Is It Wrong Everywhere, Always, and for Anyone to Believe Anything on Insufficient Evidence?

by Peter Van Inwagen

What I want to do is not so much to challenge (or to vindicate) the principle this sentence expresses as to examine what the consequences of attempting consistently to apply it in our lives would be. Various philosophers have attempted something that might be described in these words, and have argued that a strict adherence to the terms of the principle would lead to a chain of requests for further evidence that would terminate only in such presumably unanswerable questions as What evidence have you for supposing that your sensory apparatus is reliable?, or Yes, but what considerations can you adduce in support of the hypothesis that the future will resemble the past?; and they have drawn the conclusion that anyone who accepts such propositions as that one's sensory apparatus is reliable or that the future will resemble the past must do so in defiance of the principle. You will be relieved to learn that an investigation along these lines is not on the program tonight. I am not going to raise the question whether a strict adherence to the principle would land us in the one of those very abstract sorts of epistemological predicament exemplified by uncertainty about the reliability of sense perception or induction. I shall be looking at consequences of accepting the principle that are much more concrete, much closer to our concerns as epistemically responsible citizens—citizens not only of the body politic but of the community of philosophers.

I shall, as I say, be concerned with Clifford's sentence and the lecture that it epitomizes. But I am going to make my way to this topic by a rather winding path. Please bear with me for a bit.

I begin my indirect approach to Clifford's 'sentence by stating a fact about philosophy. Philosophers do not agree about anything to speak of. That is, it is not very usual for agreement among philosophers on any important philosophical issue, to be describable as being, in a quite unambiguous sense, common. Oh, this philosopher may agree with that philosopher on many philosophical points; for that matter, if this philosopher is a former student of that philosopher, they may even agree on all philosophic al points. But you don't find universal or near-universal agreement about very many important theses or arguments in philosophy. Indeed, it would be hard to find an important philosophical thesis that, say, ninety-five percent of, say, American analytical philosophers born between 1930 and 1950 agreed about in, say, 1987.

And why not? How can it be that equally intelligent and well-trained philosophers can disagree about the freedom of the will or nominalism or the covering-law model of scientific explanation when each is aware of all of the arguments and distinctions an d other relevant considerations that the others are aware of? How—and now I will drop a broad hint about where I am going—how can we philosophers possibly regard ourselves as justified in believing much of anything of philosophical significance in this embarrassing circumstance? How can I believe (as I do) that free will is incompatible with determinism or that unrealized possibilities are not physical objects or that human beings are not four-dimensional things extended in time as well as in space, when David Lewis—a philosopher of truly formidable intelligence and insight and ability—rejects these things I believe and is already aware of and understands perfectly every argument that I could produce in their defense?

Well, I do believe these things. And I believe that I am justified in believing them. And I am confident that I am right. But how can I take these positions? I don't know. That is itself a philosophical question, and I have no firm opinion about its correct answer. I suppose my best guess is that I enjoy some sort of philosophical insight (I mean in relation to these three particular theses) that, for all his merits, is somehow denied to Lewis. And this would have to be an insight that is incommunicable- -at least I don't know how to communicate it—, for I have done all I can to communicate it to Lewis, and he has understood perfectly everything I have said, and he has not come to share my conclusions. But maybe my best guess is wrong. I'm confident about only one thing in this area: the question must have some good answer. For not only do my beliefs about these questions seem to me to be undeniably true, but (quite independently of any consideration of which theses it is that seem to me to be true), I don't want to be forced into a position in which I can't see my way clear to accepting any philosophical thesis of any consequence. Let us call this unattractive position philosophical skepticism. (Note that I am not using this phrase in its usual sense of "comprehensive and general skepticism based on philosophical argument." Note also that philosophical skepticism is not a thesis—if it were, it's hard to see how it could be accepted without pragmatic contradiction—but a state: philosophical skeptics are people who can't see their way clear to being nominalists or realists, dualists or monists, ordinary-language philosophers or phenomenologists; people, in short, who are aware of many philosophical options but take none of them, people who have listened to many philosophical debates but have never once declared a winner.) I think that any philosopher who does not wish to be a philosophical skeptic—I know of no philosopher who is a philosophical skeptic—must agree with me that this question has some good answer: whatever the reason, it must be possible for one to be justified in accepting a philosophical thesis when there are philosophers who, by all objective and external criteria, are at least equally well qualified to pronounce on that thesis and who reject it.

