1. In the course of my last lecture I suggested that some part at least of the hostility of so much contemporary religious thought towards philosophy was due to a failure to keep the distinction clear between two possible offices of reason in religion: between reason as an organ for the apprehension of the Divine, and reason as the ultimate arbiter upon claims to such apprehension. It is only in the latter office, I argued, that philosophy is bound by its inherent nature to insist upon reason's competence. If it were more generally appreciated that for philosophy it is an entirely open question, not a question to which the answer is pre-determined, whether God (if there be a God) is apprehensible by reason, the gulf between the attitudes of religion and philosophy would not, I think, appear as unbridgeable as it is taken to be in many religious quarters today.

At the same time, it is only fair to observe that the confusion between these two possible offices of reason is by no means confined to reason's critics. The Rationalist too is often deceived. Profoundly convinced that reason must be held to be supreme in determining religious truth no less than in determining truth elsewhere, he slips easily into an interpretation of this which implies that if there is a God to be apprehended at all, reason is competent to apprehend Him. Now the latter proposition may, of course, be perfectly sound. But it does not follow as a mere corollary of the former proposition, as the Rationalist is apt to assume. There is no prima facie contradiction in holding that reason is not a competent organ for the apprehension of the Divine, while also insisting upon reason's supremacy in the determination of religious truth in the sense that it, and it alone, can adjudicate upon conflicting claims to an apprehension of the Divine.

In the present lecture I want to develop and illustrate the thesis that reason is ultimate arbiter by considering, more in concreto, the kind of attitude which it seems proper for reason to adopt towards claims to religious ‘revelation,’ and the principles or criteria which it seems proper for reason to apply in carrying out its task of appraisal. The need for such appraisal arises primarily from the simple fact that ‘revelations’ are not self-certifying. They carry with them no native hall-mark of authenticity, no internal characteristic by which we can confidently distinguish the genuine article from the spurious counterfeit. There is great point in Hobbes's observation that for a man to say that God ‘hath spoken to him in a dream, is no more than to say that he dreamed that God spake to him.’¹ The strength of a man's own conviction is, of course, almost worthless as a criterion; for mutually contradictory ‘revelations’ can be accompanied by equally strong subjective conviction. The authenticity or otherwise of an ostensibly revelational experience has to be determined by quite other considerations which I now wish to discuss.

2. Among the first tasks of reason in this field will be, I suggest, to reflect critically upon the general notion of ‘revelational’ experience in religion, in the light, more especially, of the vast increase of knowledge about the workings of the human mind that has been made available in the last half-century. Ostensible revelations have certain typical characteristics which can be, and often have been, described, and about which there is substantial agreement. Now it would be absurd to ignore the fact that there are reputable and competent students of the mind who, after a careful survey of ostensibly revelational experience, have reached the considered conclusion that there is nothing in that experience which cannot be explained ‘subjectively’ or ‘psychologically.’ The content of the experience, so it is held, is not as it appears to the subject to be, viz. an object disclosed to him. It is an object projected by him. It is a ‘fantasy’ whose function it is to give fulfilment to certain wishes, hopes, etc., of whose very existence in his mind the subject may be unaware. Such fantasy projection—not always, of course, simulating a ‘religious’ experience—is now admitted to be a fairly common occurrence in the maladjusted mind, and a good deal is known about the inner tensions which occasion it. It is also admitted, even by those most sympathetic towards the general notion of religious revelation, that at least some, and probably very many, of the experiences which seem to the subject of them to be religious revelations must be interpreted as falling in fact into the category of wish-fulfilments. Inevitably the question arises, if this be true of some, may it not be true of all?

The formal discussion of this and allied problems concerning the nature of religious experience I must defer until our second course. I introduce the topic now merely by way of illustrating what seems to me an obvious office of reason as arbiter in the sphere of religious truths. That the psychological facts relating to ostensible revelations do have a bearing upon questions of religious truth seems undeniable. It seems equally undeniable that only reason can appraise their significance.

