But there is another and much more obvious threat to the meaningfulness of our question. Although the expression ‘the truth of religion’ is so often on people's lips one does not require to be a philosopher to feel some doubt whether if taken strictly any precise significance can be attached to it. For ‘religion’ is instantiated in a host of different ‘religions’ whose characteristic beliefs conflict with one another at many important points. How can we possibly enquire into the truth of something that comprises mutually contradictory sets of propositions? It would seem to make sense to ask ‘Is religion true?’ only if we arbitrarily choose to mean by the term ‘religion’ not religion as such religion in its generic nature but some specific religion (e.g. the Christian religion) or at most some specific group of religions (e.g. the theistic group) which closely approximate to one another in certain cardinal tenets.

And no doubt some such arbitrary meaning is in fact in most people's minds when they raise the question of ‘the truth of religion’. Nevertheless it seems to me that the question can be intelligibly raised concerning religion as such or in its generic nature. For the name ‘religion’ is presumably given to each of the different acknowledged religions on account of something taken to be common to them all. We imply in naming them all ‘religions’ that there is some set of essential characteristics in virtue of which the term ‘religion’ is used and that each of these religions does in fact exhibit this common essence. Now if this common essence includes (as it surely must) some body of belief. we may legitimately ask ‘Is religion true?’ meaning ‘Are those beliefs which pertain to the common essence of religion true beliefs?’

Our question then would seem to be intelligible enough. But is it also a worth-while question? May it not be the case that when we have tracked down those residual elements of belief that are identical in everything that we recognise as ‘religion’ we shall find this common nucleus to be so minute that the question of its truth-value will excite nobody? That is in the abstract a genuine possibility. I do not myself believe however that it is the fact. It seems to me that the common essence of religious belief is both substantial and of the highest significance. But whether or not this claim is well-founded will depend upon the validity of the definition of the term ‘religion’ (or—what is the same thing from another angle—of the analysis of the general concept of religion) towards which I shall be gradually working my way in the course of today's lecture.

Now prima facie no enterprise would seem better suited to an opening lecture in the field of philosophy of religion than the attempt to give a definition of religion. Yet we have to recognise at once that objection may be taken to it precisely on the ground of its unsuitability for an opening lecture. For are we not constantly being told by eminent scholars in this field that a definition of religion is something that can be given only at the end—that it is the terminus ad quem of the study of religion not the terminus a quo presupposing for its intelligent statement an extensive survey of and intensive reflection upon the phenomena of religion?

This is an extremely common view; but it seems to me to be mistaken and to get such plausibility as it possesses from a confusion between two quite distinct senses of ‘definition’. When we ask for the definition of anything we may be seeking to know simply the connotation of the term in ordinary educated usage. This is in the case of most terms easy enough to get provided we are within reach of a good dictionary; for dictionaries exist to provide just such information. Or we may on the other hand be seeking to know what the thing denoted by the term really and essentially is. That is an entirely different story. The kind of knowledge sought here is the kind of knowledge which is the ultimate objective of all science and all philosophy. In the case of religion it is manifest that we could not even come within sight of the requisite knowledge until we have given the closest consideration to a whole host of difficult questions; about e.g. the origins of religion in human experience; about the impact upon religion of social forces; about the manner of religion's development—if it does develop—and the apparent goal of that development; above all about the justification of religion's claim to express an unique insight into the ultimate nature of reality. Without any doubt the definition of religion in this second sense can come if at all only at the end of ‘extensive survey of and intense reflection upon the phenomena of religion’. But notice. Definition in this second sense presupposes definition in the first sense. The student of religion would not know what phenomena to survey and reflect upon in order to attain to a definition in the second sense if he did not already have a definition in the first sense to guide and control his enquiries. And in point of fact it is usually quite clear in the case of those writers on religion who insist that a definition is impossible at the beginning that they themselves tacitly assume one from the beginning.¹ The working of an implicit definition in their minds reveals itself throughout in their selection of phenomena to be studied as phenomena of religion.