Will someone say that philosophical theses are theses of a very special sort, and that philosophy is therefore a special case? That adequacy of evidential support is much more easily achieved in respect of philosophical propositions than in respect of geological or medical or historical propositions? Perhaps because nothing really hangs on philosophical questions, and a false or unjustified philosophical opinion is therefore harmless? Or because philosophy is in some sense not about matters of empirical fact? As to the first of these two suggestions, I think it is false that nothing hangs on philosophical questions. What people have believed about the philosophical theses advanced by—for example—Plato, Locke, and Marx has had profound effects on history. I don't know what the world would be like if everyone who ever encountered philosophy immediately became, and thereafter remained, a philosophical skeptic, but I'm willing to bet it would be a vastly different world. (In any case, I certainly hope this suggestion is false. I'd hate to have to defend my own field of study against a charge of adhering to loose epistemic standards by arguing that it's all right to adopt loose epistemic standards in philosophy because philosophy is detached from life to su ch a degree that philosophical mistakes can't do any harm.) In a more general, theoretical way, Clifford has argued, and with some plausibility, that it is in principle impossible to claim on behalf of any subject-matter whatever—on the ground that mistaken beliefs about the things of which that subject-matter treats are harmless—exemption from the strict epistemic standards to which, say, geological, medical, and historical beliefs are properly held. He argues,

[That is not] truly a belief at all which has not some influence upon the actions of him who holds it. He who truly believes that which prompts him to an action has looked upon the action to lust after it, he has committed it already in his heart. If a belief is not realized immediately in open deeds, it is stored up for the guidance of the future. It goes to make a part of that aggregate of beliefs which is the link between sensation and action at every moment of all our lives, and which is so organized and compacted together that no part of it can be isolated from the rest, but every new addition modifies the structure of the whole. No real belief, however trifling and fragmentary it may seem, is ever truly insignificant; it prepares us to receive m ore of its like, confirms those which resembled it before, and weakens others; and so gradually it lays a stealthy train in our inmost thoughts, which may some day explode into overt action, and leave its stamp upon our character forever. . . And no one man's belief is in any case a private matter which concerns himself alone . . . no belief held by one man, however seemingly trivial the belief, and however obscure the believer, is actually insignificant or without its effect on the fate of mankind . . . .

Whether or not you find this general, theoretical argument convincing, it does in any case seem quite impossible to maintain, given the actual history of the relation between philosophy and our social life, that it makes no real difference what people believe about philosophical questions.

The second suggestion—that philosophy is different (and that philosophers may therefore properly, in their professional work, observe looser epistemic standards than geologists or physicians observe in theirs) because it's not about matters of empirical fact—is trickier. Its premise is not that it doesn't make any difference what people believe about philosophical questions; it's rather that the world would look exactly the same whether any given philosophical thesis was true or false. I think t hat that's a dubious assertion. If the declarative sentences that philosophers characteristically write and speak in their professional capacity are meaningful at all, then many of them express propositions that are necessary truths or necessary falsehood s, and it's at least a very doubtful assertion that the world would look the same if some necessary truth were a falsehood or if some necessary falsehood were a truth. (Would anyone argue that mathematicians may properly hold themselves to looser epistemic standards than geologists because the world would look the same whether or not there was a greatest prime?) And even if it were true that philosophy was, in no sense of this versatile word, about matters of empirical fact, one might well raise the question why this should lend any support to the suggestion that philosophers were entitled to looser epistemic standards than geologists or physiologists, given that philosophical beliefs actually do have important effects on the behavior of those who hold them. Rather than address the issues that these speculations raise, however, I will simply change the subject.

Let us consider politics. Almost everyone will admit that it makes a difference what people believe about politics—I am using the word in its broadest possible sense- and it would be absurd to say that propositions like Capital punishment is an ineffective deterrent or "Nations that do not maintain a strong military capability actually increase the risk of war" are not about matters of empirical fact. And yet people disagree about these propositions (and scores of others of equal importance), and their disagreements about them bear a disquieting resemblance to the disagreements of philosophers about nominalism and free will and the covering-law model. That is, their disagreements are matters of interminable debate, and impressive authorities can be found on both sides of many of the interminable debates.

