, that p be true and q false. [68] Accordingly, some propositions that you can deduce from propositions that are self-evident, incorrigible, or evident to the senses are also certain for you; among these propositions, Locke thinks, is the existence of God (IV, x, 1-6, pp. 306-10). Indeed, he adds, ?From what has been said, it is plain to me we have a more certain knowledge of the existence of a God, than of anything our senses have not immediately discovered to us.? When it comes to knowledge, therefore, we have no control over our giving assent; assent is elicited willy-nilly, and the question of how we should regulate assent in this area therefore does not arise. (We can't regulate it at all, anymore than I could regulate the direction in which I fall, if I fell off a cliff.) Of course, knowledge forms only a part of the beliefs to be found in a human noetic structure and, according to Locke, a relatively small part (?Our knowledge, as has been shown, being very narrow,? IV, xv, 2, p. 364). It is opinion that includes the bulk of what we ordinarily believe; and it is with respect to opinion--that which we believe but do not know--that the question of regulation does arise. 2. Reason Locke's crucial claim is that we must be guided, in the formation of opinion, by reason. Well, what is reason? First, it is ?a faculty in man, that faculty whereby man is supposed to be distinguished from beasts, and wherein it is evident he much surpasses them? (IV, xvii, 1, p. 386). Second, reason is the power whereby we can discern broadly logical relations among propositions (IV, xviii, 3, p. 417), which, of course, are the candidates for our assent, the things we believe. In particular, by virtue of employment of reason, we distinguish two kinds of relations among propositions: The greatest part of our knowledge depends upon deductions and intermediate ideas: and in those cases where we are fain to substitute assent instead of knowledge, and take propositions for true, without being certain they are so, we have need to find out, examine, and compare the grounds of their probability. In both these cases, the faculty which finds out the means, and rightly applies them, to discover certainty in the one, and probability in the other, is that which we call reason. For, as reason perceives the necessary and indubitable connexion of all the ideas or proofs one to another, in each step of any demonstration that produces knowledge; so it likewise perceives the probable connexion of all the ideas or proofs one to another, in every step of a discourse, to which it will think assent due. (IV, xvii, 2, p. 387) It is by reason, therefore, that we perceive deductive and probabilistic relations among propositions. We needn't say anything here about deductive relations between propositions; and while much needs to be said about probability, Locke doesn't say it. He does say a little, however, beginning rather inauspiciously by declaring, ?Probability is likeliness to be true? (IV, xv, 3, p. 365). The uninformative character of this, however, is ameliorated by his pointing out [69] that probability has to do with what occurs for the most part' in our experience; and he adds that testimony from others also establishes probability (IV, xv, 4, pp. 365-66). Locke seems to think of probability as an objective relation among propositions; he probably also thinks that it is a quasi-logical relationship among them; his views, therefore, may be precursors of those of J. M. Keynes, Rudolf Carnap, and others. [70] Probability, furthermore, comes in degrees: ?Upon these grounds depends the probability of any proposition: and as the conformity of our knowledge, as the certainty of observations, as the frequency and constancy of experience, and the number and credibility of testimonies do more or less agree or disagree with it, so is any proposition in itself more or less probable? (IV, xv, 6, p. 367). Here he seems to suggest that a proposition is probable to some degree ?in itself?; he is better understood, I think, as holding that probability is a relation between propositions. A proposition has a certain degree of probability for me' (i.e., relative to those propositions that are certain for me); what counts with respect to the formation of my opinion is the probability of the candidate in question with respect to what is certain for me. 3. Regulating Opinion by Reason Locke's claim is that we should regulate our opinion or assent by reason; but what does this mean? How do you do a thing like that? His answer, fundamentally, is that I must regulate my opinion in such a way that I opine only that which is probable with respect to that which is certain for me. I have no control over my assent when it comes to knowledge, what is certain for me; however, I do have control over my assent when it comes to opinion, what isn't certain. And the rule here is that I must not assent to a proposition unless it is probable with respect to what is certain for me. Assent, furthermore, comes in degrees [71] (IV, xvi, 1, p. 369). More exactly, then, the rule is that I should proportion my degree of assent to the probability of the proposition in question: ?The grounds of probability we have laid down in the foregoing chapter: as they are the foundations on which our assent is built, so are they also the measure whereby its several degrees are, or ought to be regulated? (IV, xvi, 1, p. 369). More specifically, for any proposition that comes to my attention, I should proportion my degree of assent to it to the degree to which that proposition is probable with respect to what is certain for me. Proper procedure here is ?not entertaining any proposition with greater assurance than the proofs [deductive or inductive] it is built upon will warrant? (IV, xix, 1, p. 429) (probabilistic proofs as well as deductive proofs). Another way to put this: I should proportion degree of assent to the evidence; that is, I should believe a proposition p with a firmness that is proportional to the degree to which p is probable with respect to what is certain for me. This is what it is to regulate or govern opinion according to reason. __________________________________________________________________ [67] For an account of self-evidence, see WPF, chapter 6. [68] I do not and cannot know all propositions entailed by those of the above sort, of course; some might be much too complicated and difficult for me to grasp, and others might be such that I simply can't see the connection between them and propositions of the above three sorts. For still others, the argument for them is so long and complicated that I lack the certainty required by knowledge. [69] As Aristotle also did; see WPF, p. 159. [70] What he says is consistent with other views, however, including the one proposed in chapter 9 of WPF. [71] Here I think he means to point to two phenomena: first, that one believes some propositions more firmly than others, and second, that we judge some propositions more probable than others. To illustrate the first, I believe that 7 + 5 = 12 more firmly than I believe that Glasgow is west of Aberdeen, but I do believe both of these propositions. As for the second, I believe it is reasonably probable that all the continents of Earth once formed a supercontinent; I also believe that it is more probable that the works attributed to Shakespeare were really written by Shakespeare, not by Bacon. __________________________________________________________________ B. Revelation The question that started off the whole discussion issuing in the nine hundred pages of the Essay was on ?the principles of morality and revealed religion.? But now we see that we are to regulate our opinion by reason, that is, proportion our belief in a proposition to the degree to which it is probable with respect to what is certain for us. Does this mean, then, that divine revelation, ?revealed religion,? is to play no role in the right regulation of opinion? If that regulation demands that we proportion degree of assent to the evidence, what room is there for assenting to ?the great things of the Gospel,? as Jonathan Edwards calls them, the incarnation, atonement, and other central features of Christianity? Must we conclude that God could not reveal to us propositions unavailable by the use of our natural faculties? Surely not: ?God, in giving us the light of reason, has not thereby tied up his own hands from affording us, when he thinks fit, the light of revelation? (IV, xviii, 8, p. 423). Even if he does afford us the light of revelation, however, must we not regulate assent in such a way as to believe what he reveals, only if the latter is probable with respect to what is certain for us? If so, how can we accept what he teaches by way of revelation? Incarnation, atonement, and trinity don't seem particularly probable with respect to what is self-evident or about my own mental states. Locke answers, first, that God indeed can and does reveal such truths to us, and that what he reveals should certainly be believed: ?we may as well doubt of our own being, as we can whether any revelation from God be true? (IV, xvi, 14, p. 383); ?Whatever God hath revealed is certainly true: no doubt can be made of it? (IV, xviii, 10, p. 425). But then does he think these great truths are probable with respect to what is certain for us? First, he declares repeatedly that we can't properly believe what goes against reason in the sense of going contrary to the principles of knowledge: a man can never have so certain a knowledge, that a proposition which contradicts the clear principles and evidence of his own knowledge was divinely revealed, or that he understands the words rightly wherein it is delivered, as he has that the contrary is true, and so is bound to consider and judge of it as a matter of reason, and not swallow it. . . . (IV, xviii, 8, p. 424) However, it is not required that to be worthy of assent, such a teaching must be probable with respect to what is certain for me. Rather, what has to be probable, in this way, is that the doctrine in question is indeed revealed, really is proposed for our assent by the Lord: So that faith is a settled and sure principle of assent and assurance, and leaves no manner of room for doubt or hesitation. Only we must be sure that it be a divine revelation, and that we understand it right: else we shall expose ourselves to all the extravagancy of enthusiasm . . . (IV, xvi, 14, p. 383) and Whatever God hath revealed is certainly true: no doubt can be made of it. This is the proper object of faith: but whether it be a divine revelation or no, reason must judge. (IV, xviii, 10, p. 425) Locke's constant question is how do you know that this is from God?' ?How do I know that God is the revealer of this to me; that this impression is made upon my mind by his Holy Spirit; and that therefore I ought to obey it? If I know not this, how great soever the assurance is that I am possessed with, it is groundless; whatever light I pretend to, it is but enthusiasm? (IV, xix, 10, p. 435). ?Reason,? he says, ?must be our last judge and guide in everything? (IV, xix, 14, p. 438, his emphasis). He goes on: I do not mean that we must consult reason, and examine whether a proposition revealed from God can be made out by natural principles, and if it cannot, that then we may reject it: but consult it we must, and by it examine whether it be a revelation from God or no: and if reason finds it to be revealed from God, reason then declares for it as much as for any other truth, and makes it one of her dictates. (IV, xix, 14, p. 439) Overall, then, the view is this: God can certainly reveal truths to us. We are not obliged to accept as revealed, however, anything that would go contrary to what we would otherwise know, even with respect to the lowest level of knowledge. Furthermore, a given candidate p for revelation, if it doesn't have evidence from what is certain, cannot have any more epistemic probability than is enjoyed by the proposition that p is indeed a revelation from God (IV, xvi, 14, p. 383). So we are to follow reason, in the formation of religious opinion, but so doing does not preclude accepting certain propositions as specially revealed by God, and accepting them on that basis. __________________________________________________________________ [60] Part I of the Discourse on the Method of Rightly Conducting the Reason and Seeking Truth in the Sciences, in The Philosophical Works of Descartes, tr. and ed. Elizabeth S. Haldane and G. R. T. Ross (New York: Dover, 1955 [originally published by Cambridge University Press in 1931]), pp. 85-86. [61] The Philosophical Works of Descartes, p. 144. [62] See Nicholas Wolterstorff's luminous and illuminating essay on Locke in John Locke and the Ethics of Belief (Cambridge: Cambridge University Press, 1996). [63] An Essay concerning Human Understanding, ed. with ?Prolegomena? by Alexander Fraser (New York: Dover, 1959 [first published by Oxford University Press in 1894]), vol. 1, p. 9. Subsequent page references to Locke's essay are to this edition. [64] Fraser's footnote 1, p. 9. [65] Fraser's ?Prolegomena,? p. xvii. [66] Introduction to his abridgment of the Essay (New York: NAL Penguin, 1974 [originally published Collins, 1964]), p. 15. The Essay is long and confusing; it was composed over many years and didn't receive anything like the final editing it needed. As a result, it has been published in abridged editions going all the way back to 1694 (Boston: Printed by Manning and Loring, for J. White, Thomas and Andrews, D. West, E. Larkin, J. West and the proprietor of the Boston bookstore) four years or so after its publication and ten years before Locke's death. These abridgments sometimes delete some of the passages most important to a proper understanding of the Essay; for example, A. D. Woozley's omits the absolutely crucial passage quoted on pp. 86-87, below. __________________________________________________________________ II. Classical Evidentialism, Deontologism, and Foundationalism In God and Other Minds, I took for granted what was then axiomatic: that belief in God is rationally justifiable only if there are good arguments for it, and only if the arguments in favor of it are stronger than the arguments against it. The origin--at least the proximate origin--of this idea is to be found in the work of Locke I've been outlining. A belief is acceptable, he says, only if it is either itself certain or else probable (i.e., more probable than not) with respect to propositions that are certain for me. Christian belief, clearly enough, is not certain for me: it is not self-evident, incorrigible, or a deliverance of the senses. Hence, if it is to be acceptable, it must be probable with respect to propositions of these sorts. Locke doesn't, so far as I know, explicitly raise the question whether I must know or believe that the belief is thus probable, if it is to be acceptable for me; I think he assumes that it must be. He thinks of the matter in terms of applying a test: a certain belief p comes within your purview; you are to determine whether it is probable with respect to what is certain for you in order to determine whether it is acceptable for you. But then you will accept the belief only if you see or believe that it does pass this test. Evidentialism is the claim that religious belief is rationally acceptable only if there are good arguments for it; Locke is both a paradigm evidentialist and the proximate source of the entire evidentialist tradition, [72] from him through Hume and Reid and Kant and the nineteenth century to the present. Locke's classical evidentialism is one element of a larger whole that also includes classical foundationalism and classical deontologism. This connected complex of theses and attitudes has been enormously influential in epistemology since the Enlightenment, and enormously influential especially with respect to our question, the question of the rational justifiability of religious belief: call it the classical package. The classical package includes ways of thinking about faith, reason, rationality, justification, knowledge, the nature of belief, and other related topics. It is hard to overemphasize the importance of these ways of thinking for the de jure question. We have seen how Locke is the fountainhead of the evidentialist tradition, one of the elements in the classical package; but he is also a main source, for us moderns (and postmoderns), of the other two elements: classical foundationalism and classical deontologism. I now turn to them. [73] __________________________________________________________________ A. Classical Foundationalism First, classical foundationalism is foundationalism. Here the crucial notion is that of believing one proposition on the evidential basis of others. Like any important philosophical notion, this one has its problems, complications, and perplexities. Let's ignore them. The notion is serviceable even if it is less than wholly clear, and at any rate there are clear examples. I believe that 32 × 94 is 3008 (I've just calculated it); I believe this proposition on the evidential basis of others, such as 4 × 2 = 8, 4 × 3 = 12, 8 + 2 = 10, and so on. However I don't believe those latter on the evidential basis of any other propositions at all; instead, they are basic' for me. I simply see that they are true, and accept them. I accept many propositions in this basic way: that there is snow in my backyard, for example, and that it is still white. I also believe, in the basic way, that it seems to me that I am seeing something white (I am being appeared to whitely), that I had cornflakes for breakfast, and a thousand other things. The propositions I accept in the basic way are, so to say, starting points for my thought. (This is not to say, of course, that what you take as basic doesn't depend on what else you know or believe. I believe in the basic way that what I see coming toward me is a truck; someone with no acquaintance with trucks or motor vehicles couldn't form that belief at all, let alone hold it in the basic way.) The propositions that I accept in this basic way are the foundations of my structure of beliefs--my noetic structure', as I shall call it for ease of reference. [74] And according to the foundationalist, in an acceptable, properly formed noetic structure, every proposition is either in the foundations or believed on the evidential basis of other propositions. Indeed, this much is trivially true; a proposition is in the foundations of my noetic structure if and only if it is basic for me, and it is basic for me if and only if I don't accept it on the evidential basis of other propositions. This much of foundationalism should be uncontroversial and accepted by all. [75] Further (and still properly uncontroversial), for every proposition in my noetic structure that is not in the foundations, there is an evidential path terminating in the foundations: that is, if A is nonbasic for me, then I believe it on the basis of some other proposition B, which I believe on the basis of some other proposition C, and so on down to a foundational proposition or propositions. [76] Now Locke clearly accepts this much; but he also accepts more. A foundationalist will also typically claim that not just any belief is properly basic; some propositions are such that if I accept them in the basic way, there is something wrong, something skewed, something unjustified about my noetic structure. Imagine, for example, that because of an inordinate admiration for Picasso, I suddenly find myself with the belief that he didn't die; like Elijah, he was directly transported to heaven (in a peculiarly warped sort of chariot with a great misshapen eye in the middle of its side). If I don't believe this proposition on the evidential basis of any others, it is basic for me. But there is something defective, wrong, unhappy in my believing this proposition in the basic way; this proposition is not properly basic. Noting that only some propositions seem to be properly basic, a foundationalist may go on to lay down conditions of proper basicality, admitting some kinds of propositions to this exalted condition and rejecting others. And the classical foundationalist holds that the only propositions that are properly basic for me are the ones that are certain for me. Certainty is another difficult and much contested notion; again, let us ignore the difficulties and contests, noting that classical foundationalists don't always agree as to which propositions are indeed certain in this way. Descartes admits only propositions that are self-evident or incorrigible. Locke accepts these as properly basic; he also adds, as I said earlier, propositions that are evident to the senses'--at least such propositions as something is causing me to have the ideas I do in fact have, and possibly also more robust propositions, such as that the ground is showing through the snow in my backyard. Let's say, a bit vaguely, that according to classical foundationalists, a proposition is properly basic, for a person S, if and only if it is self-evident for S, or incorrigible for S, or evident to the senses for S. Further, according to the classical foundationalists (and everyone else), you can't properly believe just any proposition on the basis of just any other. I can't properly believe, for example, the proposition that Abraham lived around 1800 b.c. on the basis of the proposition that Brutus stabbed Julius Caesar; the latter has nothing to do, evidentially speaking, with the former. Rather, I properly believe A on the basis of B only if B supports A, is in fact evidence for A. Again, this notion of evidential support is difficult and controversial; [77] once more, let us ignore the difficulties and controversies and note that different classical foundationalists propose different evidential relationships as being what is required if my belief of A on the basis of belief B is to be proper. Descartes seems to suggest that a proposition is acceptable in the superstructure of my noetic structure only if I have deduced it from or seen it to be entailed by those in the foundations. This is an extremely strenuous standard (and in fact very few of our beliefs turn out to be acceptable on this standard.) Locke admitted probabilistic support or evidence, and he also admitted testimony. Later on, Charles Sanders Peirce and others went further still and admitted also what he sometimes called abduction'--something like the relationship between a scientific theory and the evidence on which it is based. Stating classical foundationalism at its most capacious, therefore, suppose we put it as follows: (CF) A belief is acceptable for a person if (and only if) it is either properly basic (i.e., self-evident, incorrigible, or evident to the senses for that person), or believed on the evidential basis of propositions that are acceptable and that support it deductively, inductively, or abductively. In a properly run noetic structure, therefore, if you take any belief B that is not basic (not in the foundations), B will be accepted on the basis of other beliefs that are acceptable and that support B (either deductively, inductively, or abductively); if those others are not in the foundations, they will be accepted on the basis of still others that are acceptable and that support them, and so on, down to the foundations--that is, down to propositions that are self-evident, incorrigible, or evident to the senses for you. Classical foundationalism, as I say, has been enormously influential from the Enlightenment to the present. For many philosophers and others (for myself earlier on), it has amounted to a sort of unquestioned assumption, unquestioned because it isn't seen clearly enough even to recognize as an assumption. Locke's views here, particularly with respect to religion, have achieved the status of orthodoxy, and most discussions of the rational justification of religious belief have been and still are conducted in the unthinking acceptance of that framework. There may be modifications of one sort or another, analogical extensions of the original framework, departures of one sort or another; there may be a sort of unease with it, a dimly felt sense that not all is well with it; still, for most of us, the basic framework remains in the near neighborhood of classical foundationalism. __________________________________________________________________ [74] For an account of noetic structures, see WCD, pp. 72ff. [75] Even by coherentists: see WCD, pp. 78ff. [76] See ?Reason and Belief in God,? p. 54. [77] See WCD, pp. 69ff. __________________________________________________________________ B. Classical Deontologism We must now ask a question that has been clamoring for attention all along. Suppose your beliefs don't correspond to the standards the classical foundationalist or evidentialist holds before you: so what? Exactly what is the matter with you? You will be told that your belief structure is unacceptable and not rationally justified, and that you yourself are irrational; but again, so what? What is wrong with being irrational or with holding beliefs that are not rationally justified? It certainly sounds reprehensible, but what, exactly, is the problem? That is what we must know if we are to understand our de jure question. Consider, for example, John Mackie in The Miracle of Theism. [78] He believes he has shown that the central doctrines of theism are not rational or ?rationally defensible? because (as he thinks) he has shown that they are not probable with respect to what he takes to be the relevant evidence. What does he mean here by ?rational?? How is he using this protean term? Suppose he is right in thinking that it would be irrational to be a theist if theistic belief is not probable with respect to the evidence (whatever precisely that is): what is this property of irrationality that would then afflict theism or theists? Mackie doesn't say. And Mackie is not alone in failing to say. Many evidentialist objectors argue that theistic belief is irrational because there is insufficient evidence for it; they clearly think being irrational is a bad business; but they seldom say what's bad about it. Instead, they move immediately to the task of showing, as they think, that there is insufficient evidence for belief in God. This prior question, nevertheless, remains crucial: insufficient for what? What is supposed to be bad about believing in the absence of evidence? Contemporary evidentialist objectors don't (for the most part) explicitly say; their progenitor Locke, however, does say. His question, you recall, is how ?a rational creature, put in that state in which man is in this world, may and ought to govern his opinions, and actions depending thereon.? And his answer, as we have seen, is that a rational creature in our circumstances ought to govern his opinions by reason--that is, proportion his belief to what is certain for him. But how are we to understand the may' and ought' and should' that Locke employs in stating his project? At first sight, his words have a deontological ring; they are redolent of duty, obligation, permission, being within your rights and the rest of the deontological stable. Closer inspection reveals that this is, indeed, how they are to be taken. It is Locke's idea that we have a duty, an obligation to regulate opinion in the way he suggests. We enjoy high standing as rational creatures, creatures capable of belief and knowledge. Noblesse oblige, however; privilege has its obligations, and we are obliged to conduct our intellectual or cognitive life in a certain way. Our exalted station as rational creatures, creatures with reason, carries with it duties and requirements: faith is nothing but a firm assent of the mind: which, if it be regulated, as is our duty, cannot be afforded to anything but upon good reason; and so cannot be opposite to it. He that believes without having any reason for believing, may be in love with his own fancies; but neither seeks truth as he ought, nor pays the obedience due to his Maker, who would have him use those discerning faculties he has given him, to keep him out of mistake and error. He that does not this to the best of his power, however he sometimes lights on truth, is in the right but by chance; and I know not whether the luckiness of the accident will excuse the irregularity of his proceeding. This at least is certain, that he must be accountable for whatever mistakes he runs into: whereas he that makes use of the light and faculties God has given him, and seeks sincerely to discover truth by those helps and abilities he has, may have this satisfaction in doing his duty as a rational creature, that, though he should miss truth, he will not miss the reward of it. For he governs his assent right, and places it as he should, who, in any case or matter whatsoever, believes or disbelieves according as reason directs him. He that doth otherwise, transgresses against his own light, and misuses those faculties which were given him to no other end, but to search and follow the clearer evidence and greater probability. (IV, xvii, 24, pp. 413-14) Here Locke isn't speaking about specifically religious faith (faith as contrasted with reason, say), but about assent or opinion generally; and his central claim here is that there are duties and obligations with respect to its management or regulation. In particular, you are obliged to give assent only to that for which you have good reasons, good evidence: you are to accept a proposition only if it is probable with respect to what is certain for you. Someone who doesn't regulate opinion in this way ?neither seeks truth as he ought, nor pays the obedience due to his Maker? (emphasis added); God commands us to seek truth in this way and to regulate opinion in this way. Someone who does seek truth in this way, even if he should happen to miss it, still ?may have this satisfaction in doing his duty as a rational creature.? You govern your assent ?right,? he says, you place it as you ?should? if you believe or disbelieve as reason directs you. And if you don't do that, then you transgress against your own lights. One who governs his opinion thus is acting in accord with duty, is within his rights, is flouting no obligation, is not blameworthy, is, in a word, justified. The English terms justified', justification', and the like, go back at least to the King James version of the Bible. We are justified, in this use, if Christ's atoning sacrifice for sin has applied to us, so that we are now no longer blameworthy and our sin has been covered, removed, obliterated, taken away; we are no longer guilty; it is as if (so far as guilt is concerned) our sin had never existed. As a matter of fact the term taken in that sense goes back to Wycliffe's 1382 translation of the Bible; the Oxford English Dictionary cites especially Romans 5:16. And Locke is really claiming that you are justified in this sense (guiltless, conforming to your obligations and duties) in believing a proposition p only if p is either certain for you or such that it is probable with respect to propositions that are certain for you. More precisely, your assent to p is justified only if the degree of your assent to p is proportional to the degree to which p is probable with respect to what is certain for you. If you believe in some other way, then you are going contrary to your epistemic obligations; you are guilty; you are flouting epistemic duty. This is the aboriginal and basic idea of the justificationist tradition, the palimpsest in terms of which other justificationist notions are to be understood by way of analogical extensions. And of course there are analogical extensions. For example, if you follow Locke in thinking we have such a duty, then you will be inclined to transfer the term justified' from the believer to the believed, and speak, as in fact we do speak, of a proposition's being justified, or justified for someone, meaning that the person in question has a good bit of evidence for the proposition in question. You will also say, no doubt, that there is a good deal of justification or rational justification for a given proposition, meaning thereby that there is a good deal of evidence for it. [79] __________________________________________________________________ [78] Oxford: Oxford University Press, 1982. [79] There are many other analogical extensions or retrenchments of this original notion of justification, and many other analogically extended uses of the term; see WCD, chapter 1. __________________________________________________________________ [72] In ?Reason and Belief in God,? I suggested that Aquinas was also an evidentialist in this sense; various people (Alfred Freddoso, Norman Kretzmann, Eleonore Stump, Linda Zagzebski, and John Zeis in ?Natural Theology: Reformed?? in Rational Faith: Catholic Responses to Reformed Epistemology, ed. Linda Zagzebski [Notre Dame: University of Notre Dame Press, 1993], p. 72) remonstrated with me, pointing out that things were much more complicated than I thought. The fact is that Aquinas is an evidentialist with respect to scientia, scientific knowledge. But it doesn't follow that he thought a person could properly accept belief in God, say, only if he had (or there are) good theistic arguments. On the contrary, Aquinas thought it perfectly sensible and reasonable to accept this belief on faith. [73] I examine classical foundationalism in detail in WCD and ?Reason and Belief in God?; here I shall be brief and schematic. __________________________________________________________________ III. Back to the Present Locke's thought initiates the classical package: evidentialism, deontologism, and classical foundationalism. It is according to the first two that Christian belief requires evidence; that is, Christian believers are within their intellectual rights and conforming to intellectual duty, only if they have evidence for that belief. It is according to the third that the evidence must trace back, finally, to what is certain for them: what is self-evident or incorrigible or evident to the senses. This connection between justification and evidence has been at the center of the whole justificationist tradition in Western epistemology; it has been of particular importance for subsequent thought about the de jure question for Christian belief. According to this tradition, the de jure question is really the question whether Christian belief is rationally justified--that is, whether believers are justified in holding these beliefs, and whether they are conforming to intellectual duty in holding them. The main intellectual duty, however, is that of proportioning belief to the evidence, to what is certain. Hence the first version of the de jure question gets transformed into a second: do believers have sufficient evidence for their beliefs? We now see the connection between these two forms of the de jure question: the first is the basic question, but if we add (with Locke and the classical tradition) that the main duty here is that of proportioning belief to evidence, then we get the second question. I say Locke's influence--and that of the classical package--has been paramount in discussions of the de jure question. If I am right, we should expect at least two things. We should expect, first, that those who raise the de jure question would put it in terms of evidence, argument, propositional evidence, evidence from other things one thinks. And we should expect, second, that they also put it in terms of justification--justification construed deontologically or in terms of some analogical extension of deontology. Both of these expectations are amply fulfilled. Of course I don't expect you just to take my word for it, and I don't have the space for extensive documentation; I shall instead just give a bit of corroborating evidence. For the last hundred years, W. K. Clifford's essay ?The Ethics of Belief? has been cited in discussions of the de jure question; Clifford (that ?delicious enfant terrible,? as William James called him) claimed (with charming and restrained understatement) that ?it is wrong, always, everywhere, and for anyone to believe anything upon insufficient evidence.? [80] Here we have the combination of deontologism and evidentialism. This passage doesn't display classical foundationalism as well (it doesn't say what the evidence must consist in), but no doubt Clifford was a classical foundationalist; at least he thought that belief in God requires evidence. William James's essay ?The Will to Believe? [81] has been a sort of companion piece to Clifford's; it has been cited for almost as long in discussions of our question, and because James comments on and criticizes Clifford, the two essays have often been anthologized in tandem. James titled his essay ?The Will to Believe?; ?The Right to Believe? would have been more accurate. His central claim is that under certain conditions it is not contrary to duty to believe a proposition (a proposition that isn't certain) even if one has no evidence for it. If believing this proposition is a forced option and a live option, for you, and if there is no evidence against it, then you have a right to believe it, even though you don't have evidence for it. In this way James tries to make room for belief in God (even if not full Christian belief) by inserting it in the gaps of the evidence. The evidentialism and deontologism, again, are evident. [82] James and Clifford wrote a hundred years ago and more; but the last half-century has seen a host of evidentialist objectors to Christian belief, thinkers who hold both that this sort of belief, if it is to be rational, must be accepted on the basis of propositional evidence, and that the evidence is insufficient. (Among them would be Brand Blanshard, [83] Bertrand Russell, [84] Michael Scriven, [85] Antony Flew, [86] Wesley Salmon, [87] J. C. A. Gaskin, [88] Anthony O'Hear, [89] to some degree Richard Gale, [90] and John Mackie in his posthumous book, The Miracle of Theism. [91] ) Although the deontological component in these positions is often more muted than the evidentialism, it is clearly present and sometimes wholly explicit. Thus Blanshard: everywhere and always belief has an ethical aspect. There is such a thing as a general ethic of the intellect. The main principle of that ethic I hold to be the same inside and outside religion. That principle is simple and sweeping: equate your assent to the evidence. [92] Of course it isn't only evidentialist objectors to theistic belief who embrace evidentialism. John Locke himself was an evidentialist, but no evidentialist objector. Locke thought that religious belief is evidence essential' [93] in the sense that it can be rationally accepted only if believed on the basis of good evidence; he also thought the requisite good evidence was available. Several contemporary writers follow in his footsteps: they accept evidentialism, but believe the evidence is forthcoming (or at least aren't sure that it isn't). Among them would be, for example, Basil Mitchell [94] and William Abraham; [95] Stephen Wykstra defends a ?more sensible? evidentialism. [96] Anthony Kenny displays some sympathy for evidentialism; [97] as does Richard Swinburne: ?the use of symbols . . . enables me to bring out the close similarities which exist between religious theories and large scale hypotheses.? [98] Terence Penelhum is no evidentialist, but evidential considerations play a large role in his God and Skepticism; [99] the same can be said for Gary Gutting's Religious Belief and Religious Skepticism. [100] Still, it is the evidentialist objectors that most clearly display evidentialism. John Mackie's The Miracle of Theism is evidentialism at its most formidable; by way of conclusion, then, suppose we briefly note the form evidentialism takes in that book. Mackie proposes to ?examine the arguments for and against the existence of God carefully and in some detail, taking account both of the traditional concept of God and of the traditional proofs of his existence and of more recent interpretations and approaches.? He goes on: If it is agreed that the central assertions of theism are literally meaningful, it must also be admitted that they are not directly verified or directly verifiable. It follows that any rational consideration of whether they are true or not will involve arguments. . . . it [whether God exists] must be examined either by deductive or inductive reasoning or, if that yields no decision, by arguments to the best explanation; for in such a context nothing else can have any coherent bearing on the issue. (pp. 4, 6) Mackie assumes that the rational acceptability of theistic belief depends on the outcome of this examination: if, on balance, the evidence favors theism, then theistic belief is rationally acceptable; if the evidence favors atheism, then theism is not rationally acceptable. The evidentialism, of course, is palpable. Now Mackie takes it that theism is a hypothesis, something like a very large-scale scientific hypothesis (the theory of evolution, perhaps, or general relativity). He assumes further that its rational acceptability depends upon its success as a hypothesis. Speaking of religious experience, he makes the following characteristic remark: ?Here, as elsewhere, the supernaturalist hypothesis fails because there is an adequate and much more economical naturalistic alternative? (p. 198). Clearly, this remark is relevant only if we think of belief in God as or as like a scientific hypothesis, a theory designed to explain some body of evidence, and acceptable to the degree that it succeeds in explaining that evidence. On this way of looking at the matter, there is a relevant body of evidence shared by believer and unbeliever alike; theism is a hypothesis designed to explain that body of evidence; and theism is rationally defensible only to the extent that it is a good explanation thereof. Now Mackie thinks it is not a good explanation: he concludes, ?In the end, therefore, we can agree with what Laplace said about God: we have no need of that hypothesis? (p. 253); he goes on to claim, ?The balance of probabilities, therefore, comes out strongly against the existence of a god.? He clearly takes it for granted, furthermore, that if the balance of probabilities comes out as he says it does, then there is no case for theism, and the theist stands revealed as somehow irrational or intellectually deficient or perhaps intellectually out of line; as he puts it, ?It would appear from our discussion so far that the central doctrines of theism, literally interpreted, cannot be rationally defended.? [101] But why make assumptions like that? Why think that theism is rationally acceptable only if there are good arguments for it? Why think that it is, or is significantly like, a scientific hypothesis? Of course these assumptions form part of the classical package: well, why should we accept that package? Clearly there are sensible alternatives. Consider our memory beliefs, for example: obviously, one could take a Mackie-like view here as well. I believe that I had a banana for breakfast; one could hold that a belief like this (and indeed even the belief that there has been such a thing as the past) is best thought of as like a scientific hypothesis, designed to explain such present phenomena as (among other things) apparent memories; if there were a more ?economical? explanation of these phenomena that did not postulate, say, the existence of the past or of past facts, then our usual beliefs in the past ?could not be rationally defended.? But here this seems clearly mistaken; the availability of such an ?explanation? wouldn't in any way tell against our ordinary belief that there has really been a past. Why couldn't the same hold for theism or, more broadly, for Christian belief? What is to be said for (and against) the classical package, taken in particular with respect to Christian belief? __________________________________________________________________ [80] Lectures and Essays (London: Macmillan, 1901), p. 183. [81] In The Will to Believe, and Other Essays in Popular Philosophy (New York: Longmans, Green, 1897). [82] ?It's not exactly emphasized any longer, but one of James's original purposes in promoting pragmatism was not to get rid of empirically unverifiable beliefs, but to make room, in a scientistic world view, for faith and God. . . . This was explicitly the context for the 1898 lecture? (Louis Menand, ?An American Prodigy,? New York Review of Books [December 2, 1993], p. 33). The ?1898 lecture? is ?The Will to Believe.? [83] Reason and Belief (London: Allen and Unwin, 1974), pp. 400ff. [84] ?Why I Am Not a Christian? in his Why I Am Not a Christian (New York: Simon and Schuster, 1957), pp. 3ff. [85] Primary Philosophy (New York: McGraw-Hill, 1966), pp. 87ff. [86] The Presumption of Atheism (London: Pemberton, 1976), pp. 22ff. [87] ?Religion and Science: A New Look at Hume's Dialogues,? Philosophical Studies 33 (1978), pp. 176ff. [88] The Quest for Eternity: An Outline of the Philosophy of Religion (New York: Penguin, 1984). [89] Experience, Explanation, and Faith: An Introduction to the Philosophy of Religion (London, Boston: Routledge and Kegan Paul, 1984). [90] On the Nature and Existence of God (Cambridge: Cambridge University Press, 1991). [91] Oxford: Oxford University Press, 1982. [92] Reason and Belief, p. 401. More evidence for the pervasive influence of the deontological component of the classical package can be found in chapter 1 of WCD. There I argue that both the prevalence of internalism in contemporary epistemology and the multifarious and confusing array of concepts of justification among contemporary epistemologists can be understood in terms of their relation to deontology. [93] To use Stephen Wykstra's term; see his ?Towards a Sensible Evidentialism: On the Notion of Needing Evidence,'?? in Philosophy of Religion: Selected Readings, 2d ed., ed. William Rowe and William Wainwright (New York: Harcourt Brace Jovanovich, 1989). [94] See his The Existence of God (New York: Oxford University Press, 1981). [95] An Introduction to the Philosophy of Religion (Englewood Cliffs, N. J.: Prentice Hall, 1985). [96] ?Towards a Sensible Evidentialism.? [97] See his Faith and Reason (New York: Columbia University Press, 1983), especially chapters 3 and 4. [98] See his The Existence of God (Oxford: Clarendon Press, 1979). Swinburne, however, can't be considered an evidentialist in view of his ?Principle of Credulity?: ?that (in the absence of special considerations) if it seems (epistemically) to a subject that x is present, then probably x is present? (p. 254). [99] Dordrecht: D. Reidel, 1983. [100] Notre Dame: University of Notre Dame Press, 1982. [101] Indeed, Mackie takes the title of his book from Hume's ironic suggestion that upon the whole, we may conclude that the Christian Religion not only was at first attended with miracles, but even at this day cannot be believed by any reasonable person without one. . . . Whoever is moved by Faith to assent to it, is conscious of a continued miracle in his own person, which subverts all the principles of his understanding, and gives him a determination to believe what is most contrary to custom and experience. An Enquiry Concerning Human Understanding (LaSalle, Ill.: Open Court, 1966), p. 145. __________________________________________________________________ IV. Problems with the Classical Picture The classical picture has been enormously influential in guiding thought about the de jure question; its near relatives still dominate discussion of it; in particular, the evidentialism of the classical picture persists. This picture, however, like some other big pictures, doesn't survive close examination; it is subject to powerful, indeed, fatal objections. After pointing out some of the problems, I'll consider contemporary analogical extensions of the various elements of the classical picture to see whether any of them supports the evidentialism that is still widely popular and finds such a comfortable home in the classical picture. I'll conclude that in fact there is no reason at all to think that Christian belief requires argument or propositional evidence, if it is to be justified. Christians--indeed, well-educated, contemporary, and culturally aware Christians--can be justified, so I shall argue, even if they don't hold their beliefs on the basis of arguments or evidence, even if they aren't aware of any good arguments for their beliefs, and even if, indeed, there aren't any. Indeed, it is obvious that they can be justified in this way; as I shall argue, that suggests that the de jure question we seek is not this question of justification; that question is too easy to answer. So, first, what are these problems attaching to the classical picture? Recent philosophy has not been kind to classical foundationalism; many objections have been raised, many problems pointed out. I shall confine my attention to two objections, both fatal. First, as I've argued elsewhere, [102] classical foundationalism appears to be self-referentially incoherent: it lays down a standard for justified belief that it doesn't itself meet. More exactly, the classical foundationalist, in asserting (and presumably believing) his classical foundationalism, lays down a standard for being justified, blameless, within one's intellectual rights: a standard which his own belief in the classical picture doesn't meet. Stated at slightly greater length, what he claims is that (CP) A person S is justified in accepting a belief p if and only if either (1) p is properly basic for S, that is, self-evident, incorrigible, or Lockeanly evident to the senses for S, [103] or (2) S believes p on the evidential basis of propositions that are properly basic and that evidentially support p deductively, inductively, or abductively. Here I ignore the fact that the believes on the basis of' relation is not transitive. The classical picture doesn't really require that all of one's nonbasic beliefs be believed on the evidential basis of basic beliefs; some nonbasic beliefs may be believed on the basis of other nonbasic beliefs that support them, provided those others are believed on the basis of still other beliefs that support them, provided those others. . . . To put this more accurately, say that a nonbasic belief is properly based if and only if it is believed on the evidential basis of beliefs that are either properly basic or properly based. Then, according to the classical picture, every nonbasic belief must be properly based. Further, I ignore another condition that is really part of the classical picture. Suppose I believe p on the basis of propositions q[1], q[2] . . . q[n] where the q[i] in fact support p, but I can't see that they do. (Perhaps I believe that there is no greatest cardinal on the basis of ordinary axioms for set theory, but don't know, can't see, and have no reason to believe that the latter support the former.) Then, presumably, on the classical picture I am not justified in this belief. My duty is to believe a nonbasic proposition on the basis of propositions that I can see support it, not just any old propositions that happen to support it, whether or not I can see it. Hence perhaps we should add what Locke and Descartes take for granted here: if S is justified in believing p on the basis of other propositions, it must be that those other propositions support p, of course; further, S must also recognize that they do so. __________________________________________________________________ A. Self-Referential Problems Now consider (CP) itself. First, it isn't properly basic according to the classical foundationalist's lights. To be properly basic, it would have to be self-evident, incorrigible, or Lockeanly evident to the senses. But first, it isn't self-evident for the foundationalist (or for the rest of us). Even if someone claims it has some intuitive support, one couldn't with a straight face claim that it has enough intuitive support to be self-evident. For if it were self-evident, it would be such that it isn't even possible for a properly functioning human being to understand it without seeing that it is true. [104] Clearly (CP) isn't like that at all; for example, I understand it, and I don't see that it is true; and I'll bet the same goes for you. In this regard (CP) is wholly unlike 2 + 1 = 3 or If all cats are animals and Maynard is a cat, then Maynard is an animal. Second, it isn't about anyone's mental states and therefore isn't incorrigible for the foundationalist (or any of the rest of us). And third, it obviously isn't evident to the senses. According to (CP) itself, therefore, (CP) is not properly basic. That means that if (CP) is true, those who are within their rights in believing (CP) must believe it on the evidential basis of other propositions--propositions that are properly basic and that evidentially support it. And if they do, in fact, believe it in that way, then there will be good inductive, deductive, or abductive arguments to (CP) from propositions that are properly basic according to (CP). As far as I know, there aren't any such arguments. As far as I know, no classical foundationalist has produced any such arguments or proposed some properly basic propositions that support (CP). It is of course possible that there are such arguments, even if so far no one has produced them; but the probabilities seem to be against it. So probably one who accepts (CP) does so in a way that violates (CP); (CP) lays down a condition for being justified, dutiful, which is such that one who accepts it probably violates it. If it is true, therefore, the devotee of (CP) is probably going contrary to duty in believing it. So it is either false or such that one goes contrary to duty in accepting it; either way, one shouldn't accept it. But couldn't one who accepts (CP) perhaps find a sort of inductive argument for it? [105] Perhaps the defender of (CP) (the classicist', as I'll call her) reads Roderick Chisholm [106] and embraces particularism'; she proposes to develop a criterion of justified belief by assembling samples of justified and samples of unjustified belief and finding a criterion that best fits them. She assembles a reasonably large and representative sample J of cases of beliefs that, as she thinks, are justified, such that the believer is dutiful in accepting them, and another such sample U of beliefs that she takes to be unjustified, accepted in such a way as to flout intellectual duty. Then perhaps she notes that all of the beliefs in J but none in U conform to (CP); she conjectures that a belief is justified if and only if it conforms to (CP). This would be an inductive argument, of sorts, for (CP). Here is the question, however: are its premises properly basic according to the classical picture? The premises include, crucially, the propositions with respect to each member of J that it is justified and of each member of U that it is not justified. What form do such beliefs take? Well, presumably the sample classes would include such propositions as S[1] is justified in believing B[1] in circumstances C[1] and S[2] is not justified in believing P[2] in circumstances C[2]. (The sample classes need not include only actual beliefs, so to speak; they should also include clear cases of beliefs that would be justified in certain circumstances, whether or not anyone has ever actually held such beliefs in those circumstances.) And presumably these are beliefs she accepts in the basic way. (She can't, of course, use (CP) to arrive at them; that would be blatantly circular.) Clearly, beliefs of these sorts aren't either incorrigible or evident to the senses for her; so if they are properly basic, then, according to (CP), they would have to be self-evident. Now here the classicist will be told that she has a really nasty problem: she will be told that there aren't any cases at all where it is self-evident that a belief is unjustified, such that the believer has gone contrary to duty and in fact warrants disapproval and blame. The alleged reason is that our beliefs are not within our direct control; one can't just decide to hold or withhold a belief. If you offer me $1,000,000 to believe that I am under 30, or even to stop believing that I am over 30, there is no way (short of mind-altering drugs, say) I can collect. Still, this is by no means the whole story. A full examination of this question would take us too far afield, but first some of my beliefs are indirectly within my control (in the way in which, for example, my weight is), even if I can't simply decide what to believe and what not to. I can train myself not to assume automatically that people in white coats know what they are talking about; I can train myself to pay more attention to the evidence, to be less credulous and gullible (or less cynical and skeptical), and so on. Furthermore, some of my beliefs or belief states are, in a way, within my direct control. I don't at the moment have a belief on the question of the year of George Washington's birth; a quick look at my encyclopedia or a call to my eighth-grade history teacher would remedy this deficiency. It is therefore directly within my power to bring it about that I have a belief on that topic. We might even go on to say that there is a belief on that topic (the one the encyclopedia reports) such that it is directly within my power to bring it about that I have that belief. Still further, I can be in a state of epistemic sin by virtue of failing to have a certain belief. If it is my responsibility to care for a child and I see her playing with a suspicious looking bottle but don't take the trouble to examine its label, I can't expect to deflect blame by claiming that I didn't know the bottle contained poison. I should have known. (?I didn't know the gun was loaded? doesn't always suffice for self-exculpation; it might be my responsibility to know.) And there are plenty of other ways to be in epistemic sin by virtue of the beliefs you hold or don't hold. I believe that you failed to pay your income taxes last year because X, whom I would have known to be irresponsible had I made any inquiries, said so; and I was in the wrong not to make further inquiries.) I am malicious and wish you ill; the speaker says your thought is deep and rigorous; by virtue of my ill will, I form the belief that what she said is that your thought is weak and frivolous. Out of vanity and pride, I may form the belief that my work is unduly neglected when the fact is it gets more attention than it deserves. And so on. Further, in these cases it is perhaps self-evident that the beliefs in question are unjustified, formed in a way contrary to duty; at any rate I am not prepared to dispute the claim. So suppose we accommodate the classical foundationalist by stipulating that at any rate there are some cases of self-evidently unjustified belief: there still remains a real problem for the classicist. That is because these cases, at least the ones I can think of, lend no support to the claim that it is unjustified to form a belief that is neither properly basic (according to classical standards) nor believed on the basis of such propositions. More important, aren't there cases where a belief is formed according to (CP), but is nevertheless unjustified? I shouldn't form the belief that you failed to pay your taxes last year on the basis of merely casual inquiry; the stakes are too high. But suppose I do just that: your false friend Myrtle tells me you didn't pay them; I believe this in the usual way, a way, let us assume, that conforms to (CP); I am nevertheless unjustified in that belief. And on the other side, aren't there any number of cases where it is self-evident that a belief not formed in accord with (CP) is justified? Someone asks you what you had for breakfast; you reply that it was an orange and some cornflakes. You can't really think of any propositions that are properly basic according to (CP) and support your memory belief; but isn't it self-evident that you are not guilty, not worthy of reproof or blame, in so believing? And of course there will be an enormous number of examples of this sort. And the relevance of this is as follows: if the samples are chosen in any responsible and plausible way (if they are appropriately random'), they will not support that conjecture that a person is conforming to intellectual duty if and only if her beliefs conform to (CP). Hence, I can't see how a devotee of (CP) could responsibly argue for it by way of such an inductive, particularist procedure; and hence I conclude that there is probably no way in which the classicist can argue for (CP). If so, however, then (because she also holds that (CP) is not properly basic) she will be unjustified in believing (CP) if it is true; it is therefore self-referentially incoherent for her. __________________________________________________________________ [104] See WPF, p. 109. [105] See Philip Quinn's ?In Search of the Foundations of Theism,? Faith and Philosophy 2 (1985), pp. 474ff.; my response, ?The Foundations of Theism: A Reply,? Faith and Philosophy 3 (1986), p. 298; and Quinn's rejoinder, ?The Foundations of Theism Again,? in Rational Faith: Catholic Responses to Reformed Epistemology, ed. Linda Zagzebski (Notre Dame: University of Notre Dame Press, 1993), pp. 22ff. I am grateful to Quinn for showing that this possibility needs to be taken much more seriously than I had been taking it. [106] Theory of Knowledge, 3d ed. (Englewood Cliffs, N. J.: Prentice Hall, 1989), p. 7. See also ?Reason and Belief in God,? pp. 75ff. __________________________________________________________________ B. Most of Our Beliefs Unjustified? In his controversies with David Hume, Thomas Reid pointed out that the vast majority of our beliefs do not seem to conform to (CP): at least as far as justification is concerned, they are none the worse for that. This sentiment was echoed in the nineteenth century by others, in particular, Cardinal Newman. Says Newman: Nor is the assent which we give to facts limited to the range of self-consciousness. We are sure beyond all hazard of a mistake, that our own self is not the only being existing; that there is an external world; that it is a system with parts and a whole, a universe carried on by laws; and that the future is affected by the past. We accept and hold with an unqualified assent, that the earth, considered as a phenomenon, is a globe; that all its regions see the sun by turns; that there are vast tracts on it of land and water; that there are really existing cities on definite sites, which go by the names of London, Paris, Florence and Madrid. [107] But how much of this can be seen to be probable with respect to what is certain for us? How much meets the classical conditions for being properly basic? Not much, if any. I believe that I had cornflakes for breakfast, that my wife was amused at some little stupidity of mine, that there really are such external objects' as trees and squirrels, and that the world was not created ten minutes ago with all its dusty books, apparent memories, crumbling mountains, and deeply carved canyons. These things, according to classical foundationalism, are not properly basic; they must be believed on the evidential basis of propositions that are self-evident or evident to the senses (in Locke's restricted sense) or incorrigible for me. Furthermore, they must be probable and seen to be probable with respect to propositions of that sort: there must be good arguments, deductive, inductive, or abductive to these conclusions from those kinds of propositions. If there is any lesson at all to be learned from the history of modern philosophy from Descartes through Hume (and Reid), it is that such beliefs cannot be seen to be supported by, to be probable with respect to beliefs that meet the classical conditions for being properly basic. So either most of our beliefs are such that we are going contrary to epistemic obligations in holding them, or (CP) is false. It certainly doesn't seem that we must be flouting duty in holding these beliefs in the way we do. I believe in the basic way that there is a lot of snow in the backyard just now and that I met my class yesterday; I don't believe either of these things on the basis of propositions that meet the classical conditions for proper basicality; I do not believe there are any propositions of that sort with respect to which they are probable. Of course I realize I could be mistaken; but am I flouting duty in so believing? I reflect on the matter as carefully as I can; I simply see no duty here--and not because I doubt the existence of duties generally, or of epistemic duties specifically. Indeed there are duties of that sort: but is there a duty to conform belief to (CP)? I don't think so. But then how can I be guilty, blameworthy, for believing in this way? Could it be that I escape blame only because of ignorance? As we saw in WCD (pp. 15ff.), there is a distinction to be drawn between subjective and objective duty, a distinction that goes all the way back to the New Testament. The apostle Paul takes up the question whether it is wrong to eat meat sacrificed to idols. Paul holds that this isn't really wrong; however, if someone thinks (mistakenly) that it is wrong, then it is wrong for him to do so: ?I am absolutely convinced, as a Christian, that nothing is impure in itself; only, if a man considers a particular thing impure, then to him it is impure? (Romans 14:14). Certain kinds of actions (e.g., eating meat sacrificed to idols) are objectively permissible: if what makes an action wrong is that God has prohibited it, then these actions have not been prohibited by God. But if I believe they are wrong--say I mistakenly believe they have been prohibited by God--then I am blameworthy if I perform them. Conversely, certain actions in certain situations are objectively wrong; they are not to be done. Still, if I don't know that they are not to be done and justifiably believe that they are permissible, then I am not blameworthy if I do one of them. My objective duty is what I objectively ought to do; my subjective duty is what I (nonculpably) take to be my objective duty. And perhaps the classical foundationalist can take advantage of this distinction as follows: ?True,? he says, ?you are not blameworthy in failing to conform your beliefs to (CP). But that is only because of ignorance. Fortunately for you, you nonculpably can't see that you have a duty to conform your beliefs to (CP); that protects you from blame and guilt; nevertheless, you really do have an objective duty to regulate belief in the fashion I have described, even if you can't see that you do.? Here discussion seems to come to an end. All I can do is ask my interlocutor why he thinks there is such an objective duty and how he came to the knowledge, as he thinks, that there is any such thing. Can he do more than to simply repeat that as a matter of fact we all have this duty? But why should we believe that? What reason is there for thinking it true? Further, I can't properly accept (CP), even if by some wild chance it happens to be true. For if it is true, then to do my duty with respect to accepting it, I must believe it only on the basis of properly basic propositions, and ones such that I can see that they evidentially support (CP). But I don't see that any such propositions support it (and the evidentialist apparently can't help me by, e.g., giving me an appropriate argument). So if it is true and I accept it, I will be going contrary to objective duty; but if I accept it, I will (naturally enough) think it is true, and will therefore believe I am going contrary to my objective duty; hence if it is true and I believe it, I will be going contrary both to objective and subjective duty. __________________________________________________________________ [107] A Grammar of Assent (Notre Dame: University of Notre Dame Press, 1979), p. 149. __________________________________________________________________ [102] ?Reason and Belief in God,? pp. 61ff. [103] Here I am reading Locke (see above, pp. 76-77) as claiming that what I know immediately is only that my sensations are caused by external objects of some kind or other, not that those objects have the properties of trees, horses, or the other sorts of objects we think there are. __________________________________________________________________ V. Christian Belief Justified The classical package taken neat, so to speak, can't be right: there simply doesn't seem to be a duty to form belief in accordance with (CP). Of course there may be other sorts of intellectual duties. There is a duty to the truth of some kind. It may be hard to state this duty exactly; [108] perhaps it is in the neighborhood of a requirement to do your best to believe as many important truths as possible and avoid as many important falsehoods as possible. Whatever precisely our duties to the truth, I want to argue next that Christian belief can certainly be justified and can certainly be justified when taken in the basic way. We are construing justification in a broadly deontological way, so that it includes being within one's epistemic rights and also includes being epistemically responsible with respect to belief formation. (Perhaps you will think the second follows from the first.) This is a perfectly reasonable requirement; if Christian belief cannot be held in such a way as to satisfy it, then there is something wrong with Christian belief. But it isn't at all difficult for a Christian--even a sophisticated and knowledgeable contemporary believer aware of all the criticisms and contrary currents of opinion--to be justified, in this sense, in her belief; and this whether or not she believes in God or in more specific Christian doctrines on the basis of propositional evidence. Consider such a believer: as far as we can see, her cognitive faculties are functioning properly; she displays no noticeable dysfunction. She is aware of the objections people have made to Christian belief; she has read and reflected on Freud, Marx, and Nietzsche (not to mention Flew, Mackie, and Nielsen) and the other critics of Christian or theistic belief; she knows that the world contains many who do not believe as she does. She doesn't believe on the basis of propositional evidence; she therefore believes in the basic way. Can she be justified (in this broadly deontological sense) in believing in God in this way? The answer seems to be pretty easy. She reads Nietzsche, but remains unmoved by his complaint that Christianity fosters a weak, whining, whimpering, and generally disgusting kind of person: most of the Christians she knows or knows of--Mother Teresa, for instance--don't fit that mold. She finds Freud's contemptuous attitude toward Christianity and theistic belief backed by little more than implausible fantasies about the origin of belief in God [109] (patricide in the primal horde? Can he be serious?); and she finds little more of substance in Marx. She thinks as carefully as she can about these objections and others, but finds them wholly uncompelling. On the other side, although she is aware of theistic arguments and thinks some of them not without value, she doesn't believe on the basis of them. Rather, she has a rich inner spiritual life, the sort described in the early pages of Jonathan Edwards's Religious Affections; [110] it seems to her that she is sometimes made aware, catches a glimpse, of something of the overwhelming beauty and loveliness of the Lord; she is often aware, as it strongly seems to her, of the work of the Holy Spirit in her heart, comforting, encouraging, teaching, leading her to accept the ?great things of the gospel? (as Edwards calls them), helping her see that the magnificent scheme of salvation devised by the Lord himself is not only for others but for her as well. After long, hard, conscientious reflection, this all seems to her enormously more convincing than the complaints of the critics. Is she then going contrary to duty in believing as she does? Is she being irresponsible? Clearly not. There could be something defective about her, some malfunction not apparent on the surface. She could be mistaken, a victim of illusion or wishful thinking, despite her best efforts. She could be wrong, desperately wrong, pitiably wrong, in thinking these things; nevertheless, she isn't flouting any discernible duty. She is fulfilling her epistemic responsibilities; she is doing her level best; she is justified. And this is not only true, but obviously true. We may feel in some subterranean way that without evidence she isn't justified; if so, this must be because we are importing some other conception of justification. But if it is justification in the deontological sense, the sense involving responsibility, being within one's intellectual rights, she is surely justified. How could she possibly be blameworthy or irresponsible, if she thinks about the matter as hard as she can, in the most responsible way she can, and she still comes to these conclusions? Indeed, no matter what conclusions she arrived at, wouldn't she be justified if she arrived at them in this way? Even if they are wholly unreasonable, in some clear sense? An inmate of Pine Rest Christian Psychiatric Hospital once complained that he wasn't getting the credit he deserved for inventing a new form of human reproduction, ?rotational reproduction,? as he called it. This kind of reproduction doesn't involve sex. Instead, you suspend a woman from the ceiling with a rope and get her rotating at a high rate of speed; the result is a large number of children, enough to populate a city the size of Chicago. As a matter of fact, he claimed, this is precisely how Chicago was populated. He realized, he said, that there is something churlish about insisting on getting all the credit due him, but he did think he really hadn't gotten enough recognition for this important discovery. After all, where would Chicago be without it? Now there is no reason to think this unfortunate man was flouting epistemic duty, or derelict with respect to cognitive requirement, or careless about his epistemic obligations, or cognitively irresponsible. Perhaps he was doing his level best to satisfy these obligations. Indeed, we can imagine that his main goal in life is satisfying his intellectual obligations and carrying out his cognitive duties. Perhaps he was dutiful in excelsis. If so, he was justified in these mad beliefs, even if they are mad, and even though they result from cognitive dysfunction. [111] Our main quarry, of course, is the de jure objection or question. One prominent candidate is the question whether the Christian believer can be justified in believing as she does. Take that term in its original and basic deontological sense. Then the question is: can the Christian believer be within her epistemic rights and epistemically responsible in forming belief as she does? Can she be justified even if she doesn't believe on the basis of propositional evidence and even if there is no good propositional evidence? The answer to this question is obvious--too obvious, in fact, for it to be the de jure question, at least if that question is to be worthy of serious disagreement and discussion. Of course she can be justified, and my guess would be that many or most contemporary Christians are justified in holding their characteristically Christian beliefs. We must therefore look elsewhere for the de jure question. __________________________________________________________________ [108] See WCD, p. 33. [109] See below, chapter 5, pp. 137ff. [110] Ed. John Smith (New Haven: Yale University Press, 1959, first published 1746), p. 271. [111] Again, what I've really argued is that this believer is subjectively justified. Can the classical foundationalist concede this but claim that he is not objectively justified, that there really is a duty, whether he knows it or not, to believe only on the basis of evidence? But is there even the slightest reason to think there is any such duty? Here at the least the classicist owes us an argument. __________________________________________________________________ VI. Analogical Variations __________________________________________________________________ A. Variations on Classical Foundationalism The classical picture taken neat, therefore, is subject to devastating difficulty. Nowadays, however, it is seldom taken neat. Instead, there are many analogical extensions or analogically related alternatives for each of the three main components of the classical package: the evidentialism, the classical foundationalism, and the deontology. For example, John Mackie [112] retains the evidentialist component, claiming that Christian belief requires evidence on the part of the believer. But Mackie apparently construes evidence much more broadly than the classicist. In his view as in the classical picture, there is a body of knowledge--my evidence--with respect to which a belief must be probable, if it is to be justified; however, this evidence includes much more for Mackie than it does in the classical picture. It includes what is self-evident and incorrigible, of course, but it also includes ordinary perceptual judgments, memory beliefs, some basic science, some of the maxims of probability theory, and so on. Alternatively, we might follow Stephen Wykstra, who concedes that an individual Christian believer doesn't need evidence to be justified; still, Christian belief, he suggests, is evidence-essential in the sense that there must be propositional evidence for it in the Christian community. [113] Or we might go still further, following Norman Kretzmann [114] and broadening the classical requirement in such a way that what is required is only that the believer have evidence of some sort, even if the evidence in question isn't propositional. Sensuous experience might then be evidence for perceptual belief; other sorts of experience, perhaps some of the kinds of experience that go under the rubric religious experience', could also be evidence for Christian belief. These variations are all variations on the classical foundationalist component of the classical picture: according to Mackie and Kretzmann, the believer must have evidence, but evidence is more broadly construed; according to Wykstra, on the other hand, evidence is required, but it need not be possessed by the individual believer, so long as it resides somewhere in the believer's community. Mackie, Kretzmann, and Wykstra retain the evidentialism (with respect to Christian belief) of the classical picture, but modify the foundationalism. It isn't clear whether they accept the deontological component of the classical picture; suppose for the moment we keep that component fixed, modifying only the evidential requirement. It is then obvious, I think, that the believer can be justified even if there aren't good arguments from Mackie-style evidence, even if there isn't good propositional evidence in the community, and even if there isn't evidence in the broad Kretzmann sense. If it seems to me very strongly that the great things of the gospel are true, if upon reading the Scriptures I find myself convinced, and if after considerable reflection--on all the objections, for example--I still find myself convinced, how could I be properly blamed for believing as I do? Again, I could be wrong, deluded, a victim of wishful thinking, subject to some kind of cognitive disorder: nevertheless, there is no duty I am flouting. If the de jure question is whether the believer can be justified, or justified without evidence, the answer is still too easy: of course she can. __________________________________________________________________ [112] The Miracle of Theism. [113] See above, footnote 38. [114] In Our Knowledge of God: Essays on Natural and Philosophical Theology, ed. Kelly Clark (Dordrecht: Kluwer, 1992). __________________________________________________________________ B. Variations on the Deontology The above involved extensions of the classical foundationalist ingredient of the classical picture. Note that we can also ring the analogical changes on the deontological component, and we can mix and match the extensions in a dazzling variety of combinations. I can't possibly examine all these multifarious versions of evidentialism in all their permutations and combinations, [115] but I do wish to examine one particularly salient variety: Alston justification, which is believing on the basis of a reliable ground or indicator. Alston puts it as follows: to be justified in believing that p is to be in a strong position for realizing the epistemic aim of getting the truth. . . . I will begin by making the plausible assumption that to be in an epistemically strong position in believing that p is to have an adequate ground or basis for believing that p. Where the justification is mediate, this ground will consist in other things one knows or justifiably believes. Where it is immediate, it will consist typically of some experience. . . . [116] A belief is justified, therefore, if and only if it is formed on the basis of an adequate ground. Clearly, Alston justification differs radically from the original deontological notion. That is because it doesn't contain so much as a hint of the deontology of the classical picture: ?I reject all versions of a deontological concept [of justification] on the grounds that they either make unrealistic assumptions of the voluntary control of belief or they radically fail to provide what we expect of a concept of justification.? [117] Well then, why does he call what he proposes justification'? Or better, why do I consider it under the rubric justification'? How is it an analogical extension of that notion? The answer is that what it requires--that the belief in question be based on a truth-conducive ground--is an analogical extension of what, according to the classical picture, is the relevant duty. There is a complex and interesting relation between justification taken deontologically, as in the classical picture, and Alston justification (justification as truth-conducive evidence or ground). The latter discards the deontology of the former, but takes the term justification' to denote the condition which, according to the former, is sufficient for satisfying the duty that, according to the former but not the latter, is in fact laid on us human beings. (I'll leave as homework the problem of figuring out how to state this more intelligibly.) Now what sort of animal is a ground of belief? A mediate ground of a belief, according to Alston, is another belief, on the basis of which the belief in question is formed; an immediate ground of a belief is an experience, on the basis of which the belief is formed. And what is it for the ground of a belief to be adequate? ?The ground of a belief will suffice to justify it only if it is sufficiently indicative of the truth of the belief. If the ground is to be adequate to the task, it must be the case that the belief is very probably true, given that it was formed on that basis.? [118] The idea, therefore, is that the ground G of a belief B is adequate just if a certain conditional probability is high: the probability that B is true given that it has been formed on G. And here the probability in question is an objective probability [119] of some sort; if a belief B is justified, then it was formed on the basis of a ground G, such that the objective conditional probability of B on G (P(B/G)) is high. I form the belief that the largest oak in my backyard is now losing its leaves. I form this belief on the basis of experience of some kind--as Alston might state the matter, it seems to me that the tree is presenting itself to me as losing its leaves. Then that belief is justified if and only if it is objectively probable that the tree is losing its leaves, given that I undergo that experience. Putting these elements together, we can say that a belief B is justified--actually, prima facie justified--for S if and only if it is formed on the basis of a truth-conducive ground G--if and only if, that is, it is formed on the basis of some ground G, such that the objective probability that B is true, given that it has been formed on G, is high. __________________________________________________________________ [115] For some of them, see WCD, chapter 1. I leave as homework the problem of showing that Christian belief can indeed be justified on these construals. [116] Perceiving God (Ithaca: Cornell University Press, 1991), p. 73. [117] Ibid. [118] Ibid., p. 75. [119] See WPF, pp. 138ff. __________________________________________________________________ C. Is This the de Jure Question? Have we found the (or a) relevant de jure question? Is the right question to ask the question whether Christian belief is Alston justified? More specifically, the question, for a given Christian belief B I hold--the belief, say, that in Christ, God was reconciling the world to himself--is whether there is some truth-conducive ground G such that I hold the belief in question on the basis of that ground. But is this really a viable de jure question? I want to suggest that it is not. When we ask the de jure question about Christian belief, we are asking whether Christian belief is acceptable, OK, such that a sensible, intelligent, rational, informed person in something like our epistemic circumstances could or would hold such beliefs. The question as to whether such belief sometimes or typically has a truth-conducive ground, however, seems to be a very different question. I have two reasons for thinking so. In the first place, several important sorts of beliefs--a priori belief and memory belief in particular--do not seem to have a ground in Alston's sense at all, but are nonetheless perfectly in order from an epistemic point of view. Consider memory. You remember what you had for lunch: lentil soup and a doughnut. This belief isn't based on propositional evidence. You don't infer it from other things you know or believe, such things, perhaps, as your knowledge that you always have a doughnut and lentil soup for lunch, or your knowledge that it is now shortly after lunchtime and there are doughnut crumbs on your desk and an empty plastic soup dish in your trash. So it doesn't have a mediate ground. But it also isn't based on an experience. At any rate, it is clear that memory beliefs are not based on anything like sensuous experience or phenomenal imagery. [120] There may be a bit of such imagery present (a fragmentary and partial image of a doughnut or a bowl, perhaps), but you certainly don't form the belief on the basis of that image. It is clear that you could remember without having that imagery--or, indeed, any other imagery; some people report that they have no phenomenal imagery associated with memory at all. So the imagery isn't necessary. It is also insufficient; you could also have that imagery without remembering. The reason is that the imagery that goes with imagining that you had a doughnut and lentil soup for lunch, or entertaining the proposition that you did, is indistinguishable (at least in my own case) from the imagery that goes with remembering that you had a doughnut and lentil soup for lunch. And even if you do have fairly explicit phenomenal imagery in connection with this memory, you surely don't know that it was lentil soup on the basis of that imagery; the image isn't nearly clear, detailed, and explicit enough to enable you to distinguish it from, for example, imagery of pea soup, or bean soup, or many other kinds of soup. [121] Accordingly, it isn't that you know it was lentil soup on the basis of this experience; you don't form the belief that it was lentil soup with that experience as ground. (The image seems to be more like a disposable decoration.) Instead, you simply remember, simply form that belief. Or, perhaps more accurately, that belief is formed in you: you don't yourself, so to speak, take much of a hand in forming it. The same goes (though perhaps more controversially) for a priori belief. [122] I believe the proposition Necessarily, if all men are mortal and Socrates is a man, then Socrates is mortal. Now there is, indeed, a sort of imagery connected with this belief when I entertain it--perhaps something like a fragmentary image of the relevant English sentence written on a blackboard as in a logic class. But surely the belief isn't formed on the basis of that imagery; that imagery isn't anything like a ground for it; it doesn't stand to that imagery in anything like the way in which my belief that the snow in my backyard is melting stands to the visual imagery I now enjoy. Indeed, the imagery accompanying that proposition is the same, so far as I can tell, as that which accompanies entertaining Necessarily, if all men are mortal and Socrates is mortal, then Socrates is a man. So many memory and a priori beliefs are not formed on the basis of a ground in Alston's sense, either mediate or immediate. But of course many memory and a priori beliefs are eminently sensible, reasonable, rational, and the like. It therefore follows that a belief need not have a truth-conducive ground to be reasonable, sensible, or rational. Second, there are also beliefs that do have a truth-conducive ground (explained as Alston explains it) but are nonetheless not sensible or reasonable. A belief is based on an adequate ground, says Alston, if and only if it is based on a ground such that it is objectively probable that it is true, given that it is based on that ground. Note that (if objective probability conforms to the probability calculus) a necessary truth will have an objective probability of 1 on any other proposition whatever. Consider therefore the proposition 29 × 38 = 1102: the probability of this proposition is 1 on any condition whatever. Any belief in this proposition on any ground, therefore, is automatically a belief on the basis of an adequate ground. More generally, any grounded belief in any necessary proposition p is justified on this account; for the