Definition in the first sense then must be possible at the beginning. Equally certainly I think it is desirable at the beginning; for it is always a good thing to make clear to one's self and others as soon as convenient what one is going to talk about. And though it may turn out to be no simple task (for not in every case is even the ‘dictionary-type’ definition easy of achievement) it will at least be nothing like so troublesome as the effort to frame a satisfactory definition in the second sense. Anyone beginning the philosophical study of religion might well be disconcerted by the discovery that Leuba² has listed no less than forty-eight different definitions of religion offered by reputable philosophers and that Ducasse³ claims to have noted a further twenty-seven. There would indeed be good cause for alarm and despondency if this chaos of opinion related to the kind of phenomena that it is correct to classify under the heading of ‘religion’. In fact however these seventy-five definitions are almost without exception definitions in our second sense; definitions of what the thing denoted by the term really and essentially is. And in that sense great diversity of view is only to be expected since definition here involves a man's whole philosophy. The task of definition we propose for ourselves on the other hand is at least relatively simple. The requirements of this kind of definition are merely that it should include all of the characteristics that must be present and include no characteristic that may be absent where the term ‘religion’ is being correctly used in English speech or writing.

I have described our task as a relatively simple one. I must now however lay some stress upon the qualifying word ‘relatively’. Although we are looking for a dictionary-type definition of religion our problem is not to be solved just by looking up a good dictionary. Even good dictionaries we should find differ among themselves quite significantly in the definitions they propose and there is no way of escaping ultimate responsibility for our own preference. The trouble is that careless usages of the term ‘religion’ abound in common speech and to determine the ‘correct’ usage among the variations that are current is very far from being an easy matter. Indeed there must perhaps remain some element of arbitrariness in whatever definition is finally decided upon. Nevertheless I am not without hope that if we keep steadily before us as the criterion of correct usage the usage of educated persons when they are choosing their words with critical care the element of arbitrariness can be reduced to insignificant dimensions.

2. It will be helpful I think to begin by looking for a moment at two of the main sources of that loose and incorrect usage of the term ‘religion’ against which we have to be on our guard in our search for a definition.

The first is that ‘religion’ (if we may a little anticipate what will later be argued) is a term which stands for a complex phenomenon with a considerable number of defining characteristics. Wherever that is the case there is a temptation to use the term where only some of the defining characteristics are present provided that the characteristics present are of a striking sort. It is on that account no doubt that one sometimes hears of a man ‘making science his religion’—or perhaps even ‘making a religion of golf’—where all that there is in common with the traditional exemplifications of religion is an all-absorbing devotion to a given object. I fancy that in fact those who so speak more than half realise that they are using the word ‘religion’ loosely; they are anxious merely to emphasise the ‘religion-like’ character of the devotion that some people have to science or to golf. On the other hand I doubt if the same can be said of the growing tendency to speak even in ‘serious’ contexts of Communism as a ‘religion’ despite the fact that here too an ‘all-absorbing devotion’ seems to be about the only essential feature that Communism shares with what are usually called ‘religions’. I am bound to say that it seems to me a considerable disservice to the English language to use ‘religion’ in the flattened-out sense of an all-absorbing devotion to X irrespective of what X may be. Traditional usage has permitted the term ‘religion’ only where broadly speaking X denotes some supernatural being or beings believed to have real existence and to have rightful claims to human reverence. If the meaning of the term is to be gratuitously enfeebled so as to conform to the practice of those who call Communism a religion—or for that matter who call Humanism a religion—then it will become necessary to invent some new term for that highly distinctive and enormously important species of all-absorbing devotion which has for its object what is commonly meant by a ‘god’. But it would be rather an absurd situation if we found ourselves forced by such inept linguistic practices to invent a new name for something for which the old name ‘religion’ has been deemed good enough for centuries.

The type of linguistic looseness of which I have been complaining has the effect of extending the denotation of the word ‘religion’ beyond all reasonable limits. We may note now a second type of linguistic looseness that has just the opposite effect. The word ‘religion’ denotes a genus which has a wide variety of species and most people are familiar only with the species represented in their own native culture. There is a temptation in consequence to think of ‘religion’ in terms of this familiar species and to use the word in a way which implies that illegitimate restriction. Thus people often say ‘religion’ where they are clearly thinking only of Christianity or at the very most of the theistic type of religion. But that it is a loose and incorrect use even those who are guilty of it could as a rule be readily induced to acknowledge. Were the question directly put to them ‘Do you mean to imply that there are no religions except Christianity?’ they would almost certainly answer that of course this was not their intention and that in saying ‘religion’ when they meant ‘the Christian religion’ they were using inexact language.