But perhaps we are going too fast. Do facts about the psychological origins of revelational experience have a bearing upon the question of the validity of the experience? Are we not, in suggesting that this is so, falling into the ancient error of confounding questions of origin with questions of validity?

Now I agree that there is an ‘ancient error’ to be detected here; but I also think there is a good deal of confusion about wherein precisely it lies. It is this confusion, perhaps, which is responsible for the curiously ambivalent attitude on the part of so many ethical and religious writers towards it. On the one hand they are very ready with their retort to their Naturalistic critics that ‘an account of the origin of a concept tells us nothing about its validity.’ On the other hand they almost all seem extraordinarily eager to prove that as a matter of fact such concepts as ‘God’ and ‘Duty’ do not originate subjectively.

As I see it, everything turns here upon what sort of ‘origination’ we are thinking of. The term can be understood in either of two broad ways. Understood in one of these ways, the question of origin does not affect the question of validity. Understood in the other way, it does. If we fail to lay firm hold of the distinction, we naturally oscillate in attitude according as one or the other meaning is before our minds.

The distinction in meaning may perhaps be conveniently described as that between ‘explanatory’ origination and ‘factual’ origination. The antecedent mental processes which are the alleged subjective origin of an idea may be called its ‘explanatory’ origin if, in terms of them alone (plus, of course, the operation of independently established laws of psychology), we are able to understand how the idea comes to be formed; they may be called its ‘factual’ origin, on the other hand, if we are not able to understand how, but merely find that, the idea in question follows regularly upon their occurrence. To illustrate. Freud would contend that the subjective origin to which he ascribes the idea of God is an explanatory origin, in that, as he believes, given the antecedents he specifies, the generation in the mind of the idea of God is fully intelligible. An instance of subjective origination claimed to be no more than factual would be the sort of account which the non-naturalistic moralist gives of the emergence of the idea of ‘duty’ in the mind of primitive man, or in the mind of the ordinary child, where (so it is insisted) the idea when it does emerge contains something genuinely new, something quite incapable of being understood in terms of its psychical antecedents whether taken separately or in conjunction.

Let us, then, look at the relation of origin to validity in the light of this distinction. It seems obvious that in the case of the merely factual origination of an idea no inference is admissible concerning its objective validity. The idea is, ex hypothesi, not accounted for by the subjective processes, and there is therefore nothing, so far, to set against the natural presumption that we apprehend the object because it is there to be apprehended. But if, on the other hand, the origin does really explain the idea, it seems to me that we are bound to accept this as casting some shadow of doubt upon the idea's objective validity. Of course it does not disprove the independent existence of an object corresponding to the idea as conceived. But it does entail that we are no longer in a position to appeal to what is normally taken as an important part of the evidence for the independent existence of an object, namely, the fact that we have an idea of it which we apparently cannot explain as arising from within our own minds.

To repeat. To exhibit the merely factual origination of an idea in subjective processes raises no presumption against its objective validity. To exhibit its explanatory origin does. Even then, the object may have independent existence. But our having an idea of it can no longer be taken as establishing any presumption to that effect.

I suggest, therefore, that it is by no means true without qualification that the question of the origin of an idea has no bearing upon the question of its validity. It all depends upon whether the origin specified can or can not be regarded as ‘explanatory.’ It follows that psychological accounts of the subjective origination of ‘revelational’ ideas may be very relevant indeed to the attempt to determine the truth of religion: and it is, I submit, one of the functions of reason as arbiter, and thus of the philosopher as conducting reason's arbitration, to determine just how much (or how little) relevance they have.

3. Let us assume, however, that reason's examination of ostensibly revelational experience in general leaves it as at least an open question whether or not some instances of it are authentic revelations, disclosures of the nature of an objectively real being. The task remains for reason as arbiter to discriminate, with the highest measure of probability that the nature of the case permits, between the authentic and the spurious among the ostensible revelations. What, then, are the criteria which it is proper for reason to adopt in the execution of this task? I make no claim in what follows to be listing the criteria exhaustively, much less to be saying all that requires to be said about any one of them. But I shall deal in some detail with those criteria which seem to me to be the most important.