It is important to realize that this feature of philosophy and politics is not a universal feature of human discourse. It is clear, for example, that someone who believes in astrology believes in something that is simply indefensible. It would be hard to find a philosopher—I hope this is true- who believed that every philosopher who disagreed with his or her position on nominalism held a position that was indefensible in the same way that a belief in astrology was indefensible. It might be easier to fin d someone who held the corresponding position about disputed and important political questions. I suspect there really are people who think that those who disagree with them about the deterrent effect of capital punishment or the probable consequences of unilateral disarmament are not only mistaken but hold beliefs that are indefensible in the way that a belief in astrology is indefensible. I can only say that I regard this attitude as ludicrous. On each side of many interminably debated political questions—it is not necessary to my argument to say all—one can find well-informed (indeed, immensely learned) and highly intelligent men and women who adhere to the very highest intellectual standards. And this is simply not the case with debates about astrology. In fact, it is hardly possible to suppose that there could be a very interesting debate about the truth-values of the claims made by astrologers.

Everyone who is intellectually honest will admit this, will admit that there are interminable political debates with highly intelligent and well informed people on both sides. And yet few will react to this state of affairs by becoming political skeptics , by declining to have any political beliefs that are disputed by highly intelligent and well-informed people. But how can this rejection of political skepticism be defended? How can responsible political thinkers believe that the Syndicalist Party is the last, best hope for Ruritania when they know full well that there are well informed (even immensely learned) and highly intelligent people who argue vehemently—all the while adhering to the highest intellectual standards—that a Syndicalist government would be the ruin of Ruritania? Do the friends of Syndicalism claim to see gaps in the arguments of their opponents, "facts" that they have cited that are not really facts, real facts that they have chosen not to mention, a hidden agenda behind their opposition to Syndicalism? No doubt they do. Nevertheless, if they are intelligent and intellectually honest, they will be aware that if these claims were made in public debate, the opponents of Syndicalism would probably be able to muster a very respectable rebuttal. The friends of Syndicalism will perhaps be confident that they could effectively meet the points raised in this rebuttal, but, if they are intelligent and intellectually honest, they will be aware . . . and so, for all practical purposes, ad infinitum.

I ask again: what could it be that justifies us in rejecting political skepticism? How can I believe that my political beliefs are justified when these beliefs are rejected by people whose qualifications for engaging in political discourse are as impressive as David Lewis's qualifications for engaging in philosophical discourse? These people are aware of (at least) all the evidence and all the arguments that I am aware of, and they are (at least) as good at evaluating evidence and arguments as I. How, then, can I maintain that the evidence and arguments I can adduce in support of my beliefs actually justify these beliefs? If this evidence and these arguments are capable of that, then why aren't they capable of convincing these other people that these beliefs are correct? Well, as with philosophy, I am inclined to think that I must enjoy some sort of incommunicable insight that the others, for all their merits, lack. I am inclined to think that the evidence and arguments I can adduce in support of my beliefs do not constitute the totality of my justification for these beliefs. But all that I am willing to say for sure is that something justifies me in rejecting political skepticism, or at least that it is possible that something does: that it is no t a necessary truth that one is not justified in holding a political belief that is controverted by intelligent and well-informed political thinkers.

I have now accomplished one of the things I wanted to do in this talk. I have raised the question how it is possible to avoid philosophical and political skepticism. In the remainder of this talk, I am going to turn to questions about religious belief. M y point in raising the questions I have raised about philosophy and politics was primarily to set the stage for comparing religious beliefs with philosophical and political beliefs. But I think that the questions I have so far raised are interesting in their own right. Even if everything I say in the remainder of the talk is wrong, even if my comparisons of philosophical and political beliefs with religious beliefs turn out to be entirely wide of the mark, the interest of the questions I have raised so fa r would remain. How can we philosophers, when we consider the matter carefully, avoid the uncomfortable suspicion that the following words of Clifford might apply to us: Every one of them, if he chose to examine himself in foro conscientiae, would know that he had acquired and nourished a belief, when he had no right to believe on such evidence as was before him; and therein he would know that he had done a wrong thing. ?

