AN ATHEIST'S VALUES
2.83. There is no god or afterlife
Good evaluation has to be made in the light of good judgement as to what the facts are, what our situation is. Among the questions of fact on which it is important to have a right judgement are the questions whether there is a god and whether there is a life after death. My answer to each of these questions is 'No'. Ought I to give you the considerations on which I base these answers? That is a hard choice, because, on the one hand, it seems wrong to offer no reasons on these immensely important questions, but, on the other hand, all imaginable reasons have already been exhaustively discussed by both sides for centuries, and we are perfectly sick of them. I have concluded that I had better give something on this topic, but make it merely a brief indication.
Whether there is a god, and whether there is a life after death, are questions of existence, of what there is or what happens. They are not questions of mathematics, nor questions of law or necessary connexion. They are like the questions whether there is a tribe of people on the Amazon who eat their parents, and whether there are flying saucers. They cannot be answered by mathematical proofs, but only by travellers' reports, or at any rate by something far more like a traveller's report than a mathematical proof.
Mathematical proofs can show non-existence but they cannot show existence. (They can show that there 'exists' a prime number greater than the millionth power of ten. But that is another sense of 'exists'. It is not the natural existence in time or place that I am speaking of.) For mathematics can only show that certain conceptions are selfcontradictory and others are not. If a conception is selfcontradictory, as is the round square, that shows that nothing of the sort exists. But if a conception is selfconsistent, as is the round area, that does not show whether anything of the sort exists. Thus mathematics may be able to show that certain conceptions of god or immortality are selfcontradictory and therefore nothing of the kind exists. But, if there is any selfconsistent conception of god or immortality, mathematics will not show in the least whether anything fulfilling this conception exists.
In the Christian religion, though perhaps not in any other, we frequently find a conception of god that is selfcontradictory and therefore corresponds to nothing. That is the conception formed by the following three propositions taken together:
1. God is all-powerful.
2. God is all-benevolent.
3. There is much misery in the world.
A god who was all-powerful but left much misery in the world would not be all-benevolent. An all-benevolent god in a world containing much misery would not be an all-powerful god. A world containing a god who was both all-powerful and all-benevolent would contain no misery.
Here, then, we have a mathematical proof bearing on a common religious doctrine. Anyone who is confident that he frequently comes across misery in the world may conclude with equal confidence that there is no such thing as an all-powerful and all-benevolent god. And this mathematically disposes of official Christianity, as has long been known (though people have fought endlessly against it by means of the doctrine of free will).
That, however, is as far as mathematics takes us. (And even there we are relying also on our judgement and experience to get the conclusion that misery does occur.) When we turn to consider all the selfconsistent conceptions of gods and devils and angels that have been invented or might be invented, mathematics gives us no help, and we are confronted with the truth that anything whatever might exist so long as it is selfconsistent.
In order to decide what does exist, among all the things that might exist because they are selfconsistent, we have to use our judgement on the reports and experiences available to us. It is a question of using the judgement, of deciding judiciously between the two contradictories; and more than half the battle consists in getting oneself into a sincerely judicious frame of mind and out of the injudicious determination to believe the affirmative answer if one possibly can. The rest of it consists in learning by experience what kinds of inference are reliable in questions of existence. For, whereas in mathematics we learn by logic what kinds of inference are valid, in questions of fact we learn only by experience what kinds of inference are ludicrous. Only experience has taught us that, for example, when a person looks at a police line-up and says confidently that 'this is the man who attacked me', she is sometimes wrong.
What is good evidence for the presence or existence of a person, and have we such evidence for the existence of a superhuman person?
Suppose you wish to decide what kinds of mammal exist. You collect your memories of mammals you have seen. You collect zoologists' descriptions of mammals they have seen. You collect travellers' accounts. And so on. You try to judge which of these you should accept and which you should reject. 'I am sure I remember seeing a panther, but do I really remember seeing something called an "ocelot", or was that a dream? I am sure the reports of okapis in the Congo are true, but can I trust the reports of abominable snowmen in the Himalayas? Among travellers I judge that X is nearly always both honest and correct, Y is always honest but often incorrect, and Z is plain dishonest.' All this is clearly fallible, however copious our information and however much we compare and counter-check. But nothing better can possibly be found. And there are plenty of cases where the judicious man properly comes to a confident judgement as to whether a given thing exists.
Suppose now that instead of mammals I wish to make a list of all the gods there are. Here again I can only rely on my own memories, the reports of travellers, the accounts of scientists, and so on. The scientists in this case will not be zoologists but theologians, prophets, mediums, psychical researchers. Again I have to judge which of their reports to accept, which of them are both honest and aware of the dangers of error, and so on.
Many of their reports have the remarkable feature that they tell us that the god they have experienced is the only god who exists. There is nothing like this in the case of mammals. No one comes back from the Congo and tells us that the kind of mammal he saw there is the only kind of mammal there is. These reports must be unjustified because experience cannot tell us that there is only one of a certain sort in the whole universe. If you carefully examine the whole of this room you will perhaps be justified in declaring that there is one and only one Scotsman in this room. But, since you cannot examine the whole of the universe, you cannot justifiably tell us that there is one and only one god in the universe.