The criteria may be conveniently divided into (a) psychological criteria and (b) logical criteria. First, then, as to the psychological criteria.

4. These are applicable, of course, only where we have a good deal of information about the individual who claims the revelational experience. We require to know something of his personality, of his mental history, of his mode of life, and of the circumstances immediately preceding the alleged revelation. Sometimes we are not in a position to learn much about these matters; but where, as is not infrequent, fairly detailed knowledge is available, it can, I think, give legitimate aid in determining the authenticity or objectivity of the experience. For example, religious revelation, if it is anything, is revelation of a spirit to a spirit. We should therefore expect it to occur only in persons whose normal lives indicate that they are, in some measure, ‘spiritually oriented.’ Some degree of spiritual preparedness, we presume, must precede spiritual discernment. It seems unlikely that—to take an extreme case—a drunken profligate engrossed throughout most of his waking hours in fleshly pleasures should have the capacity for attaining exalted visions of the Divine nature. In a similar way it offends our sense of the fitness of things to suppose that the taking of drugs, e.g. the inhalation of nitrous oxide, can yield, as it does sometimes seem to the person concerned to be yielding, profound insights into the very heart of being (the ‘anaesthetic revelation,’ as it has been called). We cannot, of course, rule out a priori the possibility of God revealing Himself for purposes of His own through what may appear to us rather surprising and improbable media. Nevertheless we do seem justified in taking it as evidence, so far, for the authenticity of an ostensible revelation that it occurs in a person whom we have reason to believe to be attuned to things of the spirit, and as evidence against where the experiencing subject is notoriously preoccupied with material concerns.

In the second place, in view of the virtual certainty that some ostensible revelations are fantasies unconsciously produced by the subject himself, psychological study of the experiencing subject can usefully be directed to ascertaining whether he appears to be the kind of inhibited and generally maladjusted person from whom, in accordance with the teachings of psychopathology, one might expect fantasy-projection as a means of effecting substitute satisfactions for unfulfilled desires; and more particularly to ascertaining whether, if a maladjustment is diagnosed, its specific character bears a discernible relationship to the specific content of the ‘revelation.’ The excesses in the way of forced interpretations which we find in over-enthusiastic practitioners of depth psychology may reasonably incline one to observe special caution in this line of enquiry; but few, I think, who have any considerable acquaintance with the strange forms which ‘revelational’ experience often takes would wish to discount the usefulness of such psychological techniques in discriminating the spurious from the authentic.

5. From these perhaps too cursory observations upon the psychological criteria I pass on to a somewhat closer consideration of the logical criteria.

By the ‘logical’ criteria I understand criteria relating directly to the content of the ostensible revelation; to the actual doctrines it incorporates or implies concerning God's nature and His relation to the world and to the human soul. Reason requires that the propositions affirmed explicitly or implicitly in the ‘revelation’ be propositions which reason can accept as true. This does not mean, be it noted, that they must be propositions which reason can see to be logically self-evident, or to be necessary implications of propositions that are logically self-evident. To require that would be to insist that reason is, after all, not merely the arbiter but also the organ of religious truth. What it does mean is that the propositions in question must, at the very least, not violate the principle of self-consistency. So understood, the logical criterion takes two main forms. We require that the propositions inherent in the revelation be consistent with one another. And we require also—although this is a condition which must be elaborated and in some degree qualified—that they be consistent with well-accredited propositions about reality got through other channels.

So far as the first form of the logical criterion is concerned—internal self-consistency—the demand of reason is absolute. If the content of the ‘revelation’ contains a definite self-contradiction, it cannot be a revelation of the truth. There are really only the two alternatives before reason. Either it declines the task of appraising ostensible revelations altogether; in which case the question arises—as unanswerable as it is inevitable—how then are we to distinguish between the authentic and the spurious? Or, accepting the task, reason proceeds in accordance with its own inherent principles; in which case it must reject as untrue whatever contains a self-contradiction.

But there is an important caveat to be entered here. The criterion of internal self-consistency is absolute. But its application to religious utterances is very far from being always a simple matter. It must be remembered that religious utterances abound in deliberate paradox; in statements which at first glance are self-contradictions, and which if interpreted in a literal, everyday sense are self-contradictions, but which, in the meaning the words bear for the person uttering them, are not self-contradictory at all. When the mystic Tauler describes God as ‘a rich naught,’ is he committing a self-contradiction? Is it enough to say that, since ‘richness’ and ‘nothingness’ are terms which negate one another, the proposition is just self-contradictory nonsense?

Surely not. An attempt must first be made to understand the proposition in the precise sense it had for its author. We see then that the significance of applying the term ‘naught’ to God lies for Tauler in the fact that, in the case of a Being who is (as Tauler takes God to be) all-inclusive, no determinate quality can fitly be predicated. For ‘all determination is negation.’ When Eckhart proclaims that ‘God is neither this nor that,’ he is expressing virtually the same insight. To ascribe a particular predicate to God is thereby also to exclude something from Him, and therefore to limit from without that which is ex hypothesi all-inclusive, limited only by itself. In short, the real significance of the term ‘naught’ in Tauler's proposition consists not in lack of being, but rather in its very opposite, plenitude of being; and hence the prefixing to it of the term ‘rich,’ while apparently contradictory, is in reality entirely apt.

Or take Eckhart's saying that ‘the first day and the last are happening at the present instant yonder’. This too is easy enough to reduce to a prima facie contradiction. The ‘present’ instant has no meaning for us save through relation to ‘past’ and ‘future’ instants distinct from itself. But according to the words of the proposition there are no past or future instants ‘yonder’—sc. ‘in the Divine Reality’. ‘First day’ and ‘last day’ are declared to be comprehended in ‘the present instant’, which is thus the only instant. Accordingly, the proposition seems at once to be affirming a ‘present instant’, and denying the necessary condition of there being a present instant, viz. ‘past’ and ‘future’ instants.

But the reader of Eckhart knows very well that Eckhart does not mean these words to be understood as a literal expression of the truth. On the contrary, few thinkers have been more pertinacious than Eckhart in stressing that the category of Time, which he here takes for granted, is inadequate to the apprehension of God as He truly is. God's being is eternal—time-transcending. It seems clear that Eckhart would claim for his saying only a symbolic validity. The ‘truth’ that is being symbolised is that in God there are no time-distinctions. It is symbolised by the image of an all-embracing present—‘the first day and the last are happening at the present instant yonder’; which is not a bad symbol, surely, as symbols go. Eckhart is not addressing the philosopher. He is addressing the ordinary devout man, and accommodating his language thereto; giving to the truth a symbolic form which might enable the ordinary man to come as near to grasping it as it is reasonable to expect.

I have been taking examples from the utterances of the mystics, who are intensely aware, and constantly reiterate, that their language for describing the Deity has no more than a symbolic significance. It is necessary to recognise, however, that there are countless religious utterances, purporting to give the content of revelation, in which the speaker does intend his words to be understood quite literally. Against these the claims of ‘non-contradiction’ must be satisfied in a plain straightforward way. At the same time one has to bear in mind that such utterances may have had for their original authors a purely, or largely, symbolic import, even although they are now proclaimed as literal truth by uninspired disciples. Where that is the case, the exposure by reason of a self-contradiction in their literal significance may still have a value. It may stimulate, or help to stimulate, a religious creed which has become arid and formal to recapture the original signification of the revelational experiences which were its living roots.

On the relationship of symbolism to religious truth I shall in the second course have a good deal to say. Indeed the second course might not be unsuitably sub-titled ‘An Essay in Symbolic Theology’. But in the present context these few words must suffice.

6. Turning now to the second form of the logical criterion-consistency of the content of ostensible revelation with well-accredited knowledge got through other media—the great difficulty here is to decide when a piece of knowledge got through other media is sufficiently well-accredited to justify the rejection of a ‘revelational’ content inconsistent with it. Ideally, these other propositions should be so well-accredited as to be completely certain. If not, it is always possible that it is these other propositions that are in error, and that the religious utterances may be true. But most of us would agree that propositions about objective reality, got through ordinary channels, for which anything like complete certainty can be claimed, are very few indeed.

Nevertheless one is bound to hold, I think, that where an ostensible revelation is in conflict with a rationally grounded proposition for which the competent judges in the appropriate field claim a very high degree of probability, the likelihood that the ostensible revelation is authentic is, at the very least, sensibly weakened. It would be foolish to pretend, e.g. that confidence in a ‘revelation’, in terms of the Book of Genesis, about the historical origin of living things—as in George Fox's celebrated vision of the Creation—can remain unaffected by its contradiction of a biological proposition so amply evidenced as that which affirms the evolution of species. There are probably few professional biologists who would not regard a denial of the theory of evolution (as distinct, of course, from a denial of some theory concerning the mechanism by which evolution proceeds) as almost tantamount to a vote of ‘no confidence’ in the capacity of reason to achieve any sort of systematic knowledge about the natural world. And while there is a case for the view that reason is not a competent organ for the knowledge of the supernatural, the power over nature which science has enabled man to acquire seems sufficient answer to any suggestion that reason is not competent to deal even with the natural. The least that one could demand before agreeing that a ‘revelation’ of the simultaneous creation of all species should be taken seriously is a tolerably plausible alternative explanation of the vast multitude of facts enumerated by the biologist in support of his evolution theory. Of such plausible alternatives there seem to be none. I do not feel called upon to explain why I do not accept as ‘tolerably plausible’ the hypothesis that the facts upon which the biologist relies—fossil remains and the like—are deliberately deceitful evidence planted by God in order to test the firmness of man's faith when it comes into conflict with even the best founded deliverances of his reason. I suppose it is in a sense inevitable that man should construe God after the image of man; but there seems no need to construe Him after the image of a rather disagreeable kind of man.

It is well to bear in mind, however, that the just claims of scientific knowledge vis à vis the claims of religious revelation are only its claims as scientific knowledge. The task of natural science is to discover uniformities of sequence and coexistence between physical events in the space-time order. Success in that undertaking entities the scientist to make, with varying degrees of confidence, retrospective judgments about the past course of events in that order, and prospective judgments about the future course of events. There is nothing whatever in his specialised scientific training which qualifies him to talk about the universe; which, at the lowest estimate, is something much more than a space-time order of physical events. That the expert scientist is somehow possessed of such qualifications is a myth which the greater among them discourage by precept, and the less great by awful example. Yet the myth dies hard. Presumably its rational basis is the belief, plausible but quite untrue, that the scientist has acquired an exceptional competence in the assessment of evidence. He hasn't. He has probably acquired exceptional competence in assessing a special kind of evidence—that relating to his own and allied fields of specialist study. But competence in the assessment of evidence is not transferable directly from one field to a quite different field. Indeed it is hardly possible to know even what constitutes evidence in a field markedly different from one's own without the arduous preliminary of prolonged personal immersion in its subject-matter. One may know very well what kind of evidence to look for in order to evaluate the scientific proposition that the mosquito Anopheles is a carrier of malaria, and yet have not the faintest notion how to set about evaluating the philosophic proposition that a unitary self-consciousness is a condition of the possibility of significant experience. The singular naïveté which so many eminent scientists display in their obiter dicta upon social, economic, and political principles ought to be sufficient warning that their pronouncements in the spheres of religion and metaphysics—where the relevant evidence is usually still less familiar to them than the relevant evidence for social, economic and political theory’—are, to say the least of it, unlikely to be instructive.

Actually, since the primary concern of religion is with the ultimate ground of all being, and since this is in no way science's concern, it is not immediately obvious why the two should ever come into conflict at all. Conflicts do arise, however, and they are not wholly due to misunderstanding of the respective provinces of religion and science. The most important reason for this is, I think, that what goes on in the Space-Time world of Nature cannot be irrelevant to the view we take of Nature's ultimate ground. Conflict will occur—not necessarily, of course, irresolvable conflict—if science, in pursuit of her proper avocation, presents us with a picture of Nature which is, in some of its phases not easy to reconcile with the character which religion typically ascribes to Nature's ultimate ground; not easy to interpret in terms of an all-wise, all-powerful and supremely good Creator. Again, religious ‘revelations’ often enough extend (as we have seen in the case of George Fox's vision) to the manifestations of God's Will in the order of Nature, and may purport to give a comparatively detailed historical account of the course of events in certain important aspects. In so far as that is the case, a direct collision with science is in principle possible. I am inclined to think, however, that collisions induced in the latter manner are for the most part on the periphery of religion rather than at its centre. That is to say, not much of real consequence would be lost to religion if it were frankly to disavow all ‘revelations’ about the historical order that are clearly inconsistent with established scientific knowledge. For those ‘revelations’ whose religious significance is manifestly profound do not as a rule relate to occurrences about which there is well-accredited scientific knowledge. It could not be maintained, e.g. that the momentous religious proposition, founded upon revelation, that God was incarnated in the historical figure of Christ, conflicts with well-accredited scientific knowledge in the decisive way in which, e.g. the ‘revelation’ of the simultaneous creation of animal species does. I fancy that few of those who have arrived at a considered rejection of the proposition that God was incarnated in Christ, and hence of the authenticity of the ‘revelation’ on which it is based, would wish to support their denial by calling in evidence anything that would normally go by the name of ‘scientific’ knowledge at all.

But that very fact serves to remind us that the well-accredited knowledge from other sources with which reason demands that ‘revelation’ be consistent need not be confined to scientific knowledge. What, e.g. of the propositions of philosophy? It is obvious that where collision between philosophy and religion does occur, it must be of a peculiarly crucial character. For philosophy and religion are aiming, in their different ways, to do the same thing. Each of them, unlike science, aspires to apprehend the real in its ultimate nature, and in its totality. Potentially, therefore, religion would seem to have a great deal more to fear from philosophy than from science.

In actual practice, nevertheless, I think most religious people feel that they have a great deal less to fear from philosophy than from science; or at any rate from philosophy in its constructive, as distinct from its purely critical, manifestations. And the reason is not far to seek. It is simply that there are so very few, if indeed there be any, propositions of constructive philosophy which constitute ‘well-accredited knowledge’ in the sense in which a great many scientific propositions can claim that title, viz. as enjoying the assent of all competent judges. This lack of concord among the philosophers should not greatly surprise anyone who has reflected upon the nature of the philosopher's task and the kind of evidence that is appropriate to it. Still, the result is to leave the way open for a very obvious riposte from the defender of a religious view of the universe, if that should be challenged by some philosophical view. He can always say, with evident point, ‘But aren't there any number of philosophers who reject this philosophical view? And if so, why should religion trouble itself over-much about it?’

At the same time it is easy to make too much, in the religious context, of these mutual disagreements of the philosophers. After all, are there so very many religious propositions that command the assent of all competent judges in the field of religion? We cannot deduce from the differences of opinion in philosophical theory that constructive philosophy is incapable in principle of achieving a metaphysic of reality, and may therefore be justly ignored by religion. Propositions of constructive philosophy that are opposed to an ostensible revelation may be true, even although thay have behind them no solid consensus of philosophic opinion. And it remains the case that, from the standpoint of the individual person reflecting upon the claim to validity of some ostensible revelation, that revelation must be accounted at least suspect if it contradicts any proposition which the individual himself is, on philosophical grounds, strongly disposed to believe. Awareness of the extent to which philosophers disagree among themselves in their constructive thinking ought certainly to discourage sharply any tendency to dogmatism on the part of the individual thinker. But it need not, and it clearly does not, prevent him from feeling at least considerable confidence about the truth of some philosophical propositions. This confidence is probably more often felt with regard to propositions in moral philosophy that have metaphysical bearings, than with regard to what are ordinarily called ‘metaphysical’ propositions. Many thinkers by no means inclined to dogmatism have arrived on grounds of reason at strong convictions about the truth of propositions concerning, e.g. the objective reality of duty, or the objective reality of moral responsibility. Where that is so, the individual thinker has no option but to require, with a degree of stringency proportionate to the degree of his certainty, that revelation be consistent with the proposition or propositions in question. Thus if reflection upon moral experience induces in a man a strong conviction of the reality of personal freedom in moral choice, he will rightly feel doubt about the authenticity of some ‘revelation’ which proclaims or implies that human choices are all preordained from the beginning of time.

There is, indeed, one alternative (if such it can be called) in which refuge may be sought by those who are fearful of letting their philosophical beliefs bear upon their religion. There is always the ‘double-truth’ hypothesis, according to which a proposition may be ‘true for reason’, and a proposition which directly contradicts it be ‘true for religion’. I am afraid I can find nothing to commend in this endeavour to run with the hare and hunt with the hounds. To say that we can accept a proposition as true qua religious being, while rejecting it as false qua rational being, is to ignore the plain fact that we are for ourselves not two beings but one. We can acquiesce in our own self-diremption only if we fail to notice it. Admittedly this schizophrenic condition is not altogether rare. But where a man does notice his own ‘doublethink’, where he does realise that he is subscribing to two mutually contradictory ‘truths’, the question surely forces itself upon him ‘But which of the “truths” is really true?’ And since it is the ‘really true’ that ‘really matters’, the ‘not really true’ must simply give way before it.

This lecture has been (so far) an attempt to reinforce the thesis of its predecessor by illustrating what I take to be the legitimate rôle of reason vis à vis the claims of revelation. In some measure, I confess, I have seemed to myself to be labouring the obvious. I do not find it easy to understand how anyone who has even a nodding acquaintance with comparative religion and modern psychology can seriously suppose an ostensible revelation—even where he is himself the experient—to be self-sufficient, in no need of support from any other quarter. On the other hand, it is hard not to interpret the anti-rational trend in much religious writing today as a direct encouragement to men to rest content with a faith that neither knows nor seeks any justification beyond itself. Perhaps, therefore, there may still be something to be gained by reasserting the ancient truth that the only faith that is fitting in a rational being is a faith that is buttressed by reason. As Socrates almost said, ‘An unexamined faith is not worth having.’ ‘Childlike’ faith is undoubtedly a beautiful thing; but only, I would suggest, in a child.

7. I have been speaking of the need that faith should be buttressed by reason, and I have been arguing throughout for reason as the ultimate arbiter. Before concluding this general justification of reason's claims, however, I feel bound to notice, on account of its considerable popularity among theologians, a line of thought which seeks to turn the tables upon the philosopher by showing that reason needs itself to be buttressed by faith. The argument is advanced in a variety of forms, and by many theologians who are also competent philosophers. I am not aware, however, of any version that adds appreciably to what is contained in Principal Galloway's admirably fair-minded essay on ‘Knowledge and Religious Faith.’² It should suffice if we take this as our text.

Scientists and philosophers, Galloway declares, are not ‘free to object to faith as an attitude of mind to the object, for… faith enters into the attitude of the man of science as well as into that of the philosopher’ (p. 52). How does it enter in? At least three ways are indicated.

(1) In respect of the ultimate criterion of truth, ‘non-contradiction.’ ‘The principle of non-contradiction… rests, it seems to me, on a postulate of faith’ (p. 51).