These examples may serve to illustrate both certain common deviations from correct usage of the term religion and also the general manner of establishing what the correct usage is. Always the appeal must be to what people really mean by the term when they are using it not just casually but in a serious and considered way with a due recognition of the implications of their usage. Throughout what follows it is by this criterion that we shall be guided.

3. I begin with an assertion which though one cannot say it has never been questioned does not seem to me seriously disputable; viz. that ‘religion’ stands for something in which a certain state or attitude of mind is fundamental. That this state or attitude of mind generally finds expression for itself in specific forms of bodily behaviour which may for short be here summarised under the title ‘ritual’ no one would wish to deny. But to suppose—if indeed anyone really does suppose it—that mere ritual with no awareness of its meaning no awareness of any being in whose honour it is performed can ever amount to even an elementary phase of religion seems to me quite unwarranted. No doubt it may be the case (though I should myself accept it only with considerable reservations) that the mere performance of ritual dances and the like sometimes evokes in the performer for the first time the beliefs and emotions appropriate to religious worship. All that that implies however is that ritual performances may be a stimulus leading to religion not that the performances were prior to the emergence of the appropriate frame of mind themselves religious acts. It does not imply that is to say that we can have an early phase of religion in which the worshipper has no notion of what it is all about and then a later phase of the same thing in which the worshipper begins to appreciate the significance of what he is doing. On the contrary there is so complete a break of continuity between the mechanical performance of the ritual and the performance of it as a conscious mode of worship that to speak of the two as earlier and later phases of the same thing is totally inadmissible. One might as well say of the small boy who is taken to church and ‘goes through the motions’ without a clue to their meaning that he is though in an elementary way ‘practising religion’. But of course sensible people don't say anything so absurd. I see no reason to suppose that on reflection they would be any more inclined to say it in the parallel case of the savage. They would agree I think that at least some tincture of an informing spirit has got to be present in ritual if it is to be an expression of ‘religion’ in any legitimate sense of that term.

It is not to be denied of course that when people speak of ‘religion’ and still more when they speak of ‘a’ religion they have in mind a great deal more than just mental states and attitudes of a specified kind. They usually have in mind also the institutions and the systems of ritual observances and exercises in which the religious spirit commonly objectifies itself. But these are after all ‘objectifications’. Their source lies in the religious spirit which has a natural urge to seek expression for itself in outward forms. Moreover not only is ‘the spirit’ primary in as well as necessary to religion. I think we must also grant that strictly speaking it is self-sufficient for religion in the sense that we can have (though we rarely do have) what we should feel obliged to call ‘religion’ even where ritual expression of the ‘spirit’ is completely absent. No doubt we should be justly sceptical of the presence of a genuinely religious spirit in a man who gave no evidence of it whatever in the sphere of ‘action’; for it is hard to see how a genuinely religious spirit could fail to make itself manifest in the practical conduct of life. But ‘the practical conduct of life’ can scarcely be identified with what is normally understood by ‘ritual expression’; and provided we are satisfied that a man's conduct is not incompatible with the religious spirit we should I think deem it improper to deny the presence in him of that spirit and to withhold from him the right to be called ‘religious’ solely on the ground that he takes no part in those overt observances and ceremonies that constitute what is ordinarily meant by ‘ritual’.

4. I hope we may agree that religion is at least primarily a state of mind. But what kind of state of mind?

To a small but very small extent I have already anticipated the answer in speaking as though beliefs and emotions were certainly ingredient in it. But that assumption was not perhaps a very venturesome one. Only a few words seem necessary for its justification.

So far as emotion is concerned I cannot conceive that anyone nowadays would dream of describing as ‘religious’ a state of mind that did not include certain characteristic emotions. Intellectual beliefs even if they relate to the existence of supernatural beings would certainly not be deemed enough. Religion is not religion at all it would be said unless there is present an attitude of worship. And the attitude of worship is conspicuously an emotive attitude.