Many of these reports tell us of a god who is infinite or perfect. They must be unjustified, because one cannot experience infinity or perfection. However long a time I may have experienced I have not experienced infinite time. However much goodness I have experienced from a person, I have not experienced infinite goodness or perfection at his hands. It is only a finite god that we could have evidence for. Have we good evidence for such?
All of these reports have the remarkable feature that they tell us that the gods are experienced and yet not perceived. One may, it is said, sometimes perceive manifestations of the gods, visions, miracles, and, of course, images. One may also perceive a man and infer from the miracles he does that he is also a god. But one cannot perceive the god directly with eyes or ears. And yet one experiences him. Experience without perception is, of course, usually a mark of the subjective; what I experience without the aid of perception is primarily my own inner life, my thoughts and imaginations and moods and so on. But we are told that there is also experience without perception of at least one kind of objective reality, namely god or the gods.
This universal feature of the reports makes them all incredible. That is not because a claim to have experience of other persons without perception of them is always to be declared false. It is because such claims require confirmation by subsequent perception, and there is no evidential value in such a claim if it is never confirmed by perception. Even if experience without perception were good evidence, there would be nothing to show that the various believers were all experiencing the same god. A hundred people all having religious experience at the same time in the same church might each be in communication with a different god, since no god ever presents himself to their bodily perception for them to see whether they are talking about the same person.
What about inferring the existence of a god, as opposed to experiencing him? Sometimes we properly and correctly infer the existence of a person in a house, although we do not perceive anyone there. Can we similarly infer the existence of a god in the universe, although we do not perceive one? No, we cannot. The fundamental defect of all such inferences is that they require perceptual confirmation and they never get it. Inferences that there is someone in the house are sometimes confirmed by subsequently seeing him. If they never were confirmed by subsequent perception they would never be proper. But inferences from something or other to the conclusion that there is a god in the universe are never confirmed by subsequently seeing him here.
In the early days of Christianity the favourite phenomenon from which to infer the existence of a god was miracles. But the inference is worthless. It is worthless to monotheists, because all religions have their miracles, so that if miracles were good evidence for one god they would be good evidence for all the gods. It is worthless to every judicious reasoner. For a miracle is an event which astounds us by seeming to contradict the laws of nature. But our being astounded proves nothing but our ignorance. And if an event really contradicts what we thought was a law of nature, that just shows that we were wrong about the laws of nature. We very often are wrong about the laws of nature. They are not to be known by instinct, but only by the endless labours of science.
Another phenomenon from which people infer the existence of an unperceived god is the multitude of convinced and sincere testifiers. 'How could all those intelligent and honest men be mistaken?' This inference is also worthless. If we took the existence of a multitude of convinced and sincere testifiers as good evidence for a belief, we should have to believe not in one religion but in all the conflicting religions that have obtained; we should have to believe both that the sun goes round the earth and that the earth goes round the sun; we should have to believe both that the world ended at several different dates in the past and also that it is still going on; for it is a characteristic of Christianity that from its beginning down to at earliest 1900 it has produced groups of earnest persons who were convinced that the world was going to end within ten years. In 1846, for example, the Irvingites were convinced that the world was going to end in 1847. They were sober and intelligent persons living in London. They included the solicitor to Rugby School, and the banker who founded the chair of political economy in Oxford University. No one is a good reasoner about matters of fact until he has realized that it is very common for a gross falsehood to be firmly and sincerely believed by a great number of superior persons, and that therefore there is nothing in the argument from the consensus of mankind.
Another phenomenon from which many have inferred that there is an unperceived god is the designedness of the world. Since the world is designed, they have thought, it must have a designer. Here it is not the inference that is injudicious but the premiss. The world does not appear to be designed. Little bits of it appear to be designed from time to time, but as a whole it strongly appears not to be designed. When lightning kills a mean bully and leaves his victim unharmed beside him, that looks like somebody's design; but the complete picture of all the lightning-damage in a given country in a given century looks very undesigned.
Recently it has become common to infer that there is a god from the phenomenon of conscience. Human beings have consciences, uninferred convictions that certain kinds of action are wrong, and they feel guilty if they do an action belonging to one of these kinds. This phenomenon, it is thought, could only be caused by a god. But to lay it down that a certain phenomenon can only be caused by a certain something is to claim to know the laws of cause and effect by instinct and without inquiry. No one knows by instinct what the laws of nature are; they have to be laboriously hunted and proved by experiment. Since experiment is largely forbidden on human beings, the laws of human nature are particularly difficult to find; but the phenomenon of 'wolf children', that is human children brought up by wolves without contact with human beings, suggests, so far as I have read the reports, that conscience is caused by association with other human beings.