Now as to religion: Is religion different from philosophy and politics in the respects we have been discussing? Should religious beliefs perhaps be held to a stricter evidential standard than philosophical and political beliefs? Or, if they are to be held to the same standard, do typical religious beliefs fare worse under this standard than typical philosophical or political beliefs? It is an extremely popular position that religion is different. Or, at least, it must be that many anti-religious philosophers and other writers hostile to religious belief hold this position, for it seems to be presupposed by almost every aspect of their approach to the subject of religious belief. And yet this position seems never to have been explicitly formulated, much less argued for. Let us call it the Difference Thesis. An explicit formulation of the Difference Thesis is a tricky matter. I tentatively suggest that it be formulated disjunctively: Either religious beliefs should be held to a stricter epistemic standard than beliefs of certain other types—of which philosophical and political beliefs are the paradigms—or, if they are to be held to the same epistemic standard as other beliefs, they typically fare worse under this standard than typical beliefs of most other types, including philosophical and political beliefs. I use this disjunctive formulation because, while I think I see some sort of difference thesis at work in much of the hostile writing on the epistemic status of religious belief, the work of this thesis is generally accomplished at a subliminal level and it is hard to get a clear view of it. I suspect that some of the writers I have alluded to are thinking in terms of one of the disjuncts and some in terms of the other.

A good example of the Difference Thesis at work is provided by Clifford's lecture. One of the most interesting facts about The Ethics of Belief is that nowhere in it is religious belief explicitly discussed. There are, to be sure, a few glancing references to religion in the lecture, but the fact that they are references to religion, while it doubtless has its polemical function, is never essential to the point that Clifford professes to be making. Clifford's shipowner, for example, comes to his dishonest belief partly because he puts his trust in Providence, but Clifford could have made the same philosophical point if he had made the shipowner come to his dishonest belief because he had put his trust in his brother-in-law. Clifford's other main illustrative case is built round an actual Victorian scandal (described in coyly abstract terms: There was once a certain island in which . . .) involving religious persecution. But he could have made the same philosophical point if he had described a case of purely secular persecution, such as those that attended the investigations of Senator McCarthy; his illustration turned simply on the unwillingness of zealous agitators, convinced that the right was on their side, to examine certain matters of public record and to obtain easily available testimony. In both of Clifford's illustrative cases, there is a proposition that is dishonestly accepted, accepted without sufficient attention to the available evidence. In neither case is it a religious or theological proposition. And at no point does Clifford come right out and say that his arguments have any special connection with religious beliefs. It would, however, be disingenuous in the extreme to say that The Ethics of Belief is simply about th e ethics of belief in general and is no more directed at religious belief than at any other kind of belief. Everyone knows, as the phrase goes, that Clifford's target is religious belief. (Certainly the editors of anthologies know this. "The Ethics of Belief" appears in just about every anthology devoted to the philosophy of religion. It has never appeared in an anthology devoted to epistemology. I know of only one case in which anyone writing on general epistemological questions has mentioned Clifford's lecture, and that is a very brief footnote in Chisholm's Perceiving, in the chapter entitled "The Ethics of Belief." In that note, Chisholm simply says that he holds a weaker thesis about the ethics of belief than Clifford'. Given that he had borrowed Clifford's title for his chapter-title, I suppose that that was the least he could have done.) The real thesis of Clifford's lecture, its subtext as our friends in the literature departments say, is that religious beliefs—belief in God; belief in an afterlife; belief in the central historical claims of Judaism or Christianity or Islam—are always or almost always held in ways that violate the famous ethico-epistemic principle whose quotation-name is my title: It is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence. If, moreover, he is of the opinion that beliefs in any other general category are always or almost always (or typically or rather often) held in ways that violate his principle, this is certain ly not apparent.

This conviction that Clifford's specific target is religious belief is no knee-jerk reaction of overly sensitive religious believers or of anti religious polemicists eager to find yet another stick to beat churchgoers with. If the conviction is not supported by his argument, in the strictest sense of the word, it is well grounded in his rhetoric. For one thing, the lecture abounds in biblical quotations and echoes, which is not a usual feature of Clifford's prose. For another, there are the inessential religious elements in both of his illustrative examples. Much more importantly, however, there are two passing allusions to religious belief, which, although they go by rather quickly, are nevertheless writ in letters that he who runs may read. First, one of the dishonest comforts provided by certain beliefs that are not apportioned to evidence is said to be this: they . . . add a tinsel splendor to the plain straight road of our life and display a bright mirage beyond it. Secondly, when Clifford raises the question whether it is fair to blame people for holding beliefs that are not supported by evidence if they hold these beliefs as a result of their having been trained from childhood not to raise questions of evidence in certain areas, he refers to these unfortunates as those simple souls . . . who have been brought up from the cradle with a horror of doubt, and taught that their eternal welfare depends on what they believe . . . .

Let us call Clifford's principle—It is wrong always, everywhere, and for anyone . . .—Clifford's Principle, which seems an appropriate enough name for it. I should note that there seems to be another principle that Clifford seems sometimes to be appealing to and which he neither articulates nor distinguishes clearly from Clifford's Principle. Call it Clifford's Other Principle. It is something very much like this: "It is wrong always, everywhere, and for anyone to ignore evidence that is relevant to his beliefs, or to dismiss relevant evidence in a facile way." Clifford's Other Principle is obviously not Clifford's Principle. It is very doubtful whether someone who satisfied the requirements of Clifford's Principle would necessarily satisfy the requirements of Clifford's Other Principle (it could be argued that it would be possible to have evidence that justified one's accepting a certain proposition even though one had deliberately chosen not to examine certain other evidence that was relevant to the question whether to accept that proposition) and it is pretty certain that someone who satisfied the requirements of Clifford's Other Principle would not necessarily satisfy the requirements of Clifford's Principle. I suspect that Clifford tended t o conflate the two principles because of a combination of his anti-religious agenda with an underlying assumption that the evidence, such as it is, that people have for their religious beliefs is inadequate because it is incomplete, and incomplete because these believers have declined to examine certain evidence relevant to their beliefs, owing to a subconscious realization that examination of this evidence would deprive even them of the power to continue to hold their cherished beliefs. However this may be, having distinguished Clifford's Other Principle from Clifford's Principle, I am not going to discuss it further, beyond pointing out that there does not seem to be any reason to suppose, whatever Clifford may have thought, that those who hold religious beliefs are any more likely to be in violation of Clifford's Other Principle than those who hold philosophical or political beliefs. We all know that there are a lot of people who have violated Clifford's Other Principle at one point or another in the course of arriving at their political beliefs and a few who have not. As to philosophy, well, I'm sure that violations of Clifford's Other Principle are quite rare among professional philosophers. No doubt there are a few cases, however. One might cite, fo r example, a recent review of a book by John Searle, in which the author of the review (Dan Dennett) accuses Searle of gross violations of Clifford's Other Principle in his (Searle's) descriptions of current theories in the philosophy of mind. If Dennett's charge is not just, then it is plausible to suppose that he is in violation of Clifford's Other Principle. So it can happen, even among us. But let us, as the French say, return to our sheep, prominent among which is Clifford's Principle- Clifford's Principle proper, that is, and not Clifford's Other Principle.

It is interesting to note that Clifford's Principle is almost never mentioned by writers subsequent to Clifford except in hostile examinations of religious belief, and that the anti-religious writers who mention it never apply it to anything but religious beliefs. (With the exception of illustrative examples—like Clifford's example of the irresponsible ship-owner—that are introduced in the course of explaining its content and arguing for it.) It is this that provides the primary evidence for my contention that many anti-religious philosophers and other writers against religion tacitly accept the Difference Thesis: the fact that they apply Clifford's Principle only to religious beliefs is best explained by the assumption that they accept the Difference Thesis. The cases of Marxism and Freudianism are instructive examples of what I am talking about. It is easy to point to philosophers who believe that Marxism and Freudianism are nonsense: absurd parodies of scientific theories that get the real world wildly wrong. Presumably these philosophers do not believe that Marxism and Freudianism were adequately supported by the evidence that was available to Marx and Freud—or that they are adequately supported by the evidence that is available to any of the latter-day adherents of Marxism and Freudianism. But never once has any writer charged that Marx or Freud blotted his epistemic escutcheon by failing to apportion belief to evidence. I challenge anyone to find me a passage (other than an illustrative passage of the type I have mentioned) in which any devotee of Clifford's Principle has applied it to anything but religious belief. And yet practically all philosophers—the literature will immediately demonstrate this to the most casual inquirer- subscribe to theses an obvious logical consequence of which is that the world abounds in gross violations of Clifford's Principle that have nothing to do with religion.

An explanation of the widespread tacit acceptance of the Difference Thesis among those who appeal to Clifford's Principle in their attacks on religious belief is not far to seek. If Clifford's Principle were generally applied in philosophy (or in politics or history or even in many parts of the natural sciences), it would have to be applied practically everywhere. If its use became general, we'd all be constantly shoving it in one another's faces. And there would be no comfortable reply open to most of t he recipients of a charge of violating Clifford's Principle. Use every man after his desert, and who shall scape whipping? If, for example, I am an archaeologist who believes that an artifact found in a neolithic tomb was a religious object used in a fertility rite, and if my rival, Professor Graves—a professor, according to the German aphorism, is someone who thinks otherwise—, believes that it was used to wind flax, how can I suppose that my belief is supported by the evidence? If my evidence really supports my belief, why doesn't it convert Professor Graves, who is as aware of it as I am, to my position? This example, of course, is made up. But let me mention a real and not entirely dissimilar example that I recently came across in a review (by Malcolm W. Browne) of several books about the Neanderthals in the New York Times Book Review (July 4, 1993, p. 1). The review includes the following quotation from the recent book The Neandertals by Erik Trinkhaus and Pat Shipman. The authors are discussing a debate between two people called Stringer and Wolpoff, who are leading experts on the Neanderthals. What is uncanny—and disheartening—is the way in which each side can muster the fossil record into seemingly convincing and yet utterly different syntheses of the course of human evolution. Reading their review papers side by side gives the reader a distinct feeling of having awakened in a Kafka novel. Assuming that this description of the use Stringer and Wolpoff make of their evidence is accurate, can it really be that their beliefs are adequately supported by this evidence? Will someone say that Stringer and Wolpoff are scientists, and that scientists do not really believe the theories they put forward, but rather bear to them some more tentative sort of doxastic relation? Regard as the best hypothesis currently available, or some such tentative attitude as that? Well, that is certainly not the way the author of the review sees the debate. Stringer, one of the parties in the debate, has written his own book, also discussed in the review, of which the reviewer says, In Search of the Neanderthals is built around Mr Stringer's underlying (and highly controversial) belief that the Neanderthals were an evolutionary dead end, that they simply faded away after a long and unsuccessful competition with their contemporaries, the direct ancestors of modern man. (That the Neanderthals were an evolutionary dead end is, by the way, the proposition that was at issue in the debate between Stringer and Wolpoff that was said to give the reader the feeling of having awakened in a novel by Kafka.) Later in the review, summarizing the book of another expert on human origins, the reviewer says, "In another section of the book, Mr Schwartz defends his belief that modern human beings are more closely related to orangutans than to either chimpanzees or gorillas." It is hard to see how to avoid the conclusion that it is very common for scientists qua scientists to have beliefs that are vehemently rejected by other equally intelligent scientists who possess the same scientific qualifications and the same evidence. Even in the more austere and abstract parts of science, even in high-energy physics, the current queen of the sciences, where there is some real plausibility in the thesis that investigators typically hold some more tentative attitude than belief toward the content of the controversial theories they champion, it is possible to find clear examples of this. To find them, one need only direct one's attention away from the content of the theories to the judgments that physicists make about the theories, their judgments about such matters as the usefulness of the theories, their "physical interest," and their prospects. A former colleague at Syracuse University, an internationally recognized quantum-gravity theorist, has told me, using a simple declarative sentence that contained no hedges whatever, that superstring theory would come to nothing. Many prominent physicists (Sheldon Glashow, for example) agree. They really believe this. And many prominent physicists (such as Steven Weinberg and Edward Witten) vehemently disagree. They really believe that superstring theory has provided the framework within which the development of fundamental physics will take place for a century.

But let us leave the sciences and return to our central examples, philosophy and politics. If we applied Clifford's Principle generally, we'd all have to become skeptics or agnostics as regards most philosophical and political questions—or we'd have to find some reasonable answer to the challenge, In what sense can the evidence you have adduced support or justify your belief when there are many authorities as competent as you who regard this evidence as unconvincing? But no answer to this challenge is evident, and religion seems to be the only area of human life in which very many people are willing to be agnostics about the answers to very many questions. (When I say very many people, I mean very many people like us: people who write books. It is, of course, false that a very high proportion of the world population consists of people who are willing to be agnostics about religious questions.)