On the other hand the element of belief seems every bit as indispensable to the religious state. There is no need to be sure that belief should be articulated in any kind of formal creed. But the emotive attitude of worship is itself meaningless if there is not at the very least belief in the real existence of the being or beings towards whom the attitude is directed. One may remind oneself also that ‘faith’ is commonly accepted as a characteristic of the religious state and that faith whatever else it is does at least involve some kind of belief. Or again one might point to the general practice of using the terms ‘believer’ and ‘unbeliever’ as virtual synonyms for the religious and the irreligious man respectively. But it hardly seems necessary to defend further a thesis which is not likely to be denied save through some easily removable misapprehension.

We arrive at territory that admits of real debate however—indeed we come to the crux of the whole matter—when we set about trying to determine the common content of religious belief. There is in all religions belief in the real existence of the object worshipped. What common characteristics can be assigned to this ‘object’? The difficulties are obvious when we remember that the characteristics must be common to the objects of everything that we are prepared in considered judgment to describe as ‘religion’. Our conclusions must be applicable to the primitive religion of the Australian Bushmen (if we are prepared to call it ‘religion’) and to the highly sophisticated religion of Liberal Christianity (if we are prepared to call that ‘religion’). Remembering this it is self-evident that the ‘object’ of religious worship must not be identified with the ‘God’ of Theism and it is more than doubtful whether the term ‘God’ can properly be used of it at all.

Nevertheless there is I think a reasonably secure basis from which analysis may commence. In all religions the object that is believed in is an object that is worshipped and therefore presumably is deemed ‘worshipful’. We can say this at least then that religion involves belief in the existence of a being of supposedly worshipful character. Our immediate task must be to analyse this notion of the ‘worshipful’ to try to see what qualities a being must be taken to have in order to appear worthy of worship. In so far as we are successful the qualities elicited can fairly be said to be qualities belonging generically to the object of religious belief; qualities which in any religion its ‘object’ is deemed to have—though it goes without saying that there may not in actual fact be any object that possesses such qualities.

Let us ask then what is involved in the notion of the worshipful.

5. I think it will be granted that there must always be in the first place a certain element of mystery about the worshipful object. It must be something quite out of the ordinary run of nature animate or inanimate; something whose mode of being and functioning is not ‘intelligible’ to us in the way in which we suppose that the familiar processes in things and persons are ‘intelligible’. However much of power and excellence a being may be supposed to have still if that excellence and power are conceived as merely the same in kind as these qualities as they are manifested in the natural order or the human order one may admire or love or respect or fear their owner but an attitude of worship (I think we should agree) would be inappropriate. What is worshipful must always have I think something of supernatural quality.

The term ‘supernatural’ would of course be open to objection in this context if it were the case that recognition of a supernatural element presupposes clear grasp of the abstract distinction between ‘natural’ and ‘supernatural’; for it is plain that many primitive tribes which practise genuine worship cannot be supposed to have arrived at that level of intellectual sophistication. In point of fact however the recognition of particular instances of supernatural intervention would seem to presuppose nothing more than belief in a certain body of regularities of sequence and coexistence in one's environment. And so much one must presume to be present in any community however primitive if it is to live and survive at all. Its behaviour must be based upon definite habits of expectation corresponding to definite observed regularities in natural events. There is a ‘natural’ order for the savage no less than for his civilised brother though it is much more precariously founded and covers a much narrower range of ‘fact’. And when confronted by some striking deviation from his norm of the natural—some ‘extra-ordinary’ event—it is surely not beyond the capacity of the savage intelligence to frame the notion of some supernatural agency at work to account for its occurrence?

Moreover there is an abundance of evidence that extremely primitive tribes over many parts of the earth's surface do recognise the supernatural in particular instances. The languages of many are known to include words of an unmistakably supernatural import; e.g. such roughly equivalent terms as mana orenda wakanda and the like. Codrington (I quote here from Marett) has defined mana in its Polynesian use as ‘a force altogether different from physical powers which acts in all kinds of ways for good and evil’⁴ and again as ‘what works to effect everything which is beyond the ordinary powers of men outside the common processes of nature.’⁵ Recognition of the supernatural in this sense seems to be at least as primitive as anything we should ever want to describe as ‘religion’. Indeed it is almost certainly a great deal more primitive.