All the evidence and argument offered for the existence of a god, is, I judge, injudicious, in so far as we mean by a 'god' a superhuman person, and do not use the word as a mere question-mark to indicate an unknown something or other. I have shown briefly what seems to me the badness of some of the common reasons given for believing that there is a god. The eternal persistence with which people bring forward these bad arguments is probably caused by our need for the comfort and security of having a perfectly reliable father. This need often seduces us into very bad methods of argument. For example, some people will be tempted to sneer at the brevity and simplicity of these remarks of mine, saying loftily that the question of religion is not to be settled in five minutes. The question of religion will, indeed, probably never be settled. That is, there will probably never be universal agreement about it. But arguments about it are better when they are short than when they are long. Long arguments should always be suspected of being designed to intimidate or hypnotize rather than to explain. And we should reject the common assumption that no one has a right to assert atheism unless he asserts it at great length with great learning. If, however, anyone wishes for a longer statement of my position, there are plenty in existence. I will mention only David Hume's essay on miracles and his Dialogue concerning Natural Religion.
And now as to life after death. Endless life after death would be a form of infinity, and for actual infinities there can be no good evidence. But is there good evidence for any life after death?
A great deal of the evidence offered for life after death depends on the doctrine that there is a god, and is valueless because that doctrine is false. However, some people have believed that there is no god and yet is an afterlife; and spiritualists and mediums offer us evidence of survival independent of the question whether there is a god. I have never attended a spiritualist performance, but I have read some of their reports. I think I can safely say that no afterlife of any difference in quality from this life has been reported, and no afterlife of any interest. The stuff they offer us is deadly dull. However, it might be true for all that. Homer believed that there is an afterlife and it is deadly dull.
I judge that it is false, and that all these reports of messages from the dead are false. (By which I do not mean that they are all frauds or lies. No doubt many of them are sincere. It is desirable to repeat from time to time that a falsehood is not the same as a lie. A falsehood can be sincerely uttered, and a lie can be true.) These reports have the pattern of inventions, the vagueness, the poverty, the similarity to each other, the comfortingness, and the insistence on circumstances that make criticism difficult, such as darkness and reverence.
My main reason for thinking there is no afterlife is that it seems immensely probable that everything we know as life depends on there being a living body. All that part of life which consists in the activity of a living body selfevidently depends on there being a living body, for example, eating, tasting, running, laughing, kissing. The life that does not selfevidently depend on there being a living body is the interior life of the mind, including reasoning, imagining, dreaming, and other activities and experiences. But it seems quite clear that we have overwhelming physiological evidence that this mental life, too, depends on the activity of a living body, and ceases when there is no longer a good brain with good blood flowing through it in the right quantity.
For this reason I am confident that there is no life after death. However, no one who believes that there is a life after death will be disappointed, because, if there is no life after death, he will not be alive to be disappointed. Only those who believe that there is no life after death can possibly find that experience disproves their answer to this question.
So much for the questions whether there is an afterlife and whether there is a god, on our answers to which a great deal of our evaluations and actions should depend. One thing, however, which should not depend on these answers is our evaluation of religion as we know it now and have known it in the past. Religion as it has so far appeared is upon the whole a bad thing whether or not there is a god or an afterlife, because it is a fundamental rejection of the ideals of truth and reason.
2.84. Religion and reasons for morality
It is often held that religion is the only basis for ethics, that only religion can show a good reason why we should obey any moral laws, and only religion can make us in fact obey them. People holding this view tend to regard the assertion of atheism as an attempt to undermine the morality of the nation; hence they try to prevent atheism from being recommended on the B.B.C., in which they are almost completely successful.
It is false that religion is the only basis for ethics. If we had to choose between the two sweeping assertions 'religion is the only basis for ethics' and 'religion is not a basis for ethics at all', the latter would be preferable. Religion is in fact not a proper basis for ethics at all.
What is a basis for ethics? The phrase refers either to reasons for the moral law, or to causes which make people keep the moral law. It means either that the only good reasons to justify the moral law are religious reasons, or that the only stimulus which effectively causes people to obey the moral law is religion, or both. Let us consider each of these in turn, beginning with reasons. Does religion provide good reasons for the moral law, and does it provide the only good reasons for it?
Let us divide reasons into entailing and non-entailing reasons. An entailing reason for a moral law is one that entails it. A non-entailing reason for a moral law is one that does not entail it but nevertheless seems to somebody to be a good reason for it.
It has been made perfectly clear that there can be no entailing reason for a moral law except another moral law. Probably the first person to point this out unmistakably was H. A. Prichard in his article in Mind for 1911, 'Does Moral Philosophy rest on a Mistake?' (reprinted in his Moral Obligation). A sentence beginning 'thou shalt' or 'thou shalt not', or containing the words 'ought' or 'ought not' or 'right' or 'wrong', can be entailed only by a sentence also containing one of these expressions. For example, the sentence 'thou shalt not kill' is not entailed by any of the following: 'there is a god who hates killing', 'there is a god who punishