The Validity of Belief in a Personal God

C. D. Broad


Published in the Hibbert Journal 24 (1925): 32-48. Reprinted in C. D. Broad, Religion, Philosophy and Psychical Research (Routledge & Keagan Paul, 1953).

    In order to discuss the question whether there is any ground for believing in the existence of a personal God it is necessary to begin by defining our terms. For the word 'personal' and the word 'God' are both highly ambiguous. I will begin with the word 'personal'.

    The natural interpretation of the phrase 'a personal God' would be 'a God who is a person'. But, if this were the only meaning that could be attached to the phrase, we should have to say that orthodox Christians deny the existence of a personal God. For the Christian God is the Trinity; and the Trinity is not a person, though its members are persons. Now it would be extremely inconvenient to define the phrase 'personal God' in such a way that we should have to hold that all orthodox Christians deny the existence of a personal God. And, as we have seen, this inconvenient result would follow if we defined a 'personal God' to mean 'a God who is a person'. We must therefore adopt a some what wider definition of 'personal'. Now we notice that, whilst the Trinity is denied to be a person, it is asserted to be a complex unity composed of three intimately related constituents, each of which is a person. And I think that we should deny that a man believed in a personal God unless he believed that God either is a person or is a complex whole composed of nothing but interrelated persons. I therefore suggest that the phrase 'a personal God' means 'a God which either is a person or is a whole composed of nothing but interrelated persons'. This definition is certainly wide enough, whilst the first suggested definition was certainly too narrow. It might perhaps be objected that the proposed definition is now too wide. Would any and every God which is composed of nothing but interrelated persons be counted as a personal God? Or must the relations be of a specially intimate kind before we can apply the adjective 'personal' to a whole composed of nothing but persons? It is admitted that, according to the Christian doctrine of the Trinity, the relations between the constituent persons are extremely intimate; so much so that there is a constant danger of making statements about the Trinity which are true only of its constituents, and of making statements about its constituents which are true only of the Trinity as a whole. But I think that this question really arises rather under the definition of 'God' than under the definition of 'personal'. It is quite certain that we should not apply the name 'God' to any and every whole composed of interrelated persons; we should apply this name only if the relations were peculiarly intimate. I shall assume, therefore, that any whole composed of nothing but persons may be called 'personal' provided that the relations between the constituent persons are intimate enough for this whole to be called a 'God'.

    We have not, however, finished with the definition of the adjective 'personal'. We have said that 'a personal God' means 'a God which either is a person or is a whole composed of nothing but interrelated persons'. But what do we mean by a 'person'? I do not know that we can define the term; but, by considering examples of what we should call 'persons' and by contrasting them with examples of what we should refuse to call 'persons', we can see pretty well what is involved in being a person. We call a sane grown man a 'person'. We refuse to call any inanimate object, such as a chair, a 'person'. We also refuse to call a cat or a dog or a horse a 'person', though we admit that they have feelings, impulses, instincts, habits, etc. And I think that it would be felt to be a strained and metaphorical use of language to call a very young baby a 'person'. If we reflect on these examples I think we shall see that we apply the name 'person' literally to a substance if, and only if, it fulfils the following conditions:

  1. It must think, feel, will, etc.
  2. Its various contemporary states must have that peculiar kind of unity which we express by saying that they 'together make up a single total state of mind'.
  3. Its successive total states must have that peculiar kind of unity with each other which we express by saying that they are 'so many different stages in the history of a single mind'.
  4. These two kinds of unity must be recognized by itself, and not only by some external observer, i.e. it must not only be in fact a mind, but must also know that it is a mind. And this knowledge must be, in part at least, immediate and not merely inferential; though its knowledge of many details about itself may, of course, be inferential and not immediate.
It may be the case that every substance which has the kind and degree of internal unity necessary for being a mind also knows immediately that it is a mind. Still, it is one thing to have this kind and degree of unity, and it is another thing to know immediately that one has it; and it seems logically possible that the former might happen without the latter. In that case we should not, I think, refer to this mind as a 'person'. It is therefore necessary explicitly to introduce this fourth condition.