6. The worshipful then whatever else it may be is something mysterious or supernatural. But of course the converse is not true. What is mysterious or supernatural need not be worshipful. The savage is no more moved to adopt an attitude of worship towards every entity that he conceives to be endowed with mana than the modern man feels moved to do obeisance to poltergeists. The supernatural in and by itself will evoke the emotion of wonder; though a wonder admittedly that differs in quality from and is only analogous to the wonder that may be occasioned in ordinary life by striking and not immediately intelligible events which we nevertheless take for granted are in principle capable of a natural explanation. But this ‘supernatural’ wonder is only one major strand in the complex emotive attitude of worship. Worship implies adoration; a supernatural rapture of which the natural emotions of admiration joy and love afford but a pale transcript. And with this second ‘major strand’ in the attitude of worship we are led to recognise a second major component of the worshipful. For ‘adoration’ is evoked only by that which is felt to be endued with a transcendent worth or value. We may safely take it therefore that in order for a supernatural being to qualify as a fit object of worship this further condition at least must be fulfilled namely that it be felt to be endued with transcendent value.

The presence of this ‘value’ element in the notion of the worshipful and consequently in the generic object of religion is I think beyond serious doubt. And it is a not uninstructive illustration of the grave confusion which is apt to result in philosophical and anthropological works on religion from the failure to begin with a considered definition of the term ‘religion’ that many writers have not hesitated to include within the category of religion cults in which this ‘value’ element is totally wanting. A notable example is the so-called ‘apotropaic religions’. The whole object of these cults is to mollify by gifts and sacrifices certain supposed supernatural beings who are conceived to be inherently hostile to man; to be in fact ‘demons’ rather than ‘gods’. No doubt devotees of these cults go through elaborate rituals that are externally not easily distinguishable from those of religious worship. But the inner spirit is about as different as well could be. It is one thing to make gifts and sacrifices to a supernatural being in a mood of reverence and love that delights to do him honour. It is quite another thing to make them simply with a view to buying from him immunity from malicious visitations. Adhering to our criterion of correct usage viz. the duly considered linguistic usage of competent persons I cannot think that the term ‘religion’ is properly applied to organised bribery on a vast scale even if it be supernatural powers to whom the bribes are offered.

The case is little better for regarding as primitive ‘religions’ those cults which conceive their supernatural beings as favourably disposed or at worst neutral towards the worshipper and which ply them with gifts and sacrifices in the hope that this will encourage them to exercise their stupendous powers in beneficent interventions. In so far as this and this alone is the motive that inspires ritual acts we come not a step nearer to religion proper. Currying favour with gods is no more a ‘religious’ act than is bribery of demons. Indeed it seems to me that both of these types of ritual show a much closer approximation to magic than to religion. In them as in magic certain ceremonies are performed whose object is to ‘work upon’ supernatural forces in the service of natural human purposes. There are of course significant differences also. One main difference is that whereas the ritual of the cults aims at making use of supernatural forces through the agency of supernatural beings who are believed to have such forces at their disposal the ritual of magic is often designed to work directly upon the supernatural forces with no thought of supernatural beings as intermediary. In that respect since religious worship is normally and perhaps always directed to supernatural beings the cults may be said to come a little closer to religion than magic does. But if they come a little closer to religion they must still be accounted leagues away from it unless we are prepared to deny a proposition that to me at least seems indubitable viz. that religion is fundamentally a matter not of overt acts but of the inner spirit.