objective conditional probability that p on any proposition will be 1. So suppose I am extraordinarily gullible when it comes to set theory and believe, say, Cantor's Theorem (according to which the cardinality of any set is always less than that of its power set), not because I have understood a proof or been told by someone competent that it is true, but just because I picked up a comic book on the sidewalk and found therein a character who claims it is his favorite theorem. Then this belief of mine has a truth-conducive ground, but isn't rational or reasonable. Further and closer to current concerns, according to the bulk of the theistic tradition, God is a necessary being who has his most important attributes essentially: there is no possible world in which he does not exist, and none in which he lacks such attributes as omniscience, goodness, love, and the like. If this is true, then the proposition that there is such a being as God (or that he is omniscient, or loving) will have an unconditional objective probability of 1, and consequently an objective conditional probability of 1 on any other proposition. Hence for any ground at all, the probability that one of those beliefs is true, given that it is formed on the basis of that ground, is 1. In asking the de jure question about belief in God, however, we presumably do not mean to ask a question to which an affirmative answer follows just from the fact that God is a necessary being who has his primary attributes essentially. Suppose God is indeed a necessary being; then if I believe in God just to please my friends, or because I am brainwashed or hypnotized, or because I am part of an evil social system, I will be justified in the Alston sense. If so, however, it is too easy to achieve justification in this sense. No doubt there are variations on Alstonian justification, and in a complete treatment we should have to deal with them. But vita brevis est, even if philosophia longa est. I tentatively conclude, therefore, that the de jure question is not the question whether Christian belief is Alston justified. The de jure question is still elusive. __________________________________________________________________ [120] See WPF, pp. 58ff. [121] See WPF, pp. 57ff. [122] See WPF, pp. 104ff. __________________________________________________________________ __________________________________________________________________ [56] Ithaca: Cornell University Press, 1967. [57] In Faith and Rationality, ed. Alvin Plantinga and Nicholas Wolterstorff (Notre Dame: University of Notre Dame Press, 1983). [58] Actually, this is not a traditional philosophical problem in the sense that it is a problem for all philosophers or all positions in philosophy; you will find it pressing only if you accept some version of classical foundationalism. [59] The exception was William James, whose ?The Will to Believe,? in The Will to Believe and Other Essays in Popular Philosophy (New York: Longmans, Green, 1897), was widely anthologized and took the radical line (as it was then perceived) that if religious belief is a live option for you, and a forced option, then believing even without evidence is excusable. See below, p. 89. __________________________________________________________________ 4 Rationality We have seen that the relevant de jure question--the question whether Christian belief is justified, or rational, or reasonable, or intellectually respectable--can't be the question of justification strictly so called. That is, it can't be the Lockean, deontological question whether Christian believers are or can be epistemically responsible, within their epistemic rights, flouting no epistemic duties, in believing as they do. That question, we saw, is much too easy to answer: obviously, a believer--even an intelligent, well-educated, contemporary believer who has heard and considered all the objections--can be justified in this original sense. We saw also that there are analogically extended senses of the term justification'; but none of them is such that it is clear that a Christian believer can't be justified, in that sense, in holding Christian belief. Believers may be mistaken; they may be deluded; they may be foolish; they may be insufficiently critical (in a way that doesn't involve blameworthiness); but there is no reason to think either that they are inevitably derelict in their epistemic duties or that they are unjustified in one of those analogical extensions of the term. __________________________________________________________________ I. Some Assorted Versions of Rationality Of course there are other questions lurking in the nearby bushes, other ways to construe the de jure question. In particular, we can ask whether the believer is rational in believing as she does. Many who put the de jure question or urge a de jure criticism put it in terms of rationality, not justification. (More often, they put it both ways, sometimes just using the one as a synonym for the other.) So suppose we look into this matter: could it be that the appropriate de jure question is whether Christian belief (with or without evidence) is rational? But what is it for a belief to be rational? The first thing to note is that this term is multifarious, indeed, polyphonous, as our postmodern compatriots like to say. There are several importantly different ideas of rationality floating around, and the first thing we have to do is to specify the concept of rationality involved in our question. What are the main conceptions of rationality? In Warrant: The Current Debate (hereafter WCD), I specified some different but analogically related senses of the term. The basic sense is (1) Aristotelian rationality, the sense in which, as Aristotle said, ?Man is a rational animal.? Related to it in various ways are (2) rationality as proper function; (3) rationality as within or conforming to the deliverances of reason; (4) means-ends rationality, where the question is whether a particular means someone chooses is, in fact, a good means to her ends; and (5) deontological rationality. We must look briefly at these; after that, we shall turn at slightly greater length to William Alston's practical rationality. Our task will be lightened, however, by the fact that we have already dealt with (5) in the last chapter. [123] __________________________________________________________________ A. Aristotelian Rationality According to Aristotle, man is a rational animal. Fair enough--on this point as on others, Aristotle is no doubt right: but what specifically did he have in mind? Here the term rational' apparently points to or expresses a property that distinguishes human beings from other animals. As Aristotle saw it, this property is the possession of ratio, the power of reason. The idea is that human beings, unlike at least some other animals, have concepts and can hold beliefs; they can reason, reflect, and think about things, even things far removed in space or time; human beings are (or, at any rate, can be) knowers. This is what it is to be a rational creature; and this is what Aristotle saw as distinctive about human beings. Of course rational powers can come in degrees. We ordinarily think of ourselves (no doubt in a burst of specific chauvinism) as much more talented, along these lines, than other terrestrial animals, although perhaps we are prepared to concede that some of them display at least rudimentary powers of reason. We also realize there may be other creatures, perhaps in other parts of the universe, that put us absolutely in the shade when it comes to intellectual power. Now is the de jure question the question whether a creature rational in this sense can accept Christian belief? Presumably not: given the many millions of rational animals who do accept it, that question, like the question of justification, has much too easy an answer. __________________________________________________________________ B. Rationality as Proper Function If we agree that rational creatures do and therefore can accept Christian belief, we might ask whether it is only malfunctioning rational creatures that do so, creatures whose rational faculties are in some way dysfunctional. A person who suffers from pathological confusion, or flight of ideas, or the manic stage of bipolar disorder, or delusions (perhaps thinking the Martians are out to get him) is said to be irrational. Here the problem is dysfunction, malfunction of the rational faculties. The paranoid doesn't form beliefs in the way a normal, properly functioning human being does; some part of the cognitive apparatus fails to function properly. Pathologically confused people may not know what day it is or where they live. Such dysfunction can be long-term or episodic; if it is the latter, then after the episode is over, we say rationality is restored. This sense of rationality, therefore, has to do with proper function, the absence of dysfunction or pathology: you are rational if not subject to such pathology. Correlatively, irrationality, in this sense, is a matter of malfunction of (some of) the rational faculties, the faculties by virtue of which we are rational animals. So there is an analogical connection between Aristotelian rationality and rationality construed as proper function. We must distinguish two forms of rationality as proper function. On the one hand, there is what we might call internal rationality. We can initially characterize internal rationality as a matter of proper function of all belief-producing processes downstream from experience'. How can we explain this metaphor? We may begin by noting that experience comes in several varieties. First, there is sensuous imagery, the kind of experience you have most prominently in vision but in hearing, smelling, tasting, and touching as well. To use Roderick Chisholm's terminology, in this kind of experience one is appeared to in such and such a way. Sensuous imagery plays an enormously significant role in perception; perceptual beliefs are formed in response to sensuous imagery and on the basis of such imagery. Still, this isn't the only kind of experience that goes with belief formation. In chapter 3 (p. 106) and in Warrant and Proper Function (188-93, hereafter WPF), I pointed out that the formation of memory beliefs is often unaccompanied by phenomenal experience, or else accompanied only by fragmentary, fleeting, indistinct, hard-to-focus sensuous imagery. You remember that you went to a party in Novosibirsk; there is a bit of imagery, all right, although it is fleeting, partial, indistinct, and such that when you try to focus your attention on it, it disappears. But there is another kind of experience present: the belief that it was Novosibirsk (and not, say, Cleveland) seems right, acceptable, natural; it forces itself upon you; it seems somehow inevitable (the right words are hard to find). The belief feels right, acceptable, and natural; it feels different from what you think is a false belief. The same goes for a priori belief. You believe that no dogs are sets. This belief, too, involves little by way of sensuous imagery. When you consider that proposition, perhaps it is as if you catch a momentary and fleeting glimpse of part of a sentence expressing the proposition, or perhaps a fragmentary glimpse of a dog, or perhaps of a dog enclosed within braces; this imagery seems unimportant, however, more like mere decoration than something on the basis of which the belief in question is formed. And here, too, there is also this other sort of experience: it's just seeming true and indeed necessarily true that no dogs are sets. Thinking about this proposition feels different from thinking about the proposition that some dogs (your dog Tietje, for example) are sets. Still a third kind of example, also discussed in more detail in WPF (48ff.): the knowledge that it is you (as opposed to someone else) who is now perceiving the page in front of you. This too is not a matter of sensuous imagery: it is not on the basis of sensuous imagery that you believe it is you who are perceiving that page, rather than your cousin in Cleveland. Here too there is that other sort of phenomenal experience, that feeling that the proposition in question is the right one. Suppose we call this second kind of phenomenal experience doxastic experience because it always goes with the formation of belief. [124] Internal rationality includes, in the first place, forming or holding the appropriate beliefs in response to experience, including both phenomenal imagery and doxastic experience. With respect to the first, I will form beliefs appropriate to the phenomenal imagery I enjoy: for example, when appeared to in the way that goes with seeing a gray elephant, I will not form the belief that I am perceiving an orange flamingo. That sort of response is precluded by internal rationality. But perhaps the second--forming the right beliefs in response to doxastic experience--is more interesting. A pathological skeptic, for example, might have the same sort of doxastic experience as the rest of us, but still be unable to form the appropriate beliefs. I might be appeared to in the way that goes with seeing that Peter is running toward me; out of pathological caution, however, I am unable to believe that he is really running toward me (after all, it could be a cunningly contrived robot, or I could be dreaming, or a brain in a vat, or a victim of some other kind of illusion; and can I be certain that it is really me that he is running toward?). This sort of response is also precluded by internal rationality. By contrast, 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.? [125] That sort of response is not (necessarily) precluded by internal rationality. Perhaps these madmen are subjected to overwhelming doxastic experience here. Perhaps this proposition--that their heads are made of glass--seems utterly obvious to them, as obvious as that 3 + 1 = 4. Then the problem lies with this seeming, with their having this kind of doxastic experience. Given this doxastic experience, what proper function requires (all else being equal) is forming this belief; and that they do. They display external irrationality, but not internal irrationality. There is more that internal rationality requires; we can deal with it briefly. A person is internally rational only if her beliefs are coherent, or at any rate are sufficiently coherent to satisfy proper function. If she is internally rational, then if she believes that her head is made of earthenware, she will not also believe that it is made of flesh and blood--or at least won't believe these both within the confines of the same thought, so to speak. Much more ought to be said about the coherence required by proper function; it will have to await another occasion. Further, an internally rational person will draw the right inferences when the occasion arises: for example, someone who is internally rational but believes that her head is made of earthenware will probably believe that playing football (at any rate without a really good helmet) is very dangerous. Still further, given the beliefs she has, she will make the right decisions with respect to her courses of action--that is, the decisions required by proper function. Given that you do believe you are made of glass, for example, the rational thing to do is to avoid bumps. And finally, if she is internally rational, she will do what proper function requires with respect to such things as preferring to believe what is true, looking for further evidence when that is appropriate, and in general being epistemically responsible. And now that we have internal rationality in hand, external rationality is easy to explain. It requires, first, proper function with respect to the formation of the sensuous experience on which perceptual belief is based. And it consists second in the formation of the right kind of doxastic experience--that is, the sort of doxastic experience required by proper function. I suppose it would be widely conceded that Christian belief can be held by people whose rational faculties are not malfunctioning, or at any rate not malfunctioning in a way that involves clinical psychoses. [126] The fact is many Christian believers are able to hold jobs, some even as academics. (Of course, you may think this latter guarantees little by way of cognitive proper function.) So presumably the de jure question is also not the question whether Christian belief can be held by people whose cognitive or rational faculties are functioning properly, at least in this clinical sense. But this by no means settles the issue; there are subtler forms of cognitive malfunction, and impedance of cognitive proper function. As a matter of fact, the (or a) sensible version of the de jure question does lie in the neighborhood of one of these subtler forms. We'll return to the notion of proper function in more depth and detail in the next chapter, where we explore the notion of warrant. In the meantime, however, suppose we turn to still another kind of rationality. __________________________________________________________________ [124] From doxa, the Greek word for belief. In WPF I called this kind of experience impulsional evidence'. [125] Meditations, Meditation I. [126] Although Richard Rorty somewhere suggests that in the new liberal society, those who think there is such a thing as the chief end of man will have to be considered insane. See also Daniel Dennett, Darwin's Dangerous Idea (New York: Simon and Schuster, 1995), p. 516, where he suggests that perhaps Baptists should be kept in zoos and preserved as interesting cultural relics, but only if they refrain from telling their children such patent untruths as that ?man' is not a product of evolution by natural selection? (519). __________________________________________________________________ C. The Deliverances of Reason First, what are the deliverances of reason? Here we have to take the term reason' a bit more narrowly than we did in thinking about Aristotelian rationality. Among the things we know, some are self-evident. It isn't entirely easy to see what it is for a proposition to be self-evident; [127] the rough idea, however, is that a proposition is self-evident if it is so utterly obvious that we can't even understand it without seeing that it is true. Examples would be propositions like 7 + 5 = 12; if all men are mortal and Socrates is a man, then Socrates is mortal; and if Tom is taller than Sam, and Sam is taller than George, then Tom is taller than George. And the idea is that reason, taken in this narrower sense, is the faculty or power whereby we see the truth of self-evident propositions. Of course it is also reason whereby we see that one proposition entails or implies another: if I learn from the bartender that everyone at the party was drunk, and from you that Paul was at the party, I can conclude that Paul was drunk. The deliverances of reason, therefore, will be self-evident propositions, together with propositions that are self-evident consequences of deliverances of reason. (We might put this by saying that self-evidenceis closed under self-evident consequence.) And then we might say that a proposition is rational if it is among the deliverances of reason, and irrational if its denial is among the deliverances of reason. Note that many propositions--for example, the proposition that Caesar crossed the Rubicon--will then be neither rational nor irrational: neither they nor their denials are among the deliverances of reason. [128] And again, the connection with Aristotelian rationality is easy to see: reason taken in this narrow sense is one of the faculties the possession of which distinguishes us from other animals, and when it is functioning properly, what it yields are the deliverances of reason. There is a problem here. The deliverances of reason obviously come in degrees: some seem much more compelling than others, and only some have the overwhelmingly obvious nature of the propositions mentioned above. So, for example, it is obvious, I think, that there aren't any things that do not exist, although this has been disputed, and although it is not as obvious as the propositions mentioned in the above paragraph. Another example is serious actualism: the proposition that an object has properties only in worlds in which it exists. [129] This proposition has intuitive warrant, intuitive support, and can be deduced from actualism, together with other obvious principles; but it isn't just self-evident. You can understand it and nevertheless reject it, and indeed some philosophers do exactly that. [130] Should we admit these propositions that have at least some intuitive warrant to the august company of deliverances of reason, even if they are not self-evident? Indeed we should; if we do, however, we can no longer say that the deliverances of reason are closed under self-evident entailment. That is because of Russell-like paradoxes. It is a deliverance of reason that there are properties, that there is such a property as self-exemplification, and that every property has a complement, so that there is also such a property as non-self-exemplification. The rest of the sad story is well known. Is Christian belief rational in this sense? No; the central truths of Christianity are certainly not self-evident, nor, so far as anyone can see, are they such that they can be deduced from what is self-evident. Of course, that is nothing whatever against Christian belief; the same holds for, for example, what we are taught by historians, physicists, and evolutionary biologists. So the de jure question can't be the question whether Christian belief is rational in this sense. That is because a negative answer to the question is supposed to be a serious criticism of Christian belief; but it is no criticism of Christian belief (or the theory of evolution, or the belief that you live in Cleveland) that it is not a deliverance of reason in this sense. Well, is Christian belief irrational, in this sense? That is, are the denials of some of the propositions falling within Christian belief either self-evident or deducible from propositions that are self-evident? Could that be the de jure question? If Christian beliefs were irrational in this sense, that would certainly be something against them. Some have certainly argued that characteristic Christian belief is inconsistent. For example, it has often been claimed that the existence of God is incompatible with the existence of evil; Christian doctrine, however, embraces both. I believe it is clear, however, that there is no inconsistency here; [131] in fact those contemporaries who press the problem of evil against Christian or theistic belief no longer make that claim of inconsistency. Some atheologians have also urged that certain Christian doctrines (e.g., the doctrine of the trinity or the doctrine of the incarnation) are self-contradictory and hence inconsistent with the deliverances of reason. But these claims are at best inconclusive; everything depends on which precise formulation of these doctrines we consider. Some of these formulations may perhaps be inconsistent, although it is very hard to find any formulations of these doctrines that are both clearly inconsistent and also widely accepted. (In particular, the formulations to be found in the great creeds of the Christian church are not clearly inconsistent.) Other formulations clearly are not inconsistent. Further, Christians who come to realize that they have accepted an inconsistent version of one of these doctrines can easily replace that version by one that is not inconsistent. So if this were the de jure question, then even if some formulations of central Christian doctrine are contrary to the deliverances of reason, the unhappy condition of believing such a thing could be easily avoided: just move to a formulation that is not inconsistent. But those who urge the de jure question with respect to Christian belief do not, presumably, mean to claim just that Christian belief is inconsistent: even if it is perfectly consistent, they think, there is still something seriously wrong with it. We can't mollify them merely by pointing out that there are consistent versions of Christian belief. This, too, it seems, is not the de jure question. __________________________________________________________________ [127] See WPF, pp. 108ff. [128] Alternatively, we might say that a proposition is irrational if its denial is among the deliverances of reason, and rational if it is not irrational: then, of course, every proposition will be either rational or irrational. [129] See ?Replies,? in Alvin Plantinga, ed. James Tomberlin and Peter van Inwagen (Dordrecht: D. Reidel, 1985), pp. 316ff. [130] See John Pollock, ?Plantinga on Possible Worlds,? in Alvin Plantinga, pp. 126ff., and also Nathan Salmon, ?Nonexistence,? Noűs (September 1998), p. 290. [131] See chapter 9 of my The Nature of Necessity (Oxford: Oxford University Press, 1974). __________________________________________________________________ D. Means-End Rationality What about means-end rationality, what our continental cousins sometimes call Zweckrationalität? This is the sort of rationality displayed by the actions of someone who aims to achieve a certain goal and chooses means that are effective for attaining that goal. Perhaps, more exactly, we should say that this kind of rationality characterizes the actions of a rational creature--one rational in the Aristotelian and proper function sense--who is aiming to achieve a certain goal; so once again we see the connection with the basic Aristotelian sense. Means-end rationality is a matter of knowing how to get what you want; we might think of it as the cunning of reason. [132] If I want to get to Los Angeles as quickly as possible, it would be irrational to take the bus or ride my bicycle: the rational thing to do would be to take the plane. Is the de jure question about this kind of rationality? Means-end rationality is a property of actions; hence it isn't initially obvious that belief is the sort of thing that can be rational and irrational in this sense, because it isn't initially obvious that beliefs are actions. In fact it seems initially obvious that beliefs (believings) are not actions. You don't ordinarily form a given belief because you think holding that belief would be a good means to some end or other. Still, suppose we did think of belief as a sort of action (perhaps in a limiting sense); then presumably the end in view would be believing or knowing the truth. And then Christian belief would be rational in this sense if and only if a rational person, one whose cognitive faculties were functioning properly, would or could choose this means to the end of believing the truth. But there is something very peculiar about this suggestion. What you rationally choose as a means to an end depends on what you believe--for example, on what you believe about the likelihood that a given course of action will yield the result you are aiming at. But what if your aim is to believe truth? Then (pretending for the moment that what you believe is within your power in the appropriate way) you will, of course, believe a proposition if you think it is true: for if it is true, then, naturally enough, believing it is a good way to believe truth. So taking the action of believing Christian teaching will be rational for you, if, in fact, you do believe Christian teaching. (This oddness brings out the way in which belief really isn't action or, if it is, at least isn't much like other forms of action.) The real question, then, will be whether a rational person can believe the claims of Christianity, whether a rational person can accept Christian belief. And that means that the question whether Christian belief is means-end rational really reduces to the question whether it is rational in some other sense: the Aristotelian sense or, more likely, the proper function sense. So we don't have here an independent sense of rationality; because we have already dealt with that sense, what we really see is that the de jure question can't be this question of means-end rationality either. __________________________________________________________________ [132] Of course there are many variations on this notion of rationality. The rational action might be the one that would in fact lead to the achievement of your goal, or the one you think would, or the one you would think would, if your cognitive faculties were functioning properly, or the one you would think would if your cognitive faculties were functioning properly and you reflected long enough, or the one you would think would if . . . and you were sufficiently acute, or . . . and you were not distracted by lust, greed or ambition, and so on. __________________________________________________________________ [123] Deontological rationality is really justification; see above, chapter 3, p. 87. It is worth noting the analogical connection between justification and rationality. __________________________________________________________________ II. Alstonian Practical Rationality None of the varieties of rationality I have so far mentioned offers the resources for a sensible de jure question. In his magisterial book, Perceiving God, [133] William Alston proposes still another way to construe rationality (and the de jure question). Given the power and depth of Alston's account, this one merits a more careful look. __________________________________________________________________ A. The Initial Question The conclusion of Alston's book is really that it is rational--practically rational, as he says--for at least many of us to engage in what he calls Christian Mystical Practice' (CMP, for short): the practice of forming beliefs about God (or the Ultimate) on the basis of experience of God, or more exactly (putative) perception of God (or the Ultimate): My main thesis in this chapter, and indeed in the whole book, is that CMP is rationally engaged in [my emphasis] since it is a socially established doxastic practice