    If we accept this as an adequate description of 'being a person' there are certain further remarks to be made.

    (1) There are, presumably, different degrees of personality. These differences may arise in two different ways, which must be distinguished in theory even if in fact they be causally connected so that variations in one respect causally determine variations in the other respect.

  1. A mind will be more fully personal the more completely its contemporary states are united with each other to form a single total state, and the more completely its successive total states are united with each other to form the history of a single mind. In every human mind there are conflicting desires, inconsistent beliefs, and contemporary mental processes which have very little connexion with each other. And the history of every human mind is broken by gaps of dreamless sleep, fainting fits, drunkenness, and so on.
  2. A mind will be more fully personal the more fully and immediately it recognizes such unity as it in fact possesses. We have all forgotten a great many states which have in fact been experienced by us, and we cannot recall them at will. And, on the other hand, we are liable to 'remember' events that never happened, and to believe falsely that they formed parts of our mental history.
Now, since these defects are present in different degrees in different human minds, though they are present in some degree in all human minds, we can form the conception of a mind which is much more completely a person than any human being is; just as we can form the conception of a perfect gas or a frictionless fluid from our experiences of more or less imperfect gases and more or less viscous fluids. An ideal person would be a mind which is as fully unified as possible; which has no inconsistent beliefs, or conflicting desires, or mutually indifferent mental processes; which never sleeps or faints; and so on. And it must be as fully and immediately aware of this unity as possible. It must not forget anything that has belonged to it, though it is not necessary to suppose that it is always actually remembering everything that has ever happened to it. It is enough to suppose that it could remember any of these events whenever it wanted to. There seems to be no logical objection to the concept of an ideal person, if this be all that is meant by the phrase.

    (2) There are certain judgments which we make only about persons, and certain emotions which we feel only towards what we take to be persons. We should not literally ascribe moral goodness or badness to anything which we did not believe to be a person. No one seriously talks of a virtuous baby, or regards a cat as being morally responsible for its actions. And no one can strictly feel the emotions of love or gratitude to anything which he does not at the time regard as a person. It is true, I think, that a man may quite literally love his cat or dog, though he would admit, if questioned, that it is not a person. But an intelligent domestic animal probably has, in fact, the rudiments of personality, and, whether it has or not, its master almost certainly treats it in practice as a person whatever his theoretical beliefs on the subject may be. Again, it is certainly possible to feel emotions which are analogous to love and to gratitude towards certain wholes composed of interrelated persons, though we should admit that these wholes are not themselves persons. E.g. there is an emotion which we call 'love' for a public school, a college, or a country. And there is an emotion which we call 'gratitude' towards these institutions for the benefits which we believe them to have bestowed on us. But, in the first place, it is plain that we tend in practice to personify such a group of persons, although we know that it is not really a person. We tend, e.g. to substitute for Trinity College, which is a Society and not a person, a kind of idealized man who combines all the best qualities of all the nicest Trinity men that we have known. And, if we literally love certain actual Trinity men, we shall tend to feel an analogous emotion at the thought of this idealized Trinity man who represents Trinity College to us. Moreover, I think it is plain that, although some of the emotions which we feel towards certain groups of interrelated persons are analogous to love and to gratitude, they are not strictly the same emotions as love and gratitude. Love, in the strict sense, can be felt only towards something which we believe to be capable of loving us in return; it is always accompanied by a desire to be loved in return, and in the absence of such a response it tends at length to fade away. But we know perfectly well that a college or a public school cannot literally love us, though some of its members may do so. Yet this does not prevent us from feeling for it the emotion which I have described. Hence this emotion cannot be the same as love, in the strict sense. I conclude, then, that we cannot strictly love anything unless we believe it to be a person at the time when we are feeling the emotion.

    I have now, I hope, made clear what is meant by 'being a person', and have stated some important additional facts about this characteristic. The next point to be considered is what is meant by being a 'God'. I think it is quite certain that the word 'God' is extremely ambiguous, and that it has commonly been used in at least three different, though connected, senses. I distinguish these as the popular sense, the theological sense, and the philosophical sense. In the popular sense of the word 'God' a God is ipso facto a person. This person is supposed to be analogous to a human being, but to be much more powerful. It is supposed to be able to do things of a different kind from those which human beings can do, and I think that it is generally conceived as not subject to death and as exercising an important influence on the weal or woe of human beings. This is all that is involved in the notion of a God in the popular sense. It is not supposed to be necessarily unique; it is not supposed to be infinitely powerful or perfectly wise, but merely to be a great deal more powerful and a great deal wiser than any living human being. And it is not supposed of necessity to have created men or to have created the material world, nor is the continued existence of nature and of men supposed necessarily to be dependent on the continued support of a God in this sense. Lastly, a God, in the popular sense, need not be morally superior to the best human beings, though he must be wiser than the wisest and stronger than the strongest human beings. Jehovah and Apollo are Gods in the popular sense; but Jehovah inculcated a high moral tone by precept rather than by example; and Apollo, in view of his relations with Cassandra and with Hyacinthus, might have had difficulty in obtaining the College testimonial for deacon's orders, which has never been held to require a superhuman level of moral achievement.

    The word 'God', in the theological sense, has in one respect a wider meaning, and in other respects a narrower meaning, than the same word when used in the popular sense. A God, in the theological sense, need not be a person. According to orthodox Christian theology nothing can strictly be called 'God' except the Trinity as a whole. And the Trinity is certainly not a person. It is true that the Athanasian Creed says that the Father is God, and the Son is God, and the Holy Ghost is God; but it immediately adds that nevertheless there are not three Gods, but one God. If these statements are to be rendered consistent it is plain that the word 'God' must be used in different senses in the two. Nor is there the least difficulty in seeing what these senses are. The creed means that there is only one being that can with strict theological correctness be called 'God', viz. the Trinity as a whole. But each of the three persons can be called 'God' in a looser sense, because they are divine persons and essential constituents of the Trinity, which is God in the strict sense. Thus we might say loosely: 'The King is the sovereign of England, and the House of Lords is the sovereign of England, and the House of Commons is the sovereign of England.' But we should immediately add, in order to ward off possible errors, 'Of course, strictly speaking, there is only one sovereign of England, viz. the whole composed of King, Lords, and Commons in their proper constitutional relations to each other'. In the popular sense of 'God' each person of the Trinity is a God, and the Trinity as a whole is not a God; but, in the theological sense of the word, the persons are not Gods, whilst the whole composed of them is a God, and is the only God that there is.

    The theological sense of the word 'God' is thus wider than the popular sense, in so far as the former can be applied either to a person or to certain wholes composed of interrelated persons, whilst the latter can be applied only to a person. In all other respects, however, the theological conception of God is narrower and more rigid than the popular conception.

  1. Theologians push all the attributes of God to extremes. A God, in the theological sense, must be not merely very wise and very strong; he must be perfectly wise, and capable of doing anything which does not involve some internal logical inconsistency.
  2. It is an essential part of the theological conception of God that he shall be morally perfect.
  3. It is also part of the theological conception of God that he shall be unique. By this I mean that theologians are not content to hold that there happens to be only one thing answering to the definition of 'God', just as there happens to be only one thing answering to the description of 'the brother of Romulus'. They hold that, from the nature of the case, there could only be one God, just as, from the nature of the case, there could only be one individual answering to the description 'the most virtuous undergraduate in Trinity'. I think that this is one reason why theologians refuse to call the persons of the Trinity 'Gods', and confine the name 'God' to the Trinity as a whole. For, in all other respects but uniqueness, the persons of the Trinity would seem to be 'Gods' in the strict theological sense.
  4. Finally, it is, I think, part of the theological conception of God that he cannot be identified with the universe. There has to be some asymmetrical relation between God and the rest of the universe, so that there is a sense in which we can say that the latter is existentially dependent on the former whilst the former is not existentially dependent on the latter.

    I will make a few explanatory comments on the theological conception of God before passing to the philosophical conception.

(1) I do not know how far the statements of theologians about the omniscience, omnipotence, and moral perfection of God are to be taken literally. It may be that this pushing of God's attributes to extremes is only intended as a compliment; and that when God is said to be perfectly wise, and good, and powerful, these phrases are to be regarded as analogous to 'Your Serene Transparency' when applied to German princes or 'His Most Religious Majesty' when applied to Charles II. Persons who used the latter phrases plainly did not intend to deny that German princes are opaque to light, or that Charles II was sometimes inclined to be a little careless about the higher spiritual values. And it may be that theologians do not intend their statements about God's attributes to be interpreted too literally. On that hypothesis the theological conception of God may not really differ so much from the popular conception as it seems to do.

(2) We must clearly understand that not any and every group of interrelated persons would be regarded as a God even in the theological sense. It is necessary that the persons should be of a certain kind, and that their relations should have a certain high degree of intimacy. I think that the component persons must be such that each would be a God in the popular sense. And I think that the relations must be so intimate that none of the persons could exist apart from each other and outside the whole which they together form. We can see the necessity of both these conditions by taking cases where one is fulfilled and the other fails. The society of Olympus was a whole composed of interrelated persons, each of which was a God in the popular sense. But no one ever thought of regarding this whole as a God. And the reason is that the relations were not intimate enough. Zeus could have existed without Hera, and Hera could have existed without Zeus. Again, on Dr. McTaggart's view, the universe is a whole composed of persons so intimately related that none could have existed without the rest and apart from this whole. But no one would call the universe, as conceived by McTaggart, a 'God'. For its components are ourselves and other persons like us. And we are not Gods.

    I pass now to the philosophical sense of the word 'God'. This is very much wider in either the theological or the popular sense of the word. A 'God', in the philosophical sense, need not be a person or a whole composed of nothing but interrelated persons. It therefore need not be wise or good, for these epithets apply only to persons. The name 'God' has been applied by certain philosophers to the Universe as a whole. Thus Hegel calls the Absolute 'God', and Spinoza talks of 'God or Nature' as synonymous terms like Augustus and Octavius. I think, however, that even in philosophy the name 'God' would be applied to the Universe only on the supposition that the Universe has a much more intimate internal unity than it appears to have at first sight, and that this unity is of a special kind. I think that all philosophers who have asserted the existence of God have held one of three views about the internal structure of the Universe:

(1) That there is a certain part of the Universe which is not existentially dependent on any thing else, and that all the rest of the Universe is existentially dependent on this part of it. This substance is then called 'God', whatever its other characteristics may be. This is the doctrine which is known as Deism. It is held by those philosophers who talk of God as 'the great First Cause'.

(2) That there are certain characteristics of the Universe from which all its other characteristics necessarily follow. In that case the name 'God' will often be applied by philosophers to the Universe in virtue of its having this peculiar internal structure. Thus Spinoza distinguishes between Natura Naturans and Natura Naturata. Natura Naturans is the Universe, regarded as having certain fundamental characteristics from which all the rest follow. And Natura Naturata is the Universe, regarded as having all the characteristics which follow from these fundamental ones and as having no others. The Universe in its completeness is thus Natura Naturans and Natura Naturata; and Spinoza calls it 'God' in virtue of its having this kind of internal structure. This doctrine is one form of Pantheism.

(3) That many of the features which seem to characterize the Universe or parts of it do not really belong to it, but are distorted and partly illusory appearances of characteristics which really do belong to it; in particular, that the Universe is in reality purely mental (i.e. that it is a mind or a society of minds), and that matter, space, and motion are distorted appearances of this mind and its states or of these minds and their mutual relations. The name 'God' is then often applied by philosophers to the Universe as it really is on this view, as distinct from the Universe as it appears to be. This, I suppose, is why Hegel called the Universe 'God'.

    I think we may say that no philosopher asserts the existence of God unless he holds one of these three views about the nature of the Universe. On the other hand, many philosophers who do hold one of these three views would refuse to assert the existence of God, on the ground that the word has much more definite implications in theology and in ordinary life and that the use of it in the philosophic sense is misleading. E.g., the Universe, as Dr. McTaggart believed it to be, is a God in the third philosophic sense of the term. But McTaggart always refused to call it 'God' and blamed Hegel for doing so, on the ground that the phrase 'The Absolute' completely conveys his meaning whilst the word 'God' inevitably has associations and arouses emotions which are not justified by what he believed to be the facts. Here I agree with him. I think that we ought to confine the word 'God' to the theological and the popular senses of it; and that, unless we have reason to believe in the existence of a God or Gods in one of these senses, we ought not to say that we believe in the existence of God at all. Now, in these senses of the word, a God is necessarily a personal God. It is either a divine person, or it is a whole composed of nothing but divine persons so intimately related that none of them could exist apart from the rest and outside this whole. And I have defined what I mean by a 'person' and what I mean by 'divine'. The question then is: What reason, if any, have we to believe in the existence of divine persons? For it is plain that we can have no reason to believe in wholes composed of nothing but divine persons related in certain ways unless we have reason to believe in divine persons. And we might have reason to believe in the existence of divine persons whilst we had no means of deciding whether there was one or a dozen, and no means of deciding whether they stood in such and such relations to each other or not.

    A man who believes in the existence of a divine person might try to justify his belief in one of three ways:

  1. He might claim to know directly that such a being exists; or
  2. he might claim to be able to prove the existence of such a being, or to make it very probable, by argument; or
  3. he might believe it on the authority of others.
I will consider these three alleged grounds in turn.

    (1) A claim to direct knowledge of God's existence might take two different forms:

  1. A man might find the proposition 'God exists' self-evident, as most men find the proposition 2+2=4. Or
  2. he might claim to know that God exists because he has in some supersensible way perceived God; just as most people claim to know that their chairs and tables exist because they have perceived these objects with their senses.
It is quite certain that most people who believe in the existence of God do not pretend that their belief can be justified in either of these ways. Very few people would claim that they find the proposition that God exists self-evident; and still fewer people would claim to have themselves perceived God. But such claims have been made; and there is no way of positively refuting them. But there are reasons which ought to make the claimants themselves extremely doubtful, and which ought to make us still more doubtful, about accepting their claims at their face value. It is notorious that propositions may seem self-evident although they are not true. For propositions which are inconsistent with each other, and which therefore cannot both be true, have seemed to be self-evident to different people. During the war it seemed self-evident to most Englishmen that Germans are morally inferior to the English; and it seemed equally self-evident to most Germans that Englishmen are morally inferior to Germans. One of these propositions must have been false, and probably both of them were. It may be said that in this case both parties were blinded by patriotic emotion; but it might equally be suggested that those persons who find the existence of God self-evident are blinded by religious emotion. If it appears self-evident to some people that there is a perfectly wise, good, and powerful being, it appears equally self-evident to many other people that the existence of such a being is inconsistent with the amount and kind of evil which exists in the world. Lastly, we know what sort of propositions have appeared to be self-evident to nearly everyone and have never been in any danger of being refuted. They are always propositions which assert that one quality is necessarily accompanied by a certain other quality; they are never propositions which assert that there exists an object which has such and such qualities. Now the proposition that God exists is of the latter kind, and not of the former; it is therefore most unlikely that it is really self-evident in the sense which it is self-evident that 2+2=4.

    Let us now consider the claim to know directly that God exists because one has perceived him in some supersensible way. Perception may roughly be defined as being in direct cognitive contact with an existent something which manifests certain qualities to the percipient, and is instinctively regarded by him as a part or an appearance of a more extended and more enduring object which has certain other qualities that are not manifested to the percipient at the moment. E.g., when I say that I see a penny, I am in direct cognitive contact with something which manifests the qualities of brownness and approximately circular shape; and I instinctively regard this as a part or an appearance of something which is permanent, which has an inside as well as an outside, and which has qualities like hardness and coldness that are not at present being manifested to me. If this belief be mistaken, I am not perceiving what would commonly be called a 'penny'. Now it is notorious that in ordinary sense-perception we are often deluded and sometimes wildly deluded. A simple example is mistaking a mere mirror-image for a physical object, and a still more striking example is perceiving snakes or pink rats when one is suffering from delirium tremens. It is quite certain, then, that there are delusive sense-perceptions. Now, in the case of sense-perception there are several tests which we can use to tell whether a perception is delusive or not. We can check one sense by another, e.g., sight by touch. We can appeal to the testimony of others and find out whether they see anything that corresponds to what we see. Finally, we can make inferences from what we think we perceive, and find whether they are verified. We can say: 'If there are really rats running about my bed my dog will be excited, bread and cheese will disappear, and so on.' And then we can see whether anything of the kind happens. Now it does not seem to be possible to test the alleged supersensible perception which some people claim to have of God by any of these means. Very few people have had the experiences at all; they are very difficult to describe, and therefore to compare; and it is very hard to point to any verifiable consequences which would follow if, and only if, these perceptions were not delusive. And, so far as I can see, nothing comparable to supporting the testimony of one sense by that of another is here possible. This does not, of course, prove that such supersensible perceptions are delusive; but it does show that we have no means of telling whether they are or are not. And, as we already know that many perceptions are delusive, this is a serious matter. As Hobbes says: 'When a man tells me that God spoke to him in a dream all that I can be sure of is that he dreamed that God spoke to him.'

    Even if we waive this objection, and take at their face value the statements of people who say that they have perceived God, they give no support whatever to the existence of a single perfectly wise, good, and powerful being, on whom all the rest of the Universe depends. They would tend rather to support the view that there is a bewildering variety of Gods in the popular sense, many of whom possess the oddest personal peculiarities.

    (2) I pass now to arguments for the existence of God. These may be divided into deductive and inductive arguments. There are two of the former. One professes to prove from the definition of God that such a being must exist. This argument, if it were valid, would have the advantage of proving the existence of an unique individual possessed of all possible perfections, i.e. of God, in the theological sense. But it is universally admitted by philosophers and theologians that the argument is logically fallacious. It is called the Ontological Argument.

    The second deductive argument starts from the premiss that no thing or event in nature exists of intrinsic necessity. Such necessity as we find within nature is purely relative and hypothetical. We can say that, given A, B necessarily follows. But we cannot say that A's existence or B's existence is intrinsically necessary if A and B be things or events in nature. It is then argued that, since nature as a whole has this contingent character, its existence must depend on something else whose existence is intrinsically necessary. This something is called 'God'. The argument is known as the Cosmological Argument. It is not so obviously fallacious as the Ontological Argument, and it has been accepted by some very able theologians and philosophers, such as St. Thomas Aquinas and Locke. Nevertheless, I agree with Kant and Hume that it is fallacious. Fortunately it is not necessary for me to prove this here, because the argument is irrelevant for our present purpose. For it is certain that, even if it be valid, it has no tendency to prove the existence of a personal God. At best it would prove the existence of God only in one of the three philosophical senses of that term, and not in the theological or the popular sense.

    We may therefore dismiss the deductive arguments and consider the inductive ones. These start with certain admitted facts about nature and man, and argue back to the existence of God as the hypothesis which best explains these facts. Of course, the conclusions of such arguments could never be more than highly probable. But I do not think that this is a serious objection. We could quite reasonably say that the existence of God was 'proved' if it could be rendered as probable as the existence of Julius Caesar. Such arguments may be classified according to their premisses.

  1. They may start from certain facts about inorganic nature and living organisms.
  2. They may start from the fact that nature contains minds which are capable of distinguishing good and evil and of guiding their actions by ideals.
  3. They may start from the fact that certain minds have, in addition, specifically religious emotions and other experiences.
A complete inductive argument would presumably use all these facts as premisses.

    (a) The first set of facts forms the basis of the famous Design Argument. This has been so thoroughly discussed by Hume in his Dialogues on Natural Religion that there is little left to say about it. I will content myself with the following remarks:

(i) We must distinguish between the adaptation of inorganic nature to life in general, and the peculiarities of organisms as such. Let us begin with the former. It is certain that the condition of inorganic nature on the earth is, and has long been, extremely well adapted to the existence and growth of living organisms. So far as we know, the conditions under which organisms can exist are very peculiar, so that it is antecedently improbable that they should be fulfilled. Hence it is argued that they must have been deliberately brought about by a mind which wanted organisms to exist and flourish. This, I think, is a fallacious argument. It seems certain that the fulfilment of these conditions is really very local and temporary. They are probably not fulfilled now in the greater part of the Universe; they certainly were not fulfilled formerly on the earth, and they almost certainly will cease to be fulfilled there in the distant future. Now it is not antecedently improbable that even very peculiar conditions shall be fulfilled for a comparatively short time in a comparatively small region of a universe which is indefinitely extended in both Space and Time.

(ii) The position about organisms themselves is as follows. An organism is an extremely intricate system which appears, even to the most superficial view, to be extraordinarily well adapted to preserve itself in face of varying conditions and to produce things like itself. And the more minutely we examine it the more accurately true do we find this to be. Now the only other things that we know of which have the least analogy to this are artificial machines. We know that these have been designed by minds, and we have not the least reason to think that they could have existed unless there had been minds which designed them and arranged their parts in such a way as to carry out these designs. Of course organisms are now produced by other organisms, just as typewriters are produced by other machines. But in the history of any artificial machine we eventually come back to a mind which had designs and arranged matter in such a way as to carry them out. We may assume, by analogy, that if we went far enough back in the history of organisms we should come on a mind which designed them and arranged matter accordingly. This mind was certainly not human, and it must certainly have been of superhuman wisdom and power to produce such results. It may therefore fairly be called 'God'.

    I may say at once that I consider this to be an extremely strong argument if we grant two assumptions which are commonly made. The first is that organisms originated from inorganic matter. The second is that an organism really is nothing but a complicated machine, i.e. that its characteristic behaviour is wholly due to the peculiar arrangement of its parts, and is not due to entirely new properties of matter which first appear at the organic level. If we reject either of these assumptions the argument loses much of its force. If there have always been organisms of some kind, and no organism has ever originated from inorganic matter, there is no need to postulate a designing mind even though organisms be nothing but machines. And if organisms be not merely machines, there is no need to postulate a designing mind even though organisms did originate out of inorganic matter. Now, I do not see the least reason to believe that the characteristic behaviour of organisms can be wholly explained by the peculiar arrangement of their parts and the laws and properties of inorganic matter. Hence the argument for the existence of a designing mind from the peculiarities of organisms does not convince me, though I think it ought to have great weight with a purely mechanistic biologist.

    (iii) Even if we accept the argument it will not prove the existence of God, in the theological sense. In the first place, it would prove only that a designing mind had existed in the past, not that it does exist now. It is quite compatible with this argument that God should have died long ago, or that he should have turned his attention to other parts of the Universe. Again, so far from proving the existence of a being on whom the rest of the Universe is existentially dependent, it negatives this supposition. It proves the existence of a superhuman workman faced with material whose properties he has to recognize and make use of, and not of a creative being. Thirdly, there is nothing in the facts to suggest that there is only one such being. And lastly, there is nothing to suggest that he is morally perfect. We must grant him superhuman skill and power, but the actual state of the world forces us to limit either his power or his goodness, or his wisdom, or all three. So, at the very best, the argument would prove only that at some time in the remote past there had been one or more Gods in the popular sense of the word.

    (b) I will now consider the argument for the existence of God from the existence of minds like ours which can look before and after, make judgments of good and evil, and guide their conduct by them. It may be admitted that we cannot conceive of any natural process by which minds could have arisen spontaneously from mere matter. So it has been suggested that we must postulate the existence of God to account for the facts. But, in the first place, there is no reason to accept the alleged facts; and secondly, the hypothesis of a God would provide no explanation of them.