It is perhaps desirable to add in order to avert possible misunderstanding that what I have said is not intended to deny that the attitudes of propitiation and cajolement characteristic of these cults may linger on even after religion proper has emerged and may be discernible ingredients in the state of mind that informs a ritual which is in other respects authentically religious. Obviously they can and often do so linger. There is no incompatibility—if there is also no organic connection—between desiring to do honour to a worshipful God and desiring to enlist His good offices on one's own mundane behalf. Both motives can be present in the same ritual performance; and I think it must be admitted that in the actual practice of most religions both motives very often are present. But the fact that the attitudes of propitiation and cajolement are not incompatible with religion does not of course make them in any sense ‘religious’ attitudes. The externality of their relationship to religion is strikingly indicated by the acknowledged fact that as religions develop in purity and realise more adequately their true nature as religion these attitudes tend to lose their prominence and ultimately to drop out altogether. Ritual acts once dominated by such attitudes may survive in the same or almost the same external form (for there are few more conservative institutions than the organised religious body); but when they do a new interpretation is generally found for them which has nothing to do with attempts to propitiate or cajole.

Something must now be said about a phenomenon which though no doubt as rare as it is strange may well seem at first sight to cast doubt upon our contention that a value element is intrinsic to the ‘worshipful’. What about Devil-worship? Certain very peculiar ceremonies to which the name ‘Devil-worship’—or alternatively ‘Diabolism’ or ‘Satanism’—is commonly given do undeniably take place even at the present day. Now is not the essence of Devil-worship the doing obeisance to the very principle of Evil? And how is this to be reconciled with the claim that nothing can be deemed worshipful that is not also deemed good or valuable?

The difficulty has I think a fairly simple solution; but it requires that two distinct points be made.

In the first place it is extremely questionable whether much if any of what is called Devil-worship is really Devil-worship or would continue to be so called if people had clearer views about the nature of the thing they were labelling and were exercising proper care in their choice of language. There is in all cases admittedly a ritual closely resembling in externals the ritual of genuine worship and it is to the supposed ‘Devil’ that it is directed. But the frame of mind informing the ritual is probably much more often than not that of the pseudo-religious cults to which we have lately been referring. Certain stereotyped acts believed pleasing to the Devil are performed; not however out of esteem for the Prince of Darkness but in order to induce him in return to exert his supernatural power in the interests of the ‘worshippers’ securing for them the dazzling array of earthly (and earthy) prizes which it is supposed the Devil can and will bestow upon those who formally commit themselves to him. The ‘pacts with the Devil’ of which we hear so much in medieval times seem to have been of this general character. But where such is the case it is clear that so-called ‘Devil-worship’ is at bottom a mere bargaining transaction aimed at the purchase of diabolical favours; and ‘worship’ is surely quite the wrong name to give to it. It seems to me that we have here another example of inept linguistic usage where a word is applied to something which resembles in conspicuous respects that which the word primarily denotes the important difference which makes the application of the word in this context grossly improper being carelessly overlooked.

Obviously however this answer will not suffice of itself so long as we admit the possibility of any case of genuine Devil-worship any case in which the Devil is honoured as it were ‘for his own sake’; and I do not think we can by any means rule out this possibility. The solution in that case must turn I think upon the distinction between objective and subjective value; between value in the sense of that which is truly valuable and value in the sense of that which someone deems to be valuable. It would certainly be absurd if we had claimed that nothing can be deemed ‘worshipful’ except that which is truly valuable. All that in fact we have claimed is that nothing can be deemed ‘worshipful’ unless it be thought by the worshipper to have value. Now the honouring of the Devil ‘for his own sake’ implies the attribution to him by the worshipper of ‘subjective’ value. And modern developments in abnormal psychology have gone far to make intelligible to us how this can come about. The content of the subjectively good or valuable is based upon nothing more exalted than actual human desire and there seems no assignable limit to the possible perversions of human desires. It is in principle possible for the human soul to be so perverted that a man will take his greatest satisfaction to He in a programme for living that conflicts at every major point with the traditional values of Christendom—in a programme that exalts hatred and cruelty and lust and pride and avarice. If such a person also believes (as many people still do) in the actual existence of the Devil as a supernatural being who is the very prototype of all ‘fiendish’ qualities the stage is perfectly set for ‘Devil-worship’. But then Devil-worship so understood is plainly not a case of the worship of evil as evil. It is the worship of evil (or of what most people consider evil) in the belief that it is supremely good. Hence there is nothing here that is inconsistent with the analysis of the worshipful which we have been proposing. ‘Devil-worship’ so understood is a genuine enough species of religion; a very bad species it is true but the genus ‘religion’ I fear we must admit does include some very bad species.