that is not demonstrably unreliable or otherwise disqualified for rational acceptance. (194) The sort of rationality at issue here is practical' rationality; we shall therefore consider whether the de jure question might be the question whether Christian belief is practically rational in his sense. Now Alston himself does not really address specifically Christian belief--trinity, incarnation, atonement, and resurrection, for example. He is instead concerned with the sorts of beliefs that are produced by (putative) perception of God. These include such beliefs as that God is glorious, delightful, holy, majestic, all-powerful, loving, and the like, as well as such beliefs as that he is strengthening, supporting or comforting one. These are not specifically Christian beliefs, and, of course, our de jure question is about the rationality of specifically Christian beliefs. Nevertheless, perceptual beliefs about God could contribute support to Christian beliefs about God; in any event, it will be of interest to ask whether Christian belief is rational in the sense of rationality Alston identifies, even if he himself does not address that issue. Further, the question whether Christian belief is practically rational seems to me at any rate closer to the de jure question we seek than the candidates we have already canvassed. But what is this practical rationality'? How does Alston understand this protean notion, and how does he argue for the practical rationality of CMP and the beliefs it produces? __________________________________________________________________ B. Doxastic Practices Here we need a bit of stage setting. A distinctive feature of Alston's entire epistemology is its emphasis upon social doxastic practices--socially established ways of forming belief. (It makes a certain rough sense to think of Alston as judiciously blending Reid with Wittgenstein. [134] ) For example, there is sense perception (hereafter SP), the social practice of forming beliefs on the basis of perception of objects in our environment; there is also the practice of forming beliefs by way of reasoning, both deductive and nondeductive, as well as the practice of forming beliefs on the basis of memory. Together these three form what Alston calls ?the standard package,? perhaps because they are shared by all properly functioning human beings. Further, there is the practice of attributing beliefs, desires, pains and pleasures, affective states, spiritual gifts, and the like to our fellow human beings. Thomas Reid calls this practice (or, rather, the faculty or power that underlies it) sympathy'; we may think of sympathy as part of SP or, if we prefer, as a practice intimately linked with SP, but nonetheless separate and semiautonomous. (If we think of it the latter way, we should consider it part of the standard package.) These are doxastic practices: they issue in the formation of beliefs. They are also social practices in that they contain a considerable component contributed by our social environment. SP, for example, involves a substantial social component in that what we learn from others by way of teaching and testimony becomes part of the practice. For example, what we learn from others is involved in the society of checks and tests whereby we determine whether a putative perception is a real perception; I had to learn from others (parents, for example) what it is that I perceive when I perceive a tree or house or star. The contributions of nature and nurture may vary over these different practices; the contribution of nurture is perhaps maximal with respect to SP and perhaps minimal with respect to our grasp of elementary arithmetic and logic. In addition to these universally shared practices, there is also what Alston calls MP, mystical practice', the practice whereby many but not all of us form beliefs about God (or the Ultimate) on the basis of experience or perception of God (or the Ultimate). CMP is a specific variant of mystical practice, where the beliefs formed are the specifically Christian beliefs held by Christians of all stripes in many different parts of the world and at all times since the beginning of the Christian era. __________________________________________________________________ [134] See Alston's ?A Doxastic Practice Approach to Epistemology,? Knowledge and Skepticism, ed. M. Clay and K. Lehrer (Boulder, Colo.: Westview Press, 1988). __________________________________________________________________ C. Epistemic Circularity Clearly, we can raise many questions about these practices: in particular, we can ask whether they are reliable. We can also ask whether we can show that they are reliable. If we ask this latter question about SP, then we are asking whether we can show or successfully argue that the beliefs formed in this practice are for the most part true or, at any rate, close to the truth, or likely to be true or close to the truth. Our main target, of course, is CMP; but because Alston thinks of CMP as essentially involving perception of God, he attacks the question of the reliability of CMP in tandem with the counterpart question about SP. Alston concedes that we can't give a good noncircular argument for the claim that CMP is in fact reliable. He pays the same compliment to SP, however: we can't give a good noncircular argument for its reliability either. So that distressing fact about CMP is balanced by a complementary distressing fact about SP. The problem with arguments for the reliability of SP is typically what he calls epistemic circularity, a malady from which an argument for the reliability of a faculty or source of belief suffers when one of its premises is such that my acceptance of that premise originates in the operation of the very faculty or source of belief in question. If you give an epistemically circular argument for the reliability of a faculty, then you rely on that very faculty for the truth of one of your premises. An obvious example would be arguing that your intuitive arithmetical faculties are reliable by pointing out that your arithmetic intuitions seem to you to be intuitively sound. A less obviously circular project would be that of trying to determine if human cognitive faculties (including your own) are reliable by doing some science: you find out what human beings think, and then check to see whether what they think is true. Clearly enough, this procedure is epistemically circular, for you rely on human cognitive faculties--yours--in finding out what human beings think and also checking to see if what they think is true. Alston detects many examples of epistemic circularity (more than you might have thought), some obvious and some not so obvious. I believe he succeeds in establishing the important conclusion that it is not possible to show in a noncircular fashion that SP is reliable--at any rate he gets as close to establishing this conclusion as philosophers ever get to establishing any important conclusion. [135] So according to Alston, SP and CMP are in the same leaky epistemological boat. Indeed, the fact is, he argues, all of our basic doxastic practices are in the same epistemological boat; none of them can be shown in noncircular fashion to be reliable. __________________________________________________________________ [135] The argument is given at even greater depth and explicitness in his The Reliability of Sense Perception (Ithaca: Cornell University Press, 1993). __________________________________________________________________ D. The Argument for Practical Rationality The unhappy developments just explained, says Alston, present us with a ?crisis of rationality? and a ?desperate situation?: ?The course of the argument led us to the conclusion that with respect to even those sources of belief of which we are normally the most confident we have no sufficient noncircular reason for taking them to be reliable? (146). What are we to do? Well, we are obliged to settle for second best: although we can't show that any of these practices is reliable, perhaps we can show that we are rational--practically rational--to engage in them. Alston offers two connected arguments for supposing that it is practically rational to engage in these practices. According to the first, in essence, it is perfectly sensible or rational to continue to form beliefs in the SP and CMP ways, because (1) those ways do not lead to massive inconsistencies, (2) there is no reason to think them unreliable, (3) we know of no alternative doxastic practices whose reliability we could demonstrate in an epistemically noncircular fashion, and (4) changing to some other practice would be massively difficult and disruptive. According to the second argument, any socially and psychologically established doxastic practice that meets certain other plausible conditions is prima facie rational (i.e., such that it is prima facie rational to engage in it); such a practice will be all-things-considered rational if, as far as we can see, there is no reason to abandon it. These two arguments are connected, as I shall argue below; it is only the second that he explicitly employs with respect to CMP. Suppose we begin by examining the second argument; as we shall see, this argument leads back to the first. Here is how Alston puts the matter: My main thesis . . . is that CMP is rationally engaged in since it is a socially established doxastic practice that is not demonstrably unreliable or otherwise disqualified for rational acceptance. If CMP is, indeed, a socially established doxastic practice, it follows from the position defended in Chapter 4 that it is prima facie worthy of rational participation. And this means that it is prima facie rational to regard it as reliable, sufficiently reliable to be a source of prima facie justification of the beliefs it engenders. And if, furthermore, it is not discredited by being shown to be unreliable or deficient in some other way that will cancel its prima facie rationality, then we may conclude that it is unqualifiedly rational to regard it as sufficiently reliable to use in belief formation. (194) The basic contention is that it is prima facie rational to engage in CMP . . . because it is a socially established doxastic practice; and that it is unqualifiedly rational to engage in it . . . because we lack sufficient reason for regarding it as unreliable or otherwise disqualified for rational participation. (223) The main premise of this argument, then, is: It is prima facie rational (practically rational) to engage in a socially established doxastic practice, and unqualifiedly rational (rational all things considered) to engage in a socially established practice that doesn't encounter severe internal or external incompatibilities. And in chapters 5 through 7 Alston goes on to argue that CMP is indeed a socially established doxastic practice, and that it does not encounter severe internal or external incompatibilities. __________________________________________________________________ E. Practical Rationality Initially Characterized Turn now to the main premise. First, I don't know precisely how to state the second part, the part about rationality all things considered', but while this is mildly annoying, it isn't really serious, because I intend to comment only on the first part. How exactly are we to understand this proposition, and what is the sense of rational' in which it is prima facie rational to engage in a socially established doxastic practice? As to the second, we are talking about the rationality or lack thereof of taking a course of action, of doing something or other, of acting in a certain way. (That is why we're talking practical rationality.) Whether an action is rational for me will obviously have something to do with what it is I am aiming at in taking that action, what I am trying to accomplish, what my purpose, end, goal is. So the kind of rationality at issue is that means-ends rationality, that Zweckrationalität we came across above (p. 115). The rational action, for me, is the one that will contribute to the realization of my goal, or contribute more to it than any other action open to me. But is it the action that will in fact contribute to my goal that is rational for me, or the one I believe will so contribute? Presumably the second: I am not irrational, in taking a given line of action, if I make a perfectly sensible mistake about what the best means to my end is. [136] If I am thirsty and what I want is a drink of water, it will be rational for me to open the faucet and hold a glass under it; I believe that is a fine way to get a glass of water. It would be irrational for me, under these circumstances, to go (instead) for a walk in the desert; I know that water is hard to find in the desert. On the other hand, if I believed that the faucet isn't connected to any source of water, then the action of opening the faucet wouldn't be a rational way for me to get a drink; and if I believed the nearest water is in the Sonoran Desert just outside Tucson, the action of going for a walk in the desert would be rational. Now the case under consideration, of course, is the case of those doxastic practices; we are to ask whether it is rational to form beliefs by engaging in SP, CMP, or both. Here our relevant aim or goal, says Alston, is that of getting in the right relation to the truth, achieving some appropriate balance between avoiding error and believing truth. And now the question for us is whether a rational way to try to achieve that goal is to form beliefs as we always have, by employing SP, CMP, or both. Of course, this question has about it a certain air of unreality. It is up to me whether I open the faucet to get a drink of water, but it isn't really up to me whether I will form beliefs in accord with SP. I don't have any choice in the matter. And that means that the question of the practical rationality of continuing in SP is a little peculiar. I might as well ask whether it is rational for me to continue to be such that the earth attracts my body with a force that is inversely proportional to the square of the distance between us: this really isn't up to me. The same goes with respect to my major ways of forming belief: it isn't up to me whether I form beliefs in those ways. I can try as hard as I like, but (apart from such draconian measures as mind-altering drugs) I doubt that I could seriously alter my basic belief-forming proclivities. Offer me a million dollars to believe that I live in Wyoming or that I am really the president of the United States: I can strain my utmost, but I won't be able to collect. [137] Alston is perfectly well aware of the problem here, and what he suggests is that the interesting question is whether it would be rational to continue to engage in the practice in question if it were within my power to continue and within my power to refrain (168). The question is what it would be rational for me to do, if I were in a certain position: a position in which one of the things I believe, and believe truly, is that it is within my power to continue to form beliefs in the ways I have (by using the standard package and CMP), and also within my power to refrain from forming beliefs in those ways--either forming no beliefs at all of those sorts, or perhaps using some quite different belief-forming practice. __________________________________________________________________ [136] As we saw above in footnote 10, there are important distinctions within this category. [137] Well, perhaps I do have a bit more control over my belief-forming proclivities than over whether my body is attracted by the earth. The fact is there are rather standard ways in which I can influence, mold, or form my belief-producing tendencies. __________________________________________________________________ F. The Original Position Suppose (with an apologetic bow to John Rawls) we call this position the original position'. Our question, specified to the standard package, is something like this. Suppose I am in the original position: I know (or at any rate believe truly) that it is within my power to stop forming beliefs in the ordinary way, via SP, memory, and reasoning (the standard package). Perhaps I also know that it is within my power to choose some other way of forming belief. Then what would be the rational thing to do: continue to form beliefs as I have all along, try some other way, or give up on this whole belief-forming enterprise? The answer, as we have seen, depends at least in part on my aims, ends, and goals. If my aim is psychological comfort, feeling really good about myself, perhaps I should choose some belief-forming mechanisms that lead me to think I am a really fine fellow. Perhaps I should choose a way that will bring it about that I believe I have just won the Nobel Prize in chemistry, heroically overcoming such serious obstacles as that I have no training in the subject and know next to nothing about it. Naturally I should carefully avoid any belief-forming practices on the basis of which I would come to see the true extent of my failures and ineptness, my sins and miseries, as the Heidelberg Catechism puts it. In the present context, of course, my aim, according to Alston, is not personal comfort, or happiness, or psychological well-being, but getting properly in touch with the truth. The rational course depends on my aims and goals; it also depends on what I believe--that is, what I believe at the time I take the decision in question. If my aim is to feel good about myself, it would be irrational to choose belief-forming mechanisms that, as I believe, would lead to a proper knowledge of my sins and miseries. To make the rational choice, I must figure out which course is most likely to lead to the accomplishment of my goal(s), and then act on that belief by taking that course. And this leads to an important question about the main premise. As you recall, it begins thus: It is prima facie rational (practically rational) to engage in a socially established doxastic practice. . . . But why this emphasis on socially established doxastic practices? True, if in the original position I think socially established practices are especially likely to yield true beliefs, then the rational thing for me to do, in that position, is to choose socially established doxastic practices. But what if I don't think that? I unwisely read Nietzsche, becoming convinced that the common herd is commonly wrong; I develop a lordly Nietzschean disdain for the ways in which the generality of humankind form their belief. Then presumably the rational thing would be to choose practices that are not socially established. I should, instead, choose practices that are enjoyed only by the fortunate few whose Promethean efforts have taken them far beyond hoi polloi. Why is social establishment important or relevant? What counts, for practical rationality, is what I think will achieve my goal; in the original position, it may or may not be the case that I think socially established practices are especially likely to achieve my goal of believing the truth. Here we see the connection between the first and second of Alston's arguments for the practical rationality of SP and CMP. The main premise of the second argument, we might say, takes it for granted that in the original position I believe that socially established practices are as likely to lead us to the right relationship to the truth as any alternative; and indeed I suppose most of us do in fact believe that. The main premise of the first argument is different; it is that I don't know that SP (or CMP) is subject to any massive unreliability, and I also don't know of any alternative practice I could adopt which is such that I could show with respect to it that it is reliable. In the original position (the first argument continues), I would have this belief (I would believe that I know of no better alternative practice); therefore, the rational thing to do is to stick with what I've got. (Or if that seems a bit strong, it is at any rate true that sticking with my present practices is a rational thing to do.) The first argument is the basic one; the second argument takes for granted the main premise of the first argument and then incorporates something else most of us are in fact inclined to believe, namely, that socially established doxastic practices have a good chance of being reliable, perhaps a better chance than idiosyncratic doxastic practices. __________________________________________________________________ G. The Wide Original Position These thoughts lead to a crucial question: precisely what is it that I believe in the original position? In particular, what do I believe about the reliability of SP and CMP in that position? Is the idea that my beliefs, in the original position, are as much as possible like the beliefs I do in fact have, given that (in that position) I know or truly believe that it is within my power to give up SP, CMP, or both? (Call this the wide original position.) Perhaps that is the way to think of the original position. But this doesn't take us very far. The fact is that I now believe that both SP and CMP are reliable. Therefore, if my beliefs in the original position are the ones I do in fact have, the question as to the rational course is easily answered: obviously, I should continue to form beliefs in the way I have been forming them. My aim is to be in the right relationship to the truth; I propose to attain as good a mixture of achieving the truth and avoiding error as possible; but in fact I believe that SP and CMP offer a vastly better chance to achieve that goal than any alternative I can think of; therefore, the rational choice for me to make, obviously enough, is to continue both in SP and in CMP. Here there is a strong odor of triviality. I do in fact think both SP and CMP are reliable; so if, in the original position, I have the beliefs about SP and CMP that in fact I do have, then in that position, naturally enough, the rational choice would be to continue with SP and CMP. Given what I do believe about them, that would be the rational thing to do. This conclusion, while no doubt true, is pretty weak tea. Of course, if I knew I could refrain from forming beliefs in the SP and CMP way, and also believed that those ways were reliable, more reliable than any alternative way open to me, I would choose to continue to form beliefs that way. True, but not very interesting: how would this fact show or tend to show that my SP and CMP beliefs are in fact rational, in some interesting sense? We are told that if we knew it was within our power to continue to form beliefs in this way, and also within our power to abstain from so doing, then if we believed that SP and CMP are reliable, the rational thing to do would be to choose to continue to form beliefs in those ways. No doubt: nothing of interest follows. The same would go for any beliefs I have, no matter how crazy. The same would go, for example, for the insane beliefs of Descartes's madmen, who believed that they themselves were gourds--zucchini, perhaps, or summer squash--and that their heads were made of pottery. If I really do believe that I am a summer squash, then the rational thing for me to do, if offered the chance, is to continue to form beliefs in a way that yields (as I see it) this true belief. Still, that doesn't show that this belief itself is rational. We haven't yet located the de jure question. However we do have to consider another facet of the dialectical situation, one that so far I have been slighting: I am aware, in the original position, of the fact that neither SP, nor CMP, nor any other major doxastic practice can be noncircularly shown to be reliable. That, after all, is what, according to Alston, precipitated the crisis of rationality and called forth the question of rationality in the first place. It is after we realize this, he thinks, that we are in the desperate situation of which he speaks. So we must add that in the original position I am aware of the fact that we can't noncircularly establish that the practices in question are reliable. (We must also add, perhaps, that I have devoted some attention to this fact, have thought about it at least a bit; perhaps we should say that I am acutely aware of it.) This changes very little. In the original position as now conceived (the wide original position), I know that it is within my power to withhold perceptual and Christian belief; I also know that it isn't possible to give a good noncircular argument for the reliability of these sources of belief; but otherwise my beliefs are as much as possible like they are in fact. And our question remains: what would be the rational thing to do: continue with SP and CMP, or stop forming beliefs in those ways? Again, however, the answer is too easy: of course the rational thing would be to continue with SP and CMP. Once more, this is because I am in fact convinced that these sources of belief are reliable. True enough: I realize that I can't give a good noncircular argument for their reliability, but this gives me no pause. I can't see that this puts us in a desperate situation or that it should lead to a crisis of rationality: for this situation is a necessary feature of any doxastic condition. Not even God himself, necessarily omniscient as he is, can give a noncircular argument for the reliability of his ways of forming beliefs. [138] God himself is trapped inside the circle of his own ideas. About all we can say about God's ways of forming beliefs is that it is necessary, in the broadly logical sense, that a proposition p is true if and only if God believes p. [139] Of course God knows that and knows, therefore, that all of his beliefs are true. However (naturally enough), he knows this only by virtue of relying on his ways of forming beliefs. If, per impossible, he became a bit apprehensive about the reliability of those ways of forming beliefs, he would be in the same boat as we are about that question. He couldn't give an epistemically noncircular argument for the reliability of his ways of forming beliefs; for the beliefs constituting the premises of any such argument would themselves have been formed in those ways. But any epistemic debility that afflicts a necessarily omniscient being is hardly worth worrying about. In the wide original position, therefore, I would be convinced that SP and CMP are reliable sources of belief, despite the fact that I realize it isn't possible to give a good noncircular argument for their reliability; hence, in the wide original position, the rational thing to do, obviously, would be to continue with them. We are still mired in triviality. We still don't have either the de jure question or the original position quite right. The problem is that if, in the original position, we have the beliefs we actually have with respect to SP and CMP, then it is trivially obvious that the rational decision would be to continue to form beliefs in those ways. Unfortunately, the fact that this is the rational decision, given those beliefs, does nothing to show that the beliefs we form on the basis of SP and CMP are rational in any interesting sense. In particular, the atheologian who raises the de jure question with respect to Christian belief will not be mollified if told that it would be rational, given that you thought CMP reliable, to decide to continue to form beliefs in the CMP way. __________________________________________________________________ [138] Here I assume what Alston disputes: that God has beliefs. (Of course, on Alston's view, there would be something like beliefs in God.) But this is really irrelevant to the point I make here, which is that it is a necessary truth that no doxastic agent, no matter how exalted, could give a good, epistemically noncircular argument for the reliability of his doxastic faculties. [139] See my ?Divine Knowledge,? in Christian Perspectives on Religious Knowledge, ed. C. Stephen Evans and Merold Westphal (Grand Rapids: Eerdmans, 1993). __________________________________________________________________ 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-ŕ-vis belief formation. [140] 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'). [141] 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. [142] And how do I know that we have been doing this for some time? Presumably by way of memory. Memory 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. [143] 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. [144] 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 of 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? [145] Why should SP have to prove itself before the bar of reason? [146] 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, [147] 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. [148] 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 beings 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 itself. 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.? [149] 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 ideal 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. __________________________________________________________________ [140] 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. [141] 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. [142] 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. [143] 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. [144] 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. [145] ?