(2) In respect of the scientist's attitude to his hypotheses. ‘Every fresh advance in science is won through belief or trust in a theory or hypothesis which goes beyond the immediate data of experience. The labour of verification is sustained by this trust’ (p. 50).

(3) In respect of inductive arguments giving ‘the empirical generalisations we call laws’. ‘If we believe, and act on the belief, that these empirical generalisations will work in the future as they have done in the past, then our action is ultimately grounded on faith, not on proof’ (p. 50).

Now it seems to me clear that the faith of the philosopher or scientist in each of these instances differs from religious faith in a manner that wholly invalidates the argument.

To begin with (1), The assumption is that the principle of non-contradiction is not capable of being proved. Now this is no doubt the case if we limit the term ‘proof’ to deductive demonstration from some higher principle. Clearly there is in this case no higher principle. But such a limitation has surely little to be said for it? The proof of the principle of non-contradiction is that it is a principle which we cannot help accepting if we are to think at all.³ It cannot intelligibly be called in question, since the critic must himself assume it at every stage of his argument. But if a principle can be shown to be such that it cannot be significantly doubted, what stronger proof could anyone want of its validity? And if objection continues to be felt to this use of the term proof, let us, if it be preferred, speak of ‘validation’. All that really matters is the recognition that here we have a principle doubt, of which is an absurdity. Where that is the case, to speak of it as, or as resting upon, a ‘postulate of faith’, is surely a misuse of language.

In respect of (2) and (3) there is indeed something in the scientist's attitude of mind more like what is ordinarily meant by ‘faith’; but it is not, I think, very like what the religious man ordinarily means by faith. In both (2) and (3) the faith in question is a belief held provisionally; founded upon the evidence so far attained, but held in full consciousness of the possibility of fresh evidence accruing which will conflict with it and compel its revision or even total abandonment. ‘Faith’ is here in contrast with ‘certainty’. But it would seem to be of the essence of faith as normally understood in religion that it does not stand in contrast with certainty, that it just is not a ‘tentative’ attitude of mind. In so far as the religious man admits a contrast of faith with knowledge, it is with knowledge not in respect of certainty, but rather in respect of the foundations of certainty. It is a contrast between a certainty gained by some form of non-or supra-rational insight, and a certainty gained by ordinary rational processes. The kind of ‘belief’ which constitutes religious faith is thus not, like (2) and (3), held merely provisionally. Its evidence is taken to be already adequate. There is no admission that further evidence may be forthcoming which will invalidate it.

Now this is an ‘attitude of mind to the object’ against which the scientist and philosopher are ‘free to object’. No doubt their objections may be invalid; but they are not (as Galloway seems to suggest) self-stultifying, directed against an attitude of mind which they themselves adopt.

Oddly enough, Galloway himself recognises in an earlier part of his essay the difference to which we have been drawing attention between faith in its religious sense and the ‘faith’ referred to in (2) and (3) above. He tells us that ‘to believe may merely mean to hold for true on satisfactory evidence’, or again to ‘suppose or opine something for which the evidence is not complete’ (p. 25). But such belief, he goes on to point out, ‘has a fluctuating character; the attitude is a provisional one, and differs essentially from the full assurance of faith’ (p. 26). I entirely agree. But the author cannot, I suggest, have it both ways. If the faith of the scientist and philosopher is an attitude differing ‘essentially’ from religious faith, the scientist and philosopher cannot be prohibited from taking objection to religious faith on the score that they themselves evince the same attitude of mind in their own characteristic activities.

It is not to be denied, of course, that ‘faith’, even in its religious context, is sometimes used with a suggestion of imperfect certainty. But where this is so, there is no quarrel between philosophers and scientists on the one hand and the spokesmen for religion on the other. Philosophers and scientists, admittedly, are ‘not free to object’ to religious faith as so understood. But then they don't want to. What they take exception to is the faith that repudiates the possibility that it can be mistaken, and which in consequence refuses a fair hearing to all considerations hostile to it which might be advanced from the side of reason.