It might, however, be objected that what I have been representing as obvious considerations are obvious only on a certain conception of the nature of evidence. Perhaps the Difference Thesis is defensible because the evidence that some people have for their philosophical and political (and archaeological and historical . . .) beliefs consists partly of the deliverances of that incommunicable insight that I speculated about earlier. This objection would seem to be consistent with everything said in The Ethics of Belief, for Clifford nowhere tells his readers what evidence is. If evidence is evidence in the courtroom or laboratory sense (photographs, transcripts of sworn statements, the pronouncements of expert witnesses, records of meter readings—even arguments, provided that an argument is understood as simply a publicly available piece of text, and that anyone who has read and understood the appropriate piece of text thereby has the evidence that the argument is said to constitute), then the evidence pretty clearly does not support our philosophical and political beliefs. Let such evidence be eked out with logical inference and private sense experience and the memory of sense experience (my private experience and my memories, as opposed to my testimony about my experience and memories, cannot be entered as evidence in a court of law or published in Physical Review Letters, but they can be part of my evidence for my beliefs—or so the epistemologists tell us) and it still seems to be true that "the evidence" does not support our philosophical and political beliefs. It is not that evidence in this sense is necessarily impotent: it can support—I hope—many life-and-death courtroom judgments and such scientific theses as that the continents are in motion. But it does not seem to be sufficient to justify most of our philosophical and political beliefs, or our philosophical and political beliefs, surely, would be far more uniform than they are. (Socrates told Euthyphro that people do not dispute about matters that can be settled by measurement or calculation. This is certainly false, but there is nevertheless an important grain of truth in it. There is indisputably significantly greater uniformity of opinion about matters t hat can be settled by measurement and calculation than there is about the nature of justice and the other matters that interested Socrates.) If evidence must be of the courtroom-and-laboratory sort, how can the Difference Thesis be defended?

If, however, evidence can include insight or some other incommunicable element—my private experience and my memories are not necessarily incommunicable—, it may be that some of the philosophical and political beliefs of certain people are justified by the evidence available to them. (This, as I have said, is the view I find most attractive, or least unattractive.) But if evidence is understood in this way, how can anyone be confident that some of the religious beliefs of some people are not justified by the evidence available to them? (I say some people; and that is probably all that anyone would be willing to grant in the cases of philosophy and politics. Is there anyone who believes that it makes sense to talk of philosophical beliefs being justified and who also thinks that the philosophical beliefs of both Carnap and Heidegger were justified? Is there anyone who holds the corresponding thesis about the political beliefs of both Henry Kissinger and the late Kim Il-Sung?) If evidence can include incommunicable elements, how can anyone be confident that all religious believers are in violation of Clifford's Principle? If evidence can include the incommunicable, how can the Difference Thesis be defended?

What I have said so far amounts to a polemic against what I perceive as a widespread double standard in writings about the relation of religious belief to evidence and argument. This double standard consists in setting religious belief a test it could not possibly pass, and in studiously ignoring the fact that very few of our beliefs on any subject could possibly pass this test. Let me summarize this polemic by setting out some Socratic questions; a complex, in fact, of alternative lines of Socratic questioning laid out in a sort of flow-chart.

Either you accept Clifford's Principle or not. If not, game ends. If so, either you think that religious belief stands convicted of some epistemic impropriety under Clifford's Principle or not. If not, game ends. If so, do you think that other important categories of belief stand convicted of similar epistemic impropriety under Clifford's Principle—preeminently philosophical and political belief? If you do, are you a skeptic as regards these categories of belief, a philosophical and political skeptic (and, in all probability, a skeptic in many other areas)? If not, why not? If you do think that the only important category of belief that stands convicted of epistemic impropriety under Clifford's Principle is religious belief—that is, if you accept the Difference Thesis—how will you defend this position? Do you accept my disjunctive formulation of the Difference Thesis: Either religious beliefs should be held to a stricter epistemic standard than beliefs of certain other types—of which philosophical and political beliefs are the paradigms—or, if they are to be held to the same epistemic standard as other beliefs, they typically fare worse under this standard than typical beliefs of most other types, including philosophical and political beliefs? If not, how would you formulate the Difference Thesis (and how would you defend the thesis you have formulated)? If you do accept my disjunctive formulation of the Difference Thesis, which of the disjuncts do you accept? And what is your defense of that disjunct? In formulating your defense, be sure to explain how you understand evidence. Does evidence consist entirely of objects that can be publicly examined (photographs and pointer readings), or that can, at least, for purposes of setting out descriptions of the evidence available for a certain thesis, be adequately described in public language (sensations and memories, perhaps). Or may what is called evidence be, or be somehow contained in or accessible to the subject in the form of, incommunicable states of mind of the kind I have rather vaguely called insight? If the former, and if you have chosen to say that a single standard of evidence is appropriate to both religious beliefs (on the one hand) and philosophical and politic al beliefs (on the other), and if you have decided that religious beliefs fare worse under this one standard than philosophical and political beliefs—well, how can you suppose that philosophical and political beliefs are supported by that sort of evidence, public evidence, to any significant degree? If the evidence available to you provides adequate support for, say, your adherence to a certain brand of functionalism, and if it is evidence of this straightforward public sort, then it is no doubt readily available to most philosophers who have paid the same careful attention to questions in the philosophy of mind that you have. But then why aren't most of these philosophers functionalists of your particular stripe? (Why, some respectable philosophers of mind aren't even functionalists at all, shocking as that may seem to some of us.) Wouldn't the possession and careful consideration of adequate, really adequate, evidence for a proposition induce belief in that proposition? Or, if evidence that provided ad equate support for a philosophical proposition was readily available throughout a sizable population of careful, qualified philosophers, wouldn't this fact at least induce a significant uniformity of opinion as regards that proposition among those philosophers?

If you take the other option as to the nature of evidence, if you grant that evidence may include incommunicable insight, can you be sure, have you any particular reason to suppose, that it is false that there are religious believers who have "insight" t hat lends the same sort of support to their religious beliefs that the incommunicable insight that justifies your disagreement with Kripke or Quine or Davidson or Dummett or Putnam lends to your beliefs?

This is the end of my Socratic flow-chart. I will close with an attempt to forestall two possible misinterpretations. First, I have not challenged Clifford's Principle, or not unless to point out that most of us would find it awkward to live by a certain principle is to challenge it. Clifford's Principle could be correct as far as anything I have said goes. Secondly, I have not argued that religious beliefs—any religious beliefs of anyone's—are justified or enjoy any particular warrant or positive epistemic status or whatever your own favorite jargon is. (For that matter, I have not argued that philosophical and political beliefs—any philosophical or political beliefs of anyone's—are justified or enjoy any particular warrant or positive epistemic status. I have recorded my personal conviction that some philosophical and political beliefs are justified, but I have not argued for this conclusion. I do not mind—just for the sake of literary symmetry- recording my personal conviction that some religious beliefs are justified, but that they are is not a part my thesis.) There is one important question that bears on the epistemic propriety of religious belief that I have not even touched on: whether some or all religious beliefs may go clean contrary to the available evidence—as many would say the belief in a loving and all-powerful deity goes clean contrary to the plain evidence of everyone's senses. To discuss this question was not my project. My project has been to raise certain points about the relevance of Clifford's Principle to the problem of the epistemic propriety of religious belief. These are different questions: it suffices to point out that the philosopher who argues that some religious belief—or some belief of any sort—should be rejected because it goes contrary to some body of evidence is not appealing to Clifford's Principle. If what I have said is correct, then philosophers who wish to mount some sort of evidential or epistemic attack on religious belief (or, more likely, not on religious belief in general, but on particular religious beliefs) should set Clifford's Principle aside and argue that religious belief (or this or that religious belief) is refuted by the evidence they present.3


1. Lectures and Essays, (London: Macmillan, 1879). Variously reprinted.

2. Shortly after this paper was written, Louis P. Pojman published the anthology The Theory of Knowledge: Classical and Contemporary Readings (Belmont, Calif.: Wadsworth, 1993), which contains "The Ethics of Belief". It should be noted that Pojman is a Christian philosopher who has written extensively on the epistemological problems of religious belief; his writings display an abiding interest in the issues raised in The Ethics of Belief.

3. Versions of this paper were read at the 1993 Chapel Hill Philosophy Colloquium and to departmental colloquia and student groups at the University of Alabama at Tuscaloosa, Tufts University, and the University of Miami (Coral Gables). I wish to thank those who were present on these occasions for their questions and comments, especially Simon Blackburn, Jarrett Leplin, William G. Lycan, H. Scott Hestevolt, George Smith, Daniel Dennett, Susan Haack, and Eddy Zemach. I also wish to thank David Lewis, both for conversation and correspondence, over many years, that have had a profound influence on my ideas of the nature of philosophy, and for conversations on the specific topic of this paper. Some of the ideas in this paper—and, indeed, some of the paragraphs in this paper—appeared in my essay "Quam Dilecta," which was included in Thomas V. Morris, ed., God and the Philosophers (New York and Oxford: Oxford University Press, 1994).