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). [146] 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?' [147] 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. [148] As outlined in my so-far-unpublished ?Two Dozen or So Good Theistic Arguments.? [149] Meditations, Meditation I. __________________________________________________________________ [133] Ithaca: Cornell University Press, 1991. Subsequent page references will be to this book. __________________________________________________________________ __________________________________________________________________ 5 Warrant and the Freud-and-Marx Complaint The genius of a man capable of explaining religion seems to me to be of a higher order than that of a founder of religion. And that is the glory to which I aspire. Charles DuPuis What we have seen so far is what the de jure question and criticism are not: it is not the complaint that the believer is not within her intellectual rights in believing as she does; it is not the complaint that she has no good argument from propositions that are self-evident, about her own mental states, or evident to the senses for her; it is not the complaint that she has no good argument of some other sort; it is not the complaint that her Christian belief lacks Alstonian justification, or means-end rationality; and it is not the complaint that it isn't practically rational to decide to continue to form belief on the basis of experience. None of these criticisms has much of a leg to stand on. So the de jure criticism has proven elusive. In the last chapter, however, we did finally catch a glimpse of our quarry--no more than a glimpse, though--and in this chapter I want to look further into the nature of this style of criticism, in part by trying to come to an understanding of the rejection of religious belief associated with Freud and Marx. Then I will point out the connection between the de jure question, properly understood, and warrant, the subject of the two preceding books in this series. In the next few chapters, I will consider more explicitly the question whether Christian belief can have warrant even if it doesn't receive it by way of argument or propositional evidence. This is really the question (as I might have put it in ?Reason and Belief in God? [150] ) whether belief in God and Christian belief more generally can be properly basic--properly basic with respect to warrant. (It is also the question I was raising [rather inchoately] in God and Other Minds. [151] ) Perhaps another way to put this question is to ask whether Christian belief can get warrant, not by argument but by virtue of (broadly construed) religious experience. __________________________________________________________________ I. The F&M Complaint As we have seen, atheologians (those who argue against Christian belief) have often claimed that Christian belief is irrational; so far, we have failed to find a sensible version of this claim. But perhaps we can make progress by exploring the animadversions on Christian belief proposed by Freud, Marx, and the whole cadre of their nineteenth- and twentieth-century followers. [152] We could also examine here Nietzsche's similar complaint: that religion originates in slave morality, in the ressentiment of the oppressed. As Nietzsche sees it, Christianity both fosters and arises from a sort of sniveling, cowardly, servile, evasive, duplicitous, and all-around contemptible sort of character, which is at the same time envious, self-righteous, and full of hate disguised as charitable kindness. (Not a pretty picture.) I've chosen not to consider Nietzsche for two reasons: first, he really has little to add to what Marx and Freud say; second, he is harder to take seriously. He writes with a fine coruscating brilliance, his outrageous rhetoric is sometimes entertaining, and no doubt much of the extravagance is meant as overstatement to make a point. Taken overall, however, the violence and exaggeration seem pathological; for a candidate for the sober truth, we shall certainly have to look elsewhere. [153] Now Freud, Marx, and their many epigoni (and anticipators) criticize religious belief; they purport to find something wrong with it; they are masters of suspicion' and (at any rate in their own view) unmask it. And in examining their critical comments on religious belief, I think we can finally locate a proper de jure question: one that is distinct from the de facto question, is such that the answer is nontrivial, and is relevant in the sense that a negative answer to it would be a serious point against Christian belief. The first order of business, therefore, is to try to get clear as to what the Freud-Marx critical project (the F&M complaint', as I shall call it) really is. __________________________________________________________________ A. Freud There are several sides to Freud's critique of religion. For example, he was fascinated by what he saw as the Darwinian picture of early human beings coming together in packs or herds (like wolves or elk), all the females belonging to one powerful, dominant, jealous male, and he tells a dramatic story about how religion arose out of an extraordinary interaction among the members of that primal horde: The father of the primal horde, since he was an unlimited despot, had seized all the women for himself; his sons, being dangerous to him as rivals, had been killed or driven away. One day, however, the sons came together and united to overwhelm, kill, and devour their father, who had been their enemy but also their ideal. After the deed they were unable to take over their heritage since they stood in one another's way. Under the influence of failure and remorse they learned to come to an agreement among themselves; they banded themselves into a clan of brothers by the help of the ordinances of totemism, which aimed at preventing a repetition of such a deed, and they jointly undertook to forgo the possession of the women on whose account they had killed their father. They were then driven to finding strange women, and this was the origin of the exogamy which is so closely bound up with totemism. The Totem meal was the festival commemorating the fearful deed from which sprang man's sense of guilt (or original sin'). . . . . . . This view of religion throws a particularly clear light upon the psychological basis of Christianity, in which, as we know, the ceremony of the totem meal still survives, with but little distortion, in the form of Communion. [154] Strong stuff, this, displaying Freud's redoubtable imaginative powers and his ability to tell a sensational story; [155] all the elements--sex, murder, cannibalism, remorse--of a dandy Hollywood spectacular are here. Taken as a serious attempt at a historical account of the origin of religion, though, it has little to recommend it and is at best a wild guess, much less science than science fiction. [156] But perhaps Freud didn't intend it as sober and literal truth. (He himself calls it a vision'.) Perhaps it is something like a parable, maybe something like how some Christians understand early Genesis or Job, meant to illustrate and present a truth in graphic but nonliteral form. (Maybe here as elsewhere Freud is under the spell of biblical ways of writing and thinking.) And just as it isn't always easy to draw the right moral from a biblical parable, so it isn't easy to see what Freud intends us to gather from this gripping if grisly little tale. In any event, Freud offers quite a different account of the psychological origins of religious (theistic) belief: These [religious beliefs], which are given out as teachings, are not precipitates of experience or end-results of thinking: they are illusions, fulfillments of the oldest, strongest and most urgent wishes of mankind. The secret of their strength lies in the strength of those wishes. As we already know, the terrifying impressions of helplessness in childhood aroused the need for protection--for protection through love--which was provided by the father; and the recognition that this helplessness lasts throughout life made it necessary to cling to the existence of a father, but this time a more powerful one. Thus the benevolent rule of a divine Providence allays our fear of the dangers of life; the establishment of a moral world-order ensures the fulfillment of the demands of justice, which have so often remained unfulfilled in human civilization; and the prolongation of earthly existence in a future life provides the local and temporal framework in which these wish-fulfillments shall take place. [157] As we see, there is more to Freud's critique than phantasmagoric fables about the primal horde. The idea is that theistic belief arises from a psychological mechanism Freud calls wish-fulfillment'; the wish in this case is father, not to the deed, but to the belief. Nature rises up against us, cold, pitiless, implacable, blind to our needs and desires. She delivers hurt, fear, and pain; in the end, she demands our death. Paralyzed and appalled, we invent (unconsciously, of course) a Father in Heaven who exceeds our earthly fathers as much in power and knowledge as in goodness and benevolence; the alternative would be to sink into depression, stupor, paralysis, and finally death. According to Freud, belief in God is an illusion in a semitechnical use of the term: a belief that arises from the mechanism of wish-fulfillment. This illusion somehow becomes internalized. [158] An illusion (as opposed to a delusion), says Freud, is not necessarily false; and he goes on to add that it isn't possible to prove that theistic belief is mistaken. Nevertheless, there is more here than a mere antiseptic comment on the origin of religion. Although religion originates in the cognitive mechanism of wish-fulfillment, Freud apparently believes that it is within our power to resist this illusion, and that there is something condemnable, something intellectually irresponsible, in failing to do so: If ever there was a case of a lame excuse we have it here. Ignorance is ignorance; no right to believe anything can be derived from it. In other matters no sensible person will behave so irresponsibly or rest content with such feeble grounds for his opinions and for the line he takes. . . . Where questions of religion are concerned, people are guilty of every possible sort of dishonesty and intellectual misdemeanour. [159] Psychoanalysis, furthermore, provides arguments against the truth of religious belief: ?If the application of the psycho-analytic method makes it possible to find a new argument against the truths [sic] of religion, tant pis for religion . . . .? (p. 37). Once we see that religious belief takes its origin in wishful thinking, we will presumably no longer find it attractive; perhaps this will also induce in us a certain pity for those benighted souls who will never rise to our enlightened heights: The whole thing is so patently infantile, so incongruous with reality, that to one whose attitude to humanity is friendly, it is painful to think that the great majority of mortals will never be able to rise above this view of life. [160] Freud hopes and expects that we human beings will eventually give up religious belief, once we are clear about its origin, in favor of a view of the world that is closer to the actual facts of the matter: I am reminded of one of my children who was distinguished at an early age by a peculiarly marked matter-of-factness. When the children were being told a fairy story and were listening to it with rapt attention, he would come up and ask: ?Is that a true story?? When he was told it was not, he would turn away with a look of disdain. We may expect that people will soon behave in the same way towards the fairy tales of religion. . . . [161] The fundamental theme here, therefore, is that religious belief arises from wish-fulfillment. We shall have to try to see more exactly what this amounts to and what bearing, if any, it has on the rationality of Christian belief; first, however, we should briefly note Marx's rather similar criticism. __________________________________________________________________ [154] ?An Autobiographical Study,? in volume 20 of the Standard Edition of the Complete Psychological Works of Sigmund Freud (London: Hogarth Press and the Institute of Psychoanalysis, 1953-74), p. 68. See also Totem and Taboo, authorized translation by James Strachey (New York: W. W. Norton, 1950 [originally published in 1913]), pp. 140ff. [155] Freud tells a similarly fantastic story about how we human beings tamed fire--?a quite extraordinary and unexampled achievement,? he says--and turned it to our use: Psychoanalytic material, incomplete as it is and not susceptible to clear interpretation, nevertheless admits of a conjecture--a fantastic-sounding one--about the origin of this human feat. It is as though primal man had the habit, when he came in contact with fire, of satisfying an infantile desire connected with it, by putting it out with a stream of his urine. The legends that we possess leave no doubt about the originally phallic view taken of tongues of flame as they shoot upwards. Putting out fire by micturating--a theme to which modern giants, Gulliver in Lilliput and Rabelais' Gargantua, still hark back--was therefore a kind of a sexual act with a male, an enjoyment of sexual potency in a homosexual competition. The first person to renounce this desire and spare the fire was able to carry it off with him and subdue it to his own use. By damping down the fire of his own sexual excitation, he had tamed the natural force of fire. This great cultural conquest was thus the reward for his renunciation of instinct. Further, it is as though woman had been appointed guardian of the fire which was held captive on the domestic hearth, because her anatomy made it impossible for her to yield to the temptation of this desire. (Civilization and Its Discontents, tr. and ed. James Strachey [New York: W. W. Norton, 1961 (originally published in 1930 as Das Unbehagen in der Kultur)], p. 37) [156] Here see, e.g., Wilhelm Schmidt, The Origin and Growth of Religion: Facts and Theories, tr. H. J. Rose (New York: L. MacVeagh, Dial Press, 1931), p. 114, who makes an attempt to evaluate this story as serious science; see also Evan Fales, ?Scientific Explanations of Mystical Experiences, Part I: The Case of St. Teresa,? Religious Studies 32, no. 1 (June 1996), p. 148. [157] The Future of an Illusion, tr. and ed. James Strachey (New York: W. W. Norton, 1961), p. 30. This work was originally published as Die Zukunft einer Illusion (Leipzig: Internationaler Psychoanalytischer Verlag, 1927). [158] And in such a way that it (or its deliverances) rather resembles Calvin's sensus divinitatis (chapter 6, below); see Moses and Monotheism (New York: Vintage, 1967), pp. 167ff. [159] The Future of an Illusion, p. 32. [160] Civilization and Its Discontents, p. 21. [161] The Future of an Illusion, p. 29. Freud isn't unambiguously sanguine on this point; he thinks there are three powers (religion, art, and philosophy) that challenge the claims of science to cognitive supremacy, and of these three only religion ?is to be taken seriously as an enemy? (22:160). __________________________________________________________________ B. Marx Marx's most famous pronouncement on religion: The basis of irreligious criticism is man makes religion, religion does not make man. In other words, religion is the self-consciousness and the self-feeling of the man who has either not yet found himself, or else (having found himself) has lost himself once more. But man is no abstract being squatting outside the world. Man is the world of man, the state, society. This state, this society, produce religion, a perverted world consciousness, because they are a perverted world. . . . Religious distress is at the same time the expression of real distress and the protest against real distress. Religion is the sigh of the oppressed creature, the heart of a heartless world, just as it is the spirit of a spiritless situation. It is the opium of the people. The abolition of religion as the illusory happiness of the people is required for their real happiness. The demand to give up the illusions about its condition is the demand to give up a condition which requires illusions. The criticism of religion is therefore in embryo the criticism of the vale of woe, the halo of which is religion [Marx's emphasis]. [162] Marx suggests that religion arises from perverted world consciousness--perverted from a correct, or right, or natural condition. Religion involves a cognitive dysfunction, a disorder or perversion that is apparently brought about, somehow, by an unhealthy and perverted social order. Religious belief, according to Marx, is a result of cognitive dysfunction, of a lack of mental and emotional health. The believer is therefore in an etymological sense insane. Because of a dysfunctional, perverse social environment, the believer's cognitive equipment isn't working properly. If his cognitive equipment were working properly--if, for example, it were working more like Marx's--he would not be under the spell of this illusion. He would instead face the world and our place in it with the clear-eyed apprehension that we are alone, and that any comfort and help we get will have to be of our own devising. [163] And here we can see an initial difference between Freud and Marx: Freud doesn't necessarily think religious belief is produced by cognitive faculties that are malfunctioning. Religious belief--specifically belief in God--is, indeed, produced by wish-fulfillment; it is the product of illusion; still, illusion and wish-fulfillment have their functions. In this case, their function is to enable us to get along in this cold and heartless world into which we find ourselves thrown. How then is this a criticism of religious belief? Freud speaks elsewhere of a ?reality principle.? Beliefs produced by wish-fulfillment aren't oriented toward reality; their function is not to produce true belief, but belief with some other property (psychological comfort, for example). So we could initially put it like this: religious belief is produced by cognitive processes whose function is not that of producing true beliefs, but rather that of producing beliefs conducive to psychological well-being. We will look into this in more detail below; for the moment, perhaps what we can say is that the Marxist criticism of religious belief is that it is produced by disordered cognitive processes, while the Freudian criticism is that it is produced by processes that are not aimed at the production of true beliefs. __________________________________________________________________ [162] ?Contribution to the Critique of Hegel's Philosophy of Right, Introduction,? in On Religion, by Karl Marx and Friedrich Engels, tr. Reinhold Niebuhr (Chico, Calif.: Scholar's Press, 1964), pp. 41-42. Engels echoes Marx: All religion, however, is nothing but the fantastic reflection in men's minds of those external forces which control their daily life, a reflection in which the terrestrial forces assume the form of supernatural forces. In the beginnings of history it was the forces of nature which were first so reflected and which in the course of further evolution underwent the most manifold and varied personifications among the various peoples. . . . But it is not long before, side by side with the forces of nature, social forces begin to be active--forces which confront man as equally alien and at first inexplicable, dominating him with the same apparent natural necessity as the forces of nature themselves. . . . At a still further stage of evolution, all the natural and social attributes of the numerous gods are transferred to one almighty god, who is but a reflection of the abstract man. Such was the origin of monotheism. . . . It is still true that man proposes and God (that is, the alien domination of the capitalist mode of production) disposes. . . . What is above all necessary for this is a social act. And when this act has been accomplished, when society, by taking possession of all means of production and using them on a planned basis, has freed itself and all its members from the bondage in which they are now held by these means of production which they themselves have produced but which confront them as an irresistible alien force; when, therefore man not only proposes, but also disposes--only then will the last alien force which is still reflected in religion vanish; and with it will also vanish the religious reflection itself, for the simple reason that then there will be nothing left to reflect. (Anti-Dühring, pp. 147-49 in On Religion) [163] There is another possibility as to how to understand Marx here: see below, p. 162. __________________________________________________________________ C. Others We must take a deeper look at these claims. First, however, we should note that although Freud and Marx often get the credit for this alleged unmasking (perhaps with a crumb thrown in the direction of Nietzsche), its essence is to be found much earlier. Jean-Jacques Rousseau (1712-78) thought Christian belief was a product of corrupt society, and that the natural spirituality of our souls has been damaged by a Christianized civilization; he thus anticipates Marx in seeing Christian belief as a result of cognitive malfunction resulting from social malfunction. David Hume, a British contemporary of Rousseau, anticipates Freud: It must necessarily, indeed, be allowed, that, in order to carry men's intention beyond the present course of things, or lead them into any inference concerning invisible intelligent power, they must be actuated by some passion, which prompts their thought and reflection; some motive, which urges their first enquiry. But what passion shall we here have recourse to, for explaining an effect of such mighty consequences? Not speculative curiosity, surely, or the pure love of truth. That motive is too refined for such gross apprehensions; and would lead men into enquiries concerning the frame of nature, a subject too large and comprehensive for their narrow capacities. No passions, therefore, can be supposed to work upon such barbarians, but the ordinary affections of human life; the anxious concern for happiness, the dread of future misery, the terror of death, the thirst of revenge, the appetite for food and other necessaries. Agitated by hopes and fears of this nature, especially the latter, men scrutinize with a trembling curiosity, the course of future causes, and examine the various and contrary events of human life. And in this disordered scene, with eyes still more disordered and astonished, they see the first obscure traces of divinity. [164] What is crucial here is the claim that religious belief does not arise from the pure love of truth', but from other sources: desire for happiness, fear of death, and the like. In fact Hume ironically suggests that Christian belief is so contrary to experience and to the ?principles of understanding? (i.e., the deliverances of reason) that a reasonable person can accept it only by virtue of a miracle: upon the whole, we may conclude that the Christian Religion not only was at first attended with miracles, but even at this day cannot be believed by any reasonable person without one. . . . Whoever is moved by Faith to assent to it, is conscious of a continued miracle in his own person, which subverts all the principles of his understanding, and gives him a determination to believe what is most contrary to custom and experience. [165] So the fundamental thrust of Hume's suggestion, as of Freud's, is that religious belief doesn't emerge from the segment of our whole cognitive economy that is, as we might put it, aimed at the production of true belief; it comes, instead, from a desire for security or a fear of death or whatever. And of course what underlies Hume's ironic jape is the idea that Christian belief goes directly contrary to the deliverances of reason and experience. Many of our contemporaries also see religious beliefs in these terms. Thus Northrop Frye weighs in on Marx's side, but employs Freudian or semi-Freudian categories: speaking of ?the curious aberration of believing the Bible',? he says: such belief is really a voluntarily induced schizophrenia, and probably a fruitful source of the infantilism and hysterical anxieties about belief which are so frequently found in the neighborhood of religion, at least in its more uncritical areas. [166] In the same vein, we have Don Cupitt: ?Theological realism can only be actually true for [i. e. thought to be true by] a heteronomous consciousness such as no normal person ought now to have.? [167] Those who claim that they really are theological realists' (i.e., claim that they really do believe in God), he says, are hypocrites [168] or have succumbed to ?a kind of madness.? [169] Cupitt seems to think that (perhaps, as they say, given what we now know') you would have to be psychotic to actually be a theological realist (one who believes that there really is such a person as God); if you are not psychotic but nonetheless profess theistic belief, then you must be one of those hypocrites Christian churches are supposed to be full of. A final witness. Charles Daniels agrees with Freud in finding the origin of religious belief in wishful thinking: we must begin to entertain suspicions that the explanation for these [religious] experiences does not lie in any perceived religious reality, but is rather the effect of some other cause--perhaps excessive emotion and fervor. . . . It is not at all difficult, however, to construct a plausible explanation not consisting of mere possibilities like the machinations of demons, why people should come very strongly to believe there to be a divinely populated religious reality which is perceived in religious experience even when there is none . . . we very much want there to be an understandable order to the universe, we very much want our lives to be of consequence, and we very much want to know in practical detail what's right and wrong. Religion addresses what we very much want. The universe has an intelligible order because there is an intelligent powerful God who made it. We are important because God made us (as Christians say, ?in his image?) and gave us the faculties of understanding and free, intelligent action. [170] __________________________________________________________________ [164] David Hume: The Natural History of Religion, ed. H. E. Root (Stanford, Ca.: Stanford University Press, 1957), p. 166. [165] An Enquiry Concerning Human Understanding (La Salle, Ill.: Open Court, 1956), p. 145. [166] Speaking of infantilism, Frye's intemperate comments call to mind schoolyard debating styles (perhaps about fifth grade): ?Oh Yeah? Well, the trouble with you is you're crazy, and so's your whole dumb family!? [167] Taking Leave of God (New York: Crossroad, 1980), p. 12. One gathers that Cupitt thinks it is ?our modern form of consciousness? that makes this obligatory. [168] Ibid., p. 21. [169] The World to Come (London: SCM Press, 1982), p. 83. [170] ?Experiencing God,? Philosophy and Phenomenological Research (1989), pp. 497, 499. __________________________________________________________________ D. How Shall We Understand the F&M Complaint? Now the F&M (Freud-and-Marx) complaint is, naturally enough, a complaint, a (negative) criticism of religious belief, including Christian belief. But the general project under which the efforts of Freud and Marx fall is that of giving naturalistic explanations of religious belief, explanations that don't involve the truth of the beliefs in question or the truth of any other supernaturalistic beliefs or hypotheses. Many (in addition to those cited above) have joined them in this effort, and by now there is quite a variety of naturalistic explanations of religious belief. [171] But of course giving a naturalistic account of a kind of belief isn't automatically a criticism of that kind of belief. Consider a priori belief, belief in such propositions as the laws of logic, perhaps, or the basic truths of arithmetic, or the proposition that if all cats are animals, and Maynard is a cat, then Maynard is an animal. Perhaps it is possible to give a naturalistic' account of our knowledge of these truths: an account, that is, that stands in the same relation to them as a naturalistic account of religious belief stands to it. Such an account would not invoke the truth of these a priori beliefs as part of the explanation; it would proceed instead by outlining certain salient features of the causal genesis or antecedents of these beliefs, perhaps pointing to events of some kind in the nervous system. The existence of a causal explanation, of this sort, of a priori belief would not show or tend to show that such beliefs are unreliable. The same would go for religious belief. To show that there are natural processes that produce religious belief does nothing, so far, to discredit it; perhaps God designed us in such a way that it is by virtue of those processes that we come to have knowledge of him. Suppose it could be demonstrated that a certain kind of complex neural stimulation could produce theistic belief. This would have no tendency to discredit religious belief--just as memory is not discredited by the fact that one can produce memory beliefs by stimulating the right part of the brain. Clearly, it is possible both that there is an explanation in terms of natural processes of religious belief (perhaps a brain physiological account of what happens when someone holds religious beliefs), and that these beliefs have a perfectly respectable epistemic status. If we are to have a criticism of religion by way of a naturalistic explanation, what we need is something that in some way discredits religious belief, casts doubt on it, shows that it is not epistemically respectable--in a word, shows that there is something wrong with it. And the criticism, of course, is that religious belief (including Christian belief) is irrational. But irrational in just what way? What exactly is wrong with religious belief, according to the F&M complaint? How, exactly, shall we understand the F&M complaint? First, an assumption underlying it. Going all the way back to Plato and Aristotle, it has been assumed that there are intellectual or cognitive or rational powers or faculties, or (possibly) virtues: for example, perception and memory. Joining the computer craze, we might say that these faculties have inputs and outputs; their outputs are beliefs. It is these processes that produce in us the