7. I turn now to the third and last of what I take to be the essential components of the worshipful. It is one which fortunately calls for less extensive treatment than the value component since in this case we are little if at all troubled by instances of ostensible ‘worship’ which are incompatible with it. The worshipful being is always a being deemed by the worshipper to be not only supernatural and not only endued with transcendent worth or value but also to have transcendent power.

This is perhaps most easily seen by reminding ourselves of the fact (as I take it everyone will admit it to be) that no less indispensable to the attitude of genuine worship than wonder and rapture and in close integration with these is the emotion of awe. Now the objective correlate of awe is power power that is at once mysterious and overwhelming. It is clearly not enough that the power be merely mysterious (in the sense of supernatural). The supernatural as such is no doubt always credited with a mysterious power which excites our dread and wonder but this mysterious power need not be on any majestic scale; and where it is not (as in the case of the ordinary kind of ‘spook’) it does not inspire awe. Ingredient in the emotion of awe is a certain deep humility and self-abasement before the sheer magnitude of the mysterious power that confronts us; and it is pretty evident that though we may be terrified we do not feel ourselves humbled or abased in the presence of the ordinary kind of spook. If the supernatural object is to inspire awe it must then I think be accredited with a power not merely mysterious but also overwhelming; a ‘transcendent’ power if that term may be accepted as combining in its connotation the force of both adjectives.

A more debatable question but one over which we need not linger too long here is whether all awe is religious awe. The answer to that question turns as I see it on the character of the value aspect which is conceived to belong to the awe-inspiring object. McDougall seems to me right to insist that in the emotion of awe there is always present in addition to fear and wonder an element of ‘admiration’; and in that case there must presumably be recognition of value of some sort in the awe-inspiring object as the correlate of the admiration. But granted that admiration is a necessary constituent of awe it need be no more I suggest than admiration for the supernatural being in respect of such value as consists in its possession of overwhelming power—for power as such especially if it be very great does normally evoke admiration—and this is compatible with the actual disevaluation of the supernatural being in respect of other of its characteristics and as a whole. Now if this be so all that one is committed to holding in the way of ‘value’ concerning a being that inspires awe is that value is ascribed to some particular feature of it; not necessarily that value belongs to the being itself in its integrity as an individual. And it will then follow that not all awe is religious awe. For in the religious attitude it seems certain that the worshipful being is taken as admirable in itself not merely in respect of some feature or features which may be offset (as in the case perhaps of the traditional Satan) by other features that are far from admirable.

8. Mystery Power Value—in all essentials Otto's Mysterium Tremendum et Fascinans—these then seem to me to be the basic characteristics of the worshipful. Of course their degree of emphasis both absolute and relatively to one another varies enormously in different concrete religious states and likewise in different historical religions. Mystery and Power e.g. tend to be paramount in primitive religions. The first impact of the supernatural upon the primitive mind is by way of power manifestations. Mana as we saw is little if anything more than ‘supernatural power’. Even when religion proper emerges and transcendent value is integrated with transcendent power in a supernatural being who for the first time becomes an object of worship the power element as a rule remains for long the dominant one. In the later developments of religion the three elements appear in an almost infinite variety of emphases. But always I think all three are present in some degree.

Our purpose in undertaking an analysis of the ‘worshipful’ it will be remembered was to enable us to supply filling for the common content of religious beliefs. Since in all religion the object that is believed in is an object of worship and therefore (presumably) deemed worshipful it seemed to follow that whatever characteristics could be shown to belong indispensably to the worshipful would thereby be shown also to be indispensable components of the object of religious belief. We have done what we could to elicit these characteristics of the worshipful and in their light we infer that in all that is properly called ‘religion’ there is belief in a supernatural being or in supernatural beings of transcendent worth and power. Axe there any further qualities that can legitimately be ascribed to the ‘generic’ object of religious belief? For my own part I can detect none. I think in fact we are now in a position to complete our initial task in this course by offering a definition of the term ‘religion’. But I shall defer its actual formulation until the next lecture in order that I may have the opportunity of answering immediately certain objections which it may very easily though I think mistakenly provoke.