  1. It is quite possible that there have always been minds, and that no mind has ever originated from anything but another mind by a natural process. In that case the hypothesis of God is needless for the present purpose.
  2. If we make the hypothesis we have explained absolutely nothing. We are still obliged to suppose that there have always been minds, though not always non-divine minds. And the production of non-divine minds from mere matter remains just as unintelligible whether we say that it happens spontaneously or that it is miraculously accomplished by God.

    The fact is that the Argument from Design and the argument which I have just been discussing illustrate an important general principle. If you start with a sufficiently narrow and inadequate view of nature you will have to postulate a God to get you out of the difficulties in which it lands you. E.g., if you insist that living organisms are mere machines, you have to postulate God to construct them out of unorganized matter. And if you insist that nature is fundamentally material and that mind is a kind of afterthought, you have to postulate God to account for the origin of mind, though, as I have pointed out, the hypothesis does not here really help you. But why should you start with these narrow and inadequate views of nature? They have no trace of self-evidence and they conflict with the observable facts in every direction. And, unless you make this mistake at the outset, I do not think you will be able to find any inductive proof of the existence of God.

    (c) Finally, I come to the argument for the existence of God which is based on the occurrence of specifically mystical and religious experiences. I am prepared to admit that such experiences occur among people of different races and social traditions, and that they have occurred at all periods of history. I am prepared to admit that, although the experiences have differed considerably at different times and places, and although the interpretations which have been put on them have differed still more, there are probably certain characteristics which are common to all of them and which suffice to distinguish them from all other kinds of experience. In view of this I think it more likely than not that in religious and mystical experience men come into contact with some Reality or some aspect of Reality which they do not come into contact with in any other way.

    But I do not think that there is any good reason to suppose that this Reality which manifests itself to certain men in religious and mystical experiences is personal. I think that we are inclined to believe this because we are most familiar with the religious experiences of Western Europeans and of Jews, most of whom have put this interpretation upon them. We do not know, or we forget, that the mystics and religious teachers of the Far East on the whole definitely reject this interpretation. And we are inclined to forget that certain Europeans, such as Plotinus and Spinoza, who have had these experiences also reject this interpretation of them.

    I think on the whole, then, that there is no inductive argument which makes it at all highly probable that there is a personal God.

    (3) It only remains to consider whether it is reasonable to believe in the existence of a personal God on the authority of other men. We all believe many propositions on authority, and in many cases it would be most irrational not to do so. It is rational to believe a proposition on authority if one of two conditions is fulfilled.

(i) If experts agree that it can be proved, but the argument is too difficult or unfamiliar for me to follow it myself.

(ii) If persons whom I know to be competent and trustworthy tell me that they have perceived certain things which I have not perceived myself.

I accept many propositions in mathematics on the authority of Professor Hardy, who tells me that they can be proved; and I accept many propositions in physics on the authority of Professor Rutherford, whom I know to be a trustworthy person and a highly skilled experimenter and observer. But neither of these conditions is fulfilled in the case of the proposition that there exists a personal God. There is no consensus of experts about the alleged proofs, and I can see for myself that these arguments are fallacious. And I have tried to show that the claims of certain persons to have perceived God in some supersensible way are to be regarded with grave suspicion even if we accept their bona fides. Hence it would be irrational for me to believe in the existence of a personal God on the authority of others.

    To conclude. Whether there be in fact a personal God or not, it seems to me that we have no good reason to believe in the existence of such a being. I think that there are such grave difficulties in the notion of a God in the theological sense that there are strong reasons against believing that such a being exists. These objections do not apply to the notion of Gods in the popular sense. For all I can see there may be dozens of such Gods; and the only reason against being a polytheist is that there is no reason for being one.


Transcribed into hypertext by Andrew Chrucky, Aug. 20, 1998.