myriad beliefs we hold. These faculties are also something like instruments; and, like instruments, they have a function or purpose. If we thought of ourselves as created and designed either by a Master Craftsman or by evolution, these cognitive faculties would be the parts of our total cognitive establishment or total cognitive design whose purpose it is to produce beliefs in us. Their overall purpose, furthermore, is presumably to produce true beliefs in us; to put it a bit less passively, they are designed in such a way that by using them properly we can come to true belief. Our cognitive faculties work over a surprisingly large area to deliver beliefs of many different topics: beliefs about our immediate environment; about the external world at large; about the past; about numbers, propositions, and other abstract objects and the relations between them; about other people and what they are thinking and feeling; about what the future will be like; about right and wrong; about God. These faculties and processes are the instruments or organs, as we might put it, whereby we come to have knowledge. They are aimed at the truth in the sense that their purpose or function is to furnish us with true belief. Like any other instruments or organs, they can work properly or improperly; they can function well or malfunction. A wart or a tumor doesn't either malfunction (although it might be by virtue of malfunction in some system that the tumor is present) or function properly: it doesn't have a function or purpose. But an organ--your heart, for example, or liver or pancreas--does have a function, and does either work properly or malfunction. And the same goes for cognitive faculties or capacities: they too can function well or ill. The condition in which they function really badly is insanity; of course there are much milder, less intrusive forms of cognitive malfunction. Now among these faculties one of the most important is reason. Taken narrowly, reason is the faculty or power whereby we form a priori beliefs, beliefs that are prior to experience or, better, independent, in some way, of experience. [172] These beliefs include what in chapter 4 we called the deliverances of reason: first of all, simple truths of arithmetic and logic, such as 2 + 1 = 3 and if all men are mortal and Socrates is a man, then Socrates is mortal. They also include such beliefs as that nothing can be red all over and also green all over and that to be a person you must at least be potentially capable of forming beliefs and having ends or aims. Still further, they include more controversial items, such as the belief that there are properties, states of affairs, propositions, and other abstract objects, and the belief that no object has a property in a possible world in which it doesn't exist. (So I say, anyway; there are those who disagree.) The deliverances of reason also include beliefs that obviously follow from deliverances of reason. [173] And still further, reason is the power or capacity whereby we see or detect logical relationships among propositions. There are other faculties or rational powers that have as their purpose the production of true beliefs in us; [174] for example, there are perception and memory, which, along with reason, constitute the standard package of chapter 4. Further, there are introspection, by which I learn such things about myself as that I am appeared to a certain way, and believe this or that; induction, whereby (in a way that defies explicit statement) we come to expect the future to be like the past in certain respects, thereby being able to learn from experience; [175] and Thomas Reid's sympathy, whereby we come to be aware of what other people are thinking, feeling, and believing. Still further, there is testimony or credulity, whereby we learn from others, by believing what they tell us. By sympathy I learn that you are telling me that your name is Archibald; for me to believe you, however, something further is required. (Thus by perception, I see that you are in such and such a bodily state; by sympathy, I learn that you are claiming that your name is Archibald; and by testimony, I believe you.) The Enlightenment looked askance at testimony and tradition; Locke saw them as a preeminent source of error. The Enlightenment idea is that perhaps we start by learning from others--our parents, for example. Properly mature and independent adults, however, will have passed beyond all that and believe what they do on the basis of the evidence. But this is a mistake; you can't know so much as your name or what city you live in without relying on testimony. (Will you produce your birth certificate for the first, or consult a handy map for the second? In each case you are of course relying on testimony.) As Thomas Reid puts it: I believed by instinct whatever they [my ?parents and tutors?] told me, long before I had the idea of a lie, or a thought of the possibility of their deceiving me. Afterwards, upon reflection, I found they had acted like fair and honest people, who wished me well. I found that, if I had not believed what they told me, before I could give a reason for my belief, I had to this day been little better than a changeling. And although this natural credulity hath sometimes occasioned my being imposed upon by deceivers, yet it hath been of infinite advantage to me upon the whole; therefore, I consider it as another good gift of Nature. [176] In addition to the cognitive powers or rational faculties mentioned so far there may be others that are more controversial. For example, we seem to have a moral sense: certain kinds of behavior and certain kinds of character seem wrong, bad, to be avoided; others seem right, good, fitting, to be promoted. It is obviously wrong (all else being equal) to hurt young children or to refuse to care for your aging parents; perhaps we see this by way of a sort of moral sense. (It is no doubt because this moral sense can malfunction, or atrophy, that inability to tell right from wrong is a legal defense.) My point here is not to argue that indeed there is a moral sense, although I believe that there is, but rather to note that there could well be truth-aimed faculties in addition to the ones mentioned so far. Similarly a believer in God might think that there is such a thing as Calvin's sensus divinitatis, [177] a natural, inborn sense of God, or of divinity, that is the origin and source of the world's religions; perhaps there is also such a thing as the inward invitation or instigation of the Holy Spirit (to anticipate chapter 8) whereby the believer comes to accept the central truths of the Christian faith. As we have seen, these rational faculties can function either properly or improperly. We ordinarily take it for granted that when our cognitive faculties are functioning properly, when they are not subject to dysfunction or malfunction, then, for the most part, the beliefs they produce are true, or close to the truth. If your perceptual faculties are functioning properly, what you think you see is probably what you do see. (If you are suffering from delirium tremens, all bets are off.) There is, we might say, a presumption of reliability for properly functioning faculties; we are inclined (rightly or wrongly) to take it that properly functioning cognitive faculties for the most part deliver true belief. Of course there will be mistakes and disagreements, and we may be inclined to skepticism about various areas of belief: political beliefs, for example, as well as beliefs formed at the limits of our ability, as in particle physics and cosmology; but the bulk of the everyday beliefs delivered by our rational faculties, so we think, are true. At any rate, the deliverances of our rational faculties, taken broadly, comprise our best bet for achieving truth. Returning finally to the F&M complaint, it's clear that it has to do with the deliverances of our rational faculties. Freud and Marx acquiesce in the presumption of reliability; they assume (as do we all) that when our rational faculties are functioning properly and are used properly, then for the most part their deliverances are true, or at any rate close to the truth. Of course, as we saw, it is possible for cognitive faculties to function well or ill. The insane beliefs of Descartes's madmen [178] were due to cognitive malfunction of some sort. There are more subtle ways, however, in which nonrational or irrational beliefs can be formed in us. First of all, there are belief-forming processes or mechanisms that are aimed, not at the formation of true belief, but at the formation of belief with some other property--the property of contributing to survival, perhaps, or to peace of mind or psychological well-being in this sometimes dangerous and threatening world of ours. [179] Those with a lethal disease may believe their chances for recovery much higher than the statistics in their possession would warrant; again, the function of the relevant process would not be that of furnishing true beliefs but of furnishing beliefs that make it more likely that the believer will recover. A mountaineer whose survival depends on his ability to leap a crevasse may form an extremely optimistic estimate of his powers as a long-jumper; it is more likely that he will be able to leap the crevasse (or at least give it a try) if he thinks he can than if he thinks he can't. Most of us form estimates of our intelligence, wisdom, and moral fiber that are considerably higher than an objective estimate would warrant; no doubt 90 percent of us think ourselves well above average along these lines. [180] A person may be blinded (as we say) by ambition, failing to see that a certain course of action is wrong or stupid, even though it is obvious to everyone else. Our idea, here, is that inordinately ambitious people fail to recognize something they would otherwise recognize; the normal functioning of some aspect of their cognitive powers is inhibited or overridden or impeded by that excessive ambition. You may be blinded also by loyalty, continuing to believe in the honesty of your friend long after an objective look at the evidence would have dictated a reluctant change of mind. You can also be blinded by covetousness, love, fear, lust, anger, pride, grief, social pressure, and a thousand other things. In polemic, it is common to attack someone's views by claiming that the denial of what they think is patently obvious (i.e., such that any right-thinking, properly functioning person can immediately see that it is so); we then attribute their opposing this obvious truth either to dishonesty (they don't really believe what they say; after all, who could?) or to their being blinded by something or other--maybe a reluctance to change, an aversion to new ideas, personal ambition, sexism, racism, or homophobia. Thus according to Judith Plaskow, ?If the Rabbinical Assembly Law Committee cannot see that it is reflecting and supporting a long history of religious homophobia (Jewish and otherwise), then it is either willfully blind or patently dishonest.? [181] In a similar vein, Richard Dawkins insists (in a recent review in the New York Times), ?It is absolutely safe to say that if you meet someone who claims not to believe in evolution, that person is ignorant, stupid, or insane (or wicked, but I'd rather not consider that).? [182] Dawkins apparently thinks the truth of evolution is utterly clear and obvious to anyone who is not unduly ignorant, is not too stupid to follow the arguments, and is sane (i.e., with rational faculties that are functioning properly); it is therefore obvious that all who aren't just (wickedly) lying through their teeth would have to admit that they believe in evolution. What are appealed to in all these cases are mechanisms that can override or cancel what our rational faculties would ordinarily deliver, substituting a belief that is either contrary to what unimpeded rational faculties would deliver, or at any rate distinct from what reason would deliver. What we see, therefore, is that there are at least three ways in which a belief can fail to be a proper deliverance of our rational faculties: it may be produced by malfunctioning faculties, by cognitive processes aimed at something other than the truth, or by faculties whose function has been impeded and overridden by lust, ambition, greed, selfishness, grief, fear, low self-esteem, and other emotional conditions. [183] Accordingly, a belief can fail to be a proper deliverance of our rational faculties by way of malfunction and by way of being produced by a process that is not aimed at the production of true belief. And here we come to the heart of the F&M objection: when F&M say that Christian belief, or theistic belief, or even perhaps religious belief in general is irrational, the basic idea is that belief of this sort is not among the proper deliverances of our rational faculties. It is not produced by properly functioning truth-aimed cognitive faculties or processes. It is not produced by belief-producing processes that are free of dysfunction and whose purpose it is to furnish us with true belief. And this means that the presumption of the reliability of properly functioning cognitive faculties does not apply to the processes that yield belief in God or Christian belief more broadly. The fundamental idea is that religious belief has a source distinct from those of our faculties that are aimed at the truth. Alternatively, if religious belief does somehow issue from those truth-aimed faculties, their operation, when they function in such a way as to produce religious belief, is overridden and impeded by something else: a need for security, or for feeling important in the whole scheme of things, or for psychological comfort in the face of this pitiless, intimidating, and implacable world we face. Just what sort of deviation from the norm does religious belief present? Here Freud and Marx seem to diverge. Although Marx has relatively little to say about religion, there is of course that famous passage I quoted above (pp. 140-41); he seems to hold that what our rational faculties teach us (when they are unimpeded by that cognitive dysfunction produced by a perverted social order) is that there is no God and no religious meaning to life. There is no Father in Heaven to turn to and no prospect of anything, after death, but dissolution. The fundamental idea is that religious belief is irrational in a double sense: first, it is a product of cognitive faculties that are malfunctioning in response to social and political disorder; second, what these faculties produce when malfunctioning in this way is contrary to the deliverances of our rational faculties--that is, contrary to what they deliver when they function properly. For Freud, too, the main point is that theistic and religious belief, or theistic belief insofar as it is religious, does not arise from the proper function of truth-aimed cognitive processes or faculties, but rather from wishful thinking. [184] This is the force of Freud's claim that religious belief is an illusion. Of course, illusions have their functions, and a place in the human cognitive design plan; they may serve important ends, such as the end Freud thinks religious belief serves. Nevertheless, such cognitive processes as wishful thinking are not aimed at the production of true beliefs. Beliefs produced by wishful thinking are therefore irrational or nonrational in the sense that they are not produced by our rational faculties; they are not produced by truth-aimed cognitive processes. Like Marx, however, Freud thinks religious belief is also irrational in a stronger sense. Such belief runs contrary to the deliverances of our rational powers; they are ?patently infantile? and ?foreign to reality.? The F&M criticism, then, is that religious belief is not produced by cognitive faculties that are functioning properly and aimed at the truth. And this, I think, leads us finally to a viable de jure question. Those who raise this question are not interested first of all in the truth of Christian belief: their claim is that there is something wrong with believing it. Christian belief may be true, and it may be false; but at any rate it is irrational to accept it. They are best construed, I think, as complaining that Christian belief is not produced by cognitive faculties functioning properly and aimed at the truth. Now what this suggests (at least to anyone who has taken a look at the first two volumes in this series) is warrant. Freud and Marx, from the perspective of those volumes, are really complaining that theistic belief and religious belief generally lack warrant. And the de jure criticism, so it seems to me, is best construed as the claim that Christian belief, whether true or false, is at any rate without warrant. __________________________________________________________________ [171] See, for example, J. Samuel Preus, Explaining Religion (New Haven: Yale University Press, 1987). [172] See my Warrant and Proper Function (hereafter WPF), chapter 6. [173] But see above, chapter 4, p. 114. [174] For more detail, see WPF, chapters 3-9. [175] See WPF, pp. 122ff. [176] Essays on the Intellectual Powers of Man, in Thomas Reid's Inquiry and Essays, ed. R. Beanblossom and K. Lehrer (Indianapolis: Hackett, 1983), VI, 5, pp. 281-82; see also WPF, pp. 77ff. [177] See below, chapter 6. [178] Above, p. 133. [179] See WPF, pp. 11ff. [180] I can't resist repeating (from WPF, p. 12) a couple of passages from Locke: Would it not be an insufferable thing for a learned professor, and that which his scarlet would blush at, to have his authority of forty years standing wrought out of hard rock Greek and Latin, with no small expence of time and candle, and confirmed by general tradition, and a reverent beard, in an instant overturned by an upstart novelist? Can any one expect that he should be made to confess, that what he taught his scholars thirty years ago, was all errour and mistake; and that he sold them hard words and ignorance at a very dear rate? (An Essay concerning Human Understanding, ed. A. D. Woozley [New York: World Publishing, 1963], IV, xx, 11) And Let never so much probability land on one side of a covetous man's reasoning, and money on the other, it is easy to foresee which will outweigh. Tell a man, passionately in love, that he is jilted; bring a score of witnesses of the falsehood of his mistress, tis ten to one but three kind words of hers, shall invalidate all their testimonies . . . and though men cannot always openly gain-say, or resist the force of manifest probabilities, that make against them; yet yield they not to the argument. (Ibid., IV, xx, 12) [181] ?Burning in Hell, Conservative Movement Style,? Tikkun (May-June 1993), pp. 49-50. Recall in this connection Don Cupitt's charge that those who claim to accept ?theological realism? (i.e., those who claim to believe that there really is such a person as God) are ?hypocrites or psychotics?--the former, presumably, if they merely claim to be theological realists, and the latter if they really are. [182] New York Times, April 9, 1989, sec. 7, p. 34. Daniel Dennett goes Dawkins one (or two) better, claiming that one who so much as harbors doubts about evolution is ?inexcusably ignorant? (Darwin's Dangerous Idea [New York: Simon and Schuster, 1995], p. 46)--thus displaying both ignorance and wrongdoing. [183] This last (perhaps we can call it impedance') is not strictly a case of malfunction, but for present purposes I shall include it under malfunction. [184] Freud thinks of reason as the aggregate of those faculties (and he thinks of them as the ones involved in the pursuit of science); his idea, furthermore, is that reason taken this way is the only means we have for achieving the truth. Displaying that touching confidence in science characteristic of the Enlightenment, Freud assumes that scientific reason will enable us to achieve the truth in areas where for centuries we wandered in darkness; more modestly, perhaps reason so taken gives us our best shot at the truth. Ironically enough, there is excellent reason to doubt that Freud's characteristic contributions themselves constitute science in any sensible sense; see Adolf Grünbaum's The Foundations of Psychoanalysis (Berkeley: University of California Press, 1984). __________________________________________________________________ [152] Of course, it wasn't only Christian belief that drew their fire: Freud and Marx were equal-opportunity animadverters, attacking religion generally and without discrimination. [153] I don't mean for a moment to dispute Merold Westphal's contention (in Suspicion and Faith: The Religious Uses of Modern Atheism [Grand Rapids: W. B. Eerdmans, 1993]) that Christians have something to learn from Nietzsche (as from Freud and Marx). Of course they do, but the same lessons can be learned at a much subtler level from, for example, the Bible--where, as Westphal points out, Nietzsche's criticisms, insofar as they are on the mark, are anticipated. Taken as a serious account of the origin of Christianity, however, Nietzsche's intemperate scoldings can't really be seen as a serious contribution to the subject. __________________________________________________________________ II. Warrant: The Sober Truth I've said most [185] of what I have to say about the nature of warrant in Warrant: The Current Debate (WCD) and WPF. To spare the reader a trip to the library, however, I will briefly recapitulate; readers who want more depth and detail should consult those volumes (although on pp. 156ff. below I make a correction to what is said in WCD and WPF). The question is as old as Plato's Theaetetus: what is it that distinguishes knowledge from mere true belief? What further quality or quantity must a true belief have, if it is to constitute knowledge? This is one of the main questions of epistemology. (No doubt that is why it is called theory of knowledge'.) Along with nearly all subsequent thinkers, Plato takes it for granted that knowledge is at least true belief: you know a proposition p only if you believe it, and only if it is true. But Plato goes on to point out that true belief, while necessary for knowledge, is clearly not sufficient: it is entirely possible to believe something that is true without knowing it. You are congenitally given to pessimism; you believe that the stock market will plunge tomorrow, even though you have no evidence; even if you turn out to be right, you didn't know. You have traveled two thousand miles to the North Cascades for a climbing trip; you are desperately eager to climb. Being an incurable optimist, you believe it will be bright, sunny, and warm tomorrow, despite the forecast, which calls for high winds and a nasty mixture of rain, sleet, and snow. As it turns out, the forecasters were wrong, and tomorrow turns out sunny and beautiful: your belief was true, but didn't constitute knowledge. Suppose we use the term warrant' to denote that further quality or quantity (perhaps it comes in degrees), whatever precisely it may be, enough of which distinguishes knowledge from mere true belief. Then our question (the subject of WPF): what is warrant? My suggestion (WPF, chapters 1 and 2) begins with the idea that a belief has warrant only if it is produced by cognitive faculties that are functioning properly, subject to no disorder or dysfunction--construed as including absence of impedance as well as pathology. The notion of proper function is fundamental to our central ways of thinking about knowledge. But that notion is inextricably bound with another: that of a design plan. [186] Human beings and their organs are so constructed that there is a way they should work, a way they are supposed to work, a way they work when they work right; this is the way they work when there is no malfunction. There is a way in which your heart is supposed to work: for example, your pulse rate should be about 50 to 80 beats per minute when you are at rest and (if you are under age forty) achieve a maximum rate of some 180 to 200 beats per minute when you are exercising really hard. If your resting pulse is 160, or if you can't get your pulse above 60 beats per minute no matter how hard you work, then your heart isn't functioning properly. (Then again, a bird whose resting heart rate is 160 might be perfectly healthy.) We needn't initially take the notions of design plan and way in which a thing is supposed to work to entail conscious design or purpose. I don't here mean to claim that organisms are created by a conscious agent (God) according to a design plan, in something like the way in which human artifacts are constructed and designed (although in fact I think something like that is true). I am not supposing, initially at least, that having a design plan implies having been created by God or some other conscious agent; it is perhaps possible that evolution (undirected by God or anyone else) has somehow furnished us with our design plans. [187] I mean, instead, to point to something nearly all of us, theists or not, believe: there is a way in which a human organ or system works when it works properly, works as it is supposed to work; and this way of working is given by its design or design plan. Proper function and design go hand in hand with the notion of purpose or function. The various organs and systems of the body (and the ways in which they work) have their functions, their purposes: the function or purpose of the heart is to pump the blood; of the immune system, to fight off disease; of the lungs to provide oxygen; of peristalsis, to move nutrients along the intestinal tract, and so on. If the design is a good design, then when the organ or system functions properly (i.e., according to its design plan), that purpose will be achieved. The design plan specifies a particular way of working that subserves that purpose. Of course, the design plan for human beings will include specifications for our cognitive system or faculties, as well as for noncognitive systems and organs. Like the rest of our organs and systems, our cognitive faculties can work well or ill; they can malfunction or function properly. They too work in a certain way when they are functioning properly--and work in a certain way to accomplish their purpose. Accordingly, the first element in our conception of warrant (so I say) is that a belief has warrant for someone only if her faculties are functioning properly, are subject to no dysfunction, in producing that belief. [188] But that's not enough. Many systems of your body, obviously, are designed to work in a certain kind of environment. You can't breathe under water; your muscles atrophy in zero gravity; you can't get enough oxygen at the top of Mount Everest. Clearly, the same goes for your cognitive faculties; they too will achieve their purpose only if functioning in an environment much like the one for which they were designed (by God or evolution). Thus they won't work well in an environment (on some other planet, for example) in which a certain subtle radiation impedes the function of memory. And this is still not enough. It is clearly possible that a belief be produced by cognitive faculties that are functioning properly in an environment for which they were designed, but nonetheless lack warrant; the above two conditions are not sufficient. We think that the purpose or function of our belief-producing faculties is to furnish us with true (or verisimilitudinous) belief. As we saw above in connection with the F&M complaint, however, it is clearly possible that the purpose or function of some belief-producing faculties or mechanisms is the production of beliefs with some other virtue--perhaps that of enabling us to get along in this cold, cruel, threatening world, or of enabling us to survive a dangerous situation or a life-threatening disease. So we must add that the belief in question is produced by cognitive faculties such that the purpose of those faculties is that of producing true belief. More exactly, we must add that the portion of the design plan governing the production of the belief in question is aimed at the production of true belief (rather than survival, or psychological comfort, or the possibility of loyalty, or something else). Even this isn't sufficient. We can see why by reflecting on a fantasy of David Hume's: This world, for aught he knows, is very faulty and imperfect, compared to a superior standard; and was only the first rude essay of some infant Deity, who afterwards abandoned it, ashamed of his lame performance; it is the work only of some dependent, inferior Deity; and is the object of derision to his superiors; it is the production of old age and dotage in some superannuated Deity; and ever since his death, has run on at adventures, from the first impulse and active force, which it received from him. [189] So imagine that a young and untutored apprentice deity sets out to build cognitive beings, beings capable of belief and knowledge. Immaturity and incompetence triumph; the design contains serious glitches. In fact, in some areas of the design, when the faculties work just as they were designed to, the result is ludicrously false belief: thus when the cognitive faculties of these beings are working according to their design plan, they constantly confuse horses and hearses, forming the odd beliefs that cowboys in the old West rode hearses and that corpses are usually transported in horses. These beliefs are then produced by cognitive faculties working properly in the right sort of environment according to a design plan aimed at truth, but they still lack warrant. What is missing? Clearly enough, what must be added is that the design plan in question is a good one, one that is successfully aimed at truth, one such that there is a high (objective) probability that a belief produced according to that plan will be true (or nearly true). Put in a nutshell, then, a belief has warrant for a person S only if that belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S's kind of cognitive faculties, according to a design plan that is successfully aimed at truth. We must add, furthermore, that when a belief meets these conditions and does enjoy warrant, the degree of warrant it enjoys depends on the strength of the belief, the firmness with which S holds it. This is intended as an account of the central core of our concept of warrant; there is a penumbral area surrounding the central core where there are many analogical extensions of that central core; and beyond the penumbral area, still another belt of vagueness and imprecision, a host of possible cases and circumstances where there is really no answer to the question whether a given case is or isn't a case of warrant. [190] This means that the sort of classical analysis in which necessary and sufficient conditions are set out in a stylishly austere clause or two is of limited value here. What we need, instead, is an explanation and description of how the account works in the main areas of our cognitive life; that was the task of WPF. Responses to the above account of warrant have made it abundantly clear that it needs a certain kind of supplementation and fine tuning. [191] To see this, consider the following kind of Gettier example. I own a Chevrolet van, drive to Notre Dame on a football Saturday, and unthinkingly park in one of the many spaces reserved for the football coach. Naturally, his minions tow my van away and, as befits such lčse-majesté, destroy it. By a splendid piece of good luck, however, I have won the Varsity Club's Win-a-Chevrolet-Van contest, although I haven't yet heard the good news. You ask me what sort of automobile I own; I reply, both honestly and truthfully, ?A Chevrolet van.? My belief that I own such a van is true, but just by accident' (more accurately, it is only by accident that I happen to form a true belief); hence it does not constitute knowledge. All of the nonenvironmental conditions for warrant, furthermore, are met. It also looks as if the environmental condition is met: after all, isn't the cognitive environment here on earth and in South Bend just the one for which our faculties were designed? What is important about the example is this: it is clear that if the coach's minions had been a bit less zealous and had not destroyed my van, the conditions for warrant outlined above would have obtained and I would have known that I owned a Chevrolet van. In the actual situation, however, the one in which the van is destroyed, my belief is produced by the very same processes functioning the very same way in (apparently) the same cognitive environment. Hence, on my account, either both of these situations are ones in which I know that I own a Chevrolet van, or neither is. But clearly one is, and the other isn't. Therefore my account is apparently defective. [192] Consider another Gettier example, this one antedating Gettier's birth (it was proposed by Bertrand Russell). I glance at a clock, forming the opinion that it is 3:43 p.m.; as luck would have it, the clock stopped precisely twenty-four hours ago. The belief I form is indeed true; again, however, it is true just by accident' (the clock could just as well have stopped an hour earlier or later); it does not constitute knowledge. As in the previous case, if the clock had been running properly and I had formed the same belief by the same exercise of cognitive powers, I would have known; here, therefore, we have another example that apparently refutes my account. Still another example: I am not aware that Paul's look-alike brother Peter is staying at his house; if I'm across the street, take a quick look, and form the belief that Paul is emerging from his house, I don't know that it's Paul, even if in fact it is (it could just as well have been Peter emerging); again, if Peter hadn't been in the neighborhood, I would have known. What is crucial, in each of these cases, is that my cognitive faculties display a certain lack of resolution. I am unable, by a quick glance, to distinguish the state of affairs in which the clock is running properly and telling the right time from a state of affairs in which it stopped just twelve or twenty-four hours earlier. I cannot distinguish Paul from Peter just by a quick look from across the street. Of course, this lack of resolution is in each case relative to the particular exercise of cognitive powers in question. If I had watched the clock for ten minutes, say, I would have known that it isn't running, and if I had walked across the street and taken a good look, I'd have known that it wasn't Paul but Peter at the door. What I can't distinguish by those exercises of my epistemic powers are different cognitive minienvironments. In ?Respondeo,? there is a fuller development of the distinction between cognitive maxienvironments and cognitive minienvironments; here the following will suffice. First, a cognitive maxienvironment is more general and more global than a cognitive minienvironment. Our cognitive maxienvironment here on earth would include such macroscopic features as the presence and properties of light and air, the presence of visible objects, of other objects detectable by cognitive systems of our kind, of some objects not so detectable, of the regularities of nature, of the existence and general nature of other people, and so on. Our cognitive faculties are designed (by God or evolution) to function in this maxienvironment, or one like it. They are not designed for a maxienvironment in which, for example, there is constant darkness, or where everything is in a state of constant random flux, or where the only food available contains a substance that destroys short-term memory, or where there aren't any distinguishable objects, or no regularities of a kind we can detect; in such an environment, our faculties will not fulfill their function of providing us with true beliefs. Now a given cognitive maxienvironment can contain many different minienvironments--for example, the one where the clock stops, but also one where it doesn't; the one where Peter is visiting Paul, but also one where he isn't; the one where the coach's minions destroy my van, but also one where they magnanimously temper the punishment I so richly deserve, contenting themselves with painting the windshield black. And now here's the point: some cognitive minienvironments--such as those of the Notre Dame van case, the clock that stopped, Peter's visit to Paul--are misleading for some exercises of cognitive faculties, even when those faculties are functioning properly and even when the maxienvironment is favorable. The maxienvironment is right, but the minienvironment isn't; in those minienvironments the cognitive faculties in question (more exactly, particular exercises of the cognitive faculties in question) can't be counted on to produce true beliefs. The basic idea is this: our cognitive faculties have been designed for a certain kind of maxienvironment. Even within that maxienvironment, however, they don't function perfectly (they sometimes produce false belief), although they do function reliably. (Perhaps perfectly functioning cognitive faculties would require too much brain size, thus interfering with the achievement of other desiderata.) In some minienvironments, therefore, they can't be counted on to produce a true belief: if they do, it is just by accident and does not constitute knowledge. So even if the maxienvironment is favorable and the other conditions of warrant are met, a belief could still be true just by accident', thus not constituting knowledge. It is clear, therefore, that S knows p, on a given occasion, only if S's cognitive minienvironment, on that occasion, is not misleading--more exactly, not misleading with respect to the particular exercise of cognitive powers producing the belief that p. So the conditions of warrant (i.e., for the degree of warrant sufficient for knowledge [193] ) need an addition: the maxienvironment must, indeed, be favorable or appropriate, but so must the cognitive minienvironment. What must then be added to the other conditions of warrant is the resolution condition: (RC) A belief B produced by an exercise E of cognitive powers has warrant sufficient for knowledge only if MBE (the minienvironment with respect to B and E) is favorable for E. What does appropriateness' or favorability', or nonmisleadingness', for a cognitive minienvironment, consist in: can we say anything more definite? Intuitively, a minienvironment is favorable, for an exercise of cognitive powers, if that exercise can be counted on to produce a true belief in that minienvironment. Perhaps this is as specific as we can sensibly get; in ?Respondeo,? however, I went on to make a tentative suggestion as to how we could say a bit more precisely what this favorability consists in. Where B is a belief, E the exercise of cognitive powers that produces B, and MBE a minienvironment with respect to B and E, say that (F) MBE is favorable for E if and only if, if S were to form a belief by way of E in MBE, S would form a true belief. [194] Sadly enough, though, (F) won't do the trick at all; the relevant counterfactual itself can be true just by accident'--that is, by accident from the point of view of the design plan. [195] There are plenty of possible cases to demonstrate this: here is one. Return to those impecunious Wisconsinites trying to put the best face on things by erecting a lot of fake barns. [196] Suppose I am driving through the area on an early September morning when there is a good deal of mist and fog. I glance to the right and see a real barn; as it happens, all the nearby fake barns (which outnumber the real ones) are obscured by the morning mist; I say to myself, ?Now that is a fine barn!? The belief I form is true; the relevant counterfactual is also true because of the way the fake barns are obscured by mist; but the belief does not have warrant sufficient for knowledge. What to do? Here is another (also tentative) suggestion. Recall that the resolution problem arises because I can't (for example) distinguish Paul from Peter from across the street just by looking; this particular exercise of cognitive powers displays insufficient resolution for that. So consider a given exercise of cognitive powers E, the belief B formed on that occasion, and a relevant cognitive minienvironment MBE. If the conditions of warrant have been met, B will be probable (ordinarily very probable) with respect to MBE. Of course, MBE is a state of affairs. Among the states of affairs it includes are some that E is competent to detect, that are cognitively accessible to E. Thus in the twin case the appearance of a person, of a man, of someone across the road, and the like, are all detectable by E--that is, just by taking a look. On the other hand, it's being Paul rather than Peter who appears in the doorway is not thus detectable; they look just alike at this distance, and I know nothing entailing that Peter isn't there. So consider the conjunction of circumstances C contained in MBE such that C is detectable by E; call this conjunctive state of affairs DMBE. In the case in question, these circumstances will be observable, and observable by way of taking a look from across the road. In the typical case, furthermore (assuming that the general conditions of warrant are met), B will also be probable with respect to DMBE. And now we can say what it is for a minienvironment to be favorable: MBE is favorable just if there is no state of affairs S included in MBE but not in DMBE such that the objective probability of B with respect to the conjunction of DMBE and S falls below r, where r is some real number representing a reasonably high probability. In the twin case, for example, a state of affairs S such that B is not probable enough with respect to the conjunction of DMBE and S would be Peter's being in the house as well as Paul, and being indistinguishable from him from across the street. In the case of the impecunious Wisconsinites, it is that there are more fake barns than real barns in the neighborhood. Also, of course, I don't specify the requisite level of probability r, which in any case will display a certain contextual character, differing from case to case. This suggestion seems promising, although induction leads me to be less than wholly confident that it is right. It may be that in the long run we can't say more than that the minienvironment must be favorable. The overall picture, then, is as follows. Our faculties are designed for a certain kind of cognitive maxienvironment, one that sufficiently resembles the one in which we do, in fact, find ourselves. And when a belief is formed by properly functioning faculties in an environment of that sort (and the bit of the design plan that governs its production is successfully aimed at truth), then the belief in question has some degree of warrant, even if it happens to be false. But our cognitive faculties are not maximally effective--not only in that there is much we aren't capable of coming to know but also in that we are sometimes prone to err, even when the maxienvironment is right and the relevant faculties are functioning properly. Another way to put the same point: within a favorable cognitive maxienvironment, there can be minienvironments for a given exercise of our faculties, in which it is just by accident, dumb luck, that a true belief is formed, if one is indeed formed. A true belief formed in such a minienvironment doesn't have warrant sufficient for knowledge, even if it has some degree of warrant. To achieve that more exalted degree of warrant, the belief must be formed in a minienvironment such that the exercise of the cognitive powers producing it can be counted on to produce a true belief. Hence the resolution condition. Beliefs that meet all of the conditions will then constitute knowledge (provided they are accepted with sufficient firmness). I have neglected several important components of our epistemic establishment. First, I have said nothing here about defeaters; in chapter 11, I'll address this topic. Another very important topic ignored here [197] is that of epistemic probability. Further, knowledge or warrant seems to have a contextual character; the degree of warrant necessary for knowledge seems to depend, to some extent, on circumstances and context. I don't have the space to go into these matters here. __________________________________________________________________ [185] The rest is to be found in my reply to Alston, Ginet, Steup, Swinburne, and Taylor in ?Reliabilism, Analyses and Defeaters,? Philosophy and Phenomenological Research 55/2 (1995), pp. 427ff.; ?Respondeo,? in Warrant in Contemporary Epistemology: Essays in Honor of Plantinga's Theory of Knowledge, ed. J. Kvanvig (Lanham, Md.: Rowman and Littlefield, 1996); ?Warrant and Accidentally True Belief,? Analysis 57, n. 2 (April 1, 1997), p. 140; and pp. 156ff., below. [186] See WPF, pp. 11ff. [187] Although in WPF, chapter 11, I argue that there is no viable naturalistic account of proper function. [188] For necessary qualifications, see WPF, pp. 9ff. and 22-42. [189] Dialogues Concerning Natural Religion, ed. Nelson Pike (Indianapolis and New York: Bobbs-Merrill, 1970), p. 53. [190] As I argue in WPF, pp. 212-13. [191] By Robert Shope in ?Gettier Problems,? in Routledge Encyclopedia of Philosophy, ed. Edward Craig (London: Routledge, 1998), and his forthcoming book Knowledge as Power; Richard Feldman in ?Plantinga, Gettier, and Warrant,? in Warrant in Contemporary Epistemology: Essays in Honor of Plantinga's Theory of Knowledge, ed. Jonathan Kvanvig (New York: Rowman and Littlefield, 1996), p. 216; and Peter Klein, ?Warrant, Proper Function, Reliabilism, and Defeasibility,? in Warrant in Contemporary Epistemology, p. 105. I am grateful to all three for instruction and enlightenment. For my reply and an effort at repair, see ?Respondeo,? in Warrant in Contemporary Epistemology. [192] For fuller development here, see ?Respondeo,? in Warrant in Contemporary Epistemology, pp. 314ff. [193] The thought is not that a belief produced in an unfavorable minienvironment has no warrant at all, but only that it doesn't have a degree of warrant sufficient for knowledge. See Trenton Merricks's ?Warrant Entails Truth,? Philosophy and Phenomenological Research, 55, no. 4 (December 1995), p. 841; see also Sharon Ryan's reply, ?Does Warrant Entail Truth?? Philosophy and Phenomenological Research 56, no. 1 (March 1996), p. 183, and Merricks's rejoinder, ?More on Warrant's Entailing Truth,? Philosophy and Phenomenological Research 57, no. 3 (September 1997), p. 627. [194] Here I am assuming (contrary to the usual semantics for counterfactuals) that truth of antecedent and consequent is not sufficient for truth of the counterfactual (a counterfactual can be false even if it has a true antecedent and a true consequent). What is also required is that there be no sufficiently close possible world in which the counterfactual has a true antecedent and false consequent. [195] As was pointed out to me by Thomas Crisp. [196] See WPF, pp. 32-33. [197] But treated in WPF, chapter 9. __________________________________________________________________ III. The F&M Complaint Again Now we are ready to return to the F&M complaint. What we see is a clear if surprising connection between the topic of warrrant and the F&M complaint: the latter is really the claim that theistic belief lacks warrant. According to Freud, theistic belief is produced by cognitive faculties that are functioning properly, but the process that produces it--wishful thinking--does not have the production of true belief as its purpose; it is aimed, instead, at something like enabling us to carry on in the grim and threatening world in which we find ourselves. Therefore, theistic belief does not meet the third condition of warrant; as a result, the presumption of reliability that goes with warranted beliefs does not apply to it. Theistic belief is no more respectable, epistemically speaking, than propositions selected entirely at random. Suppose I have a random generator of English declarative sentences (sentences that express propositions); it randomly chooses one of a stockpile of a million sentences and their negations, flashing its selection on a big screen. I use the machine, recommending the resulting proposition to you for belief. You quite properly demur, pointing out that there isn't the slightest reason to think the belief in question true. Theistic belief, thinks Freud, has no better epistemic credentials, for the believer, than the propositions expressed by those sentences would have for someone for whom they have no source of warrant in addition to their appearing on the screen. It is baseless superstition. Still further, Freud thinks, once we see that theistic and religious belief has its origin in wishful thinking, we will also see that it is very probably false. There is no good argument from this fact about its origin to the conclusion that it is false; nor is it that someone who recognizes its origin in wishful thinking will simply see that it is false. It is rather just that people of sense who know something about how the world works will take it to be probably false. They will take the same attitude toward theistic and Christian belief that they take toward the stories in Greek or Aztec or Persian mythology: we can't really prove that these stories are false, but their chances of being true are pretty slim. So the proper intellectual attitude toward these beliefs isn't merely agnosticism; it is that the beliefs in question are unwarranted and furthermore are very probably false. Marx's views are similar. He thinks first that theistic and religious belief is produced by cognitive faculties that are not functioning properly. Those faculties are, to the extent that they produce such belief, dysfunctional; the dysfunction is due to a sort of perversion in social structure, a sort of social malfunction. Religious belief therefore doesn't meet the first condition of warrant; it is therefore without warrant, and an intellectually healthy person will reject it. Further, Marx also thinks that a person whose cognitive faculties are functioning properly and who knows what was known by the middle of the nineteenth century will see that materialism is very probably true, in which case Christian and theistic belief is very likely false. So he would join Freud in the contention that Christian and theistic belief is without warrant, a baseless superstition, and very probably false. We could see the matter slightly differently. Perhaps the problem with religious belief, according to Marx, is not that it is produced by malfunctioning faculties, but rather that capitalist society constitutes a hostile environment for the operation of human cognitive faculties; then the problem would be the second condition rather than the first. Still another possibility: perhaps the production of theistic or religious belief is like a damage-control mechanism. When people are subjected to the nasty conditions of capitalism, they come to believe these tales of a God and another world as a means of coping with their otherwise intolerable situation. Then Marx's view would be more like Freud's, and religious belief could be seen as an illusion in the Freudian sense. There would remain the following difference. According to Freud, the inclination to form religious belief arises out of our nature and is therefore to be expected, no matter what the social structure. According to this version of Marx, however, religious belief is a response to the very special social circumstances of misery and injustice generated by capitalist society, so that there need be no inclination toward it among people in a society that doesn't display that or a similar perversion. Of course Marx actually says little about religion, not enough to make it possible to distinguish one of these possibilities as the one he had in mind. The F&M complaint, therefore, is that theistic belief and religious belief in general lack warrant. So say Freud and Marx--but are they right? In the next chapter, we shall turn to a model for the possession of warrant by Christian belief. Model in hand, we shall then evaluate the F&M complaint. __________________________________________________________________ [150] In Faith and Rationality, ed. Alvin Plantinga and Nicholas Wolterstorff (Notre Dame: University of Notre Dame Press, 1983). [151] Ithaca: Cornell University Press, 1967. __________________________________________________________________ __________________________________________________________________ PART III WARRANTED CHRISTIAN BELIEF __________________________________________________________________ 6 Warranted Belief in God To know in a general and confused way that God exists is implanted in us by nature. . . . Thomas Aquinas for since the creation of the world God's invisible qualities--his eternal power and divine nature--have been clearly seen, being understood from what has been made. . . . St. Paul The de jure challenge to Christian (or theistic) belief, as we have seen, is the claim that such belief is irrational or unreasonable or unjustified or in some other way properly subject to invidious epistemic criticism; it contrasts with the de facto challenge, according to which the belief in question is false. Put just like that, the de jure rebuke is pretty vague and general; we can't do much by way of evaluating the proposed complaint without achieving a clearer and more specific formulation of it. As we have seen, clear and sensible formulation of the de jure criticism--at any rate of one that isn't just obviously mistaken--has proven elusive. In the last chapter, however, we were able to make progress by considering the F&M (Freud and Marx) complaint. What we saw is that this complaint is really the claim that Christian and other theistic belief is irrational in the sense that it originates in cognitive malfunction (Marx) or in cognitive proper function that is aimed at something other than the truth (Freud)--comfort, perhaps, or the ability to soldier on in this appalling world in which we find ourselves. To put it another way, the claim is that such belief doesn't originate in the proper function of cognitive faculties successfully aimed at producing true beliefs. To put it in still another way, the charge is that theistic and Christian belief lacks warrant. By way of response, in this chapter I shall first offer a model--a model based on a claim made jointly by Thomas Aquinas and John Calvin--for a way in which theistic belief could have warrant. Once we see how theistic belief might have warrant, we can also see the futility of the F&M complaint and its contemporary successors. In the remaining chapters of part III, I shall extend the model to cover specifically Christian belief. Chapter 7 will deal with sin and its noetic results. The extended model crucially involves the notion of faith. Following Aquinas and Calvin, I shall argue that faith has both an intellectual and an affective component: chapter 8 will therefore examine the way in which, as Calvin says, the great truths of the gospel are ?revealed to our minds,? and chapter 9 will examine the way in which, as he also says, they are ?sealed upon our hearts.? Then in chapter 10, I'll consider and reply to objections to the original and extended models. __________________________________________________________________ I. The Aquinas/Calvin Model __________________________________________________________________ A. Models I say I propose in this chapter to give a model of theistic belief's having warrant; but what sort of animal is a model, and what would it be good for? There are models of many different kinds: model airplanes, artists' models, models in the sense of exemplars, models of a modern major general. There is also the logician's sense of model in which, for example, any consistent first-order theory has a model in the natural numbers. My use of the term here is more abstract than the first and more concrete than the second. The rough idea is this: to give a model of a proposition or state of affairs S is to show how it could be that S is true or actual. The model itself will be another proposition (or state of affairs), one such that it is clear (1) that it is possible and (2) that if it is true, then so is the target proposition. From these two, of course, it follows that the target proposition is possible. In this chapter, I shall give a model of theistic belief's having warrant: the Aquinas/Calvin (A/C) model. Then in chapters 7, 8, and 9, I will extend the A/C model to a model in which specific and full-blooded Christian belief has warrant. I claim four things for these two models. First, they are possible, and thus show it is possible that theistic and Christian belief have warrant. The sense of possibility here, however, isn't just broadly logical possibility--after all, such obvious falsehoods as the population of China is less than a thousand are possible in that sense--but something much stronger. I claim that these models are epistemically possible: they are consistent with what we know, where ?what we know? is what all (or most) of the participants in the discussion agree on. [198] Second, and related to the first assertion, I claim that there aren't any cogent objections to the model--that is, to the proposition that the model is in fact true or actual. More exactly, there are no cogent objections of a philosophic or scientific kind (or indeed any other kind) to the model that are not also cogent objections to theism or Christian belief. Another way to put it: any cogent objection to the model's truth will also have to be a cogent objection to the truth of theistic or Christian belief. I shall go on to argue that if Christian belief is indeed true, then the model in question or one very like it is also true. If I am successful, therefore, the upshot will be that there is no viable de jure (as opposed to de facto) challenge either to theistic or to Christian belief. There is no sensible challenge to the rationality or rational justification or warrant of Christian belief that is not also a challenge to its truth. That is, there is no de jure challenge that is independent of a de facto challenge. That means that a particularly popular way of criticizing Christian belief--to be found in the evidentialist objection, in the F&M complaint, in many versions of the argument from evil, and in still other objections--is not viable. This is the sort of challenge that goes as follows: ?I don't know whether Christian (or theistic) belief is true--how could anyone know a thing like that? But I do know that it is irrational, or rationally unacceptable or unjustified or without warrant (or in some other way epistemically challenged).? If my argument is right, no objection of this sort has any force. Third, I believe that the models I shall present are not only possible and beyond philosophical challenge but also true, or at least verisimilitudinous, close to the truth. Still, I don't claim to show that they are true. That is because the A/C model entails the truth of theism and the extended A/C model the truth of classical Christianity. To show that these models are true, therefore, would also be to show that theism and Christianity are true; and I don't know how to do something one could sensibly call showing' that either of these is true. I believe there are a large number (at least a couple dozen) good arguments for the existence of God; none, however, can really be thought of as a showing or demonstration. As for classical Christianity, there is even less prospect of demonstrating its truth. [199] Of course this is nothing against either their truth or their warrant; very little of what we believe can be demonstrated' or shown'. Fourth, there is a whole range of models for the warrant of Christian belief, all different but similar to the A/C and extended A/C models. (In claiming that models I present are close to the truth, what I am claiming is that they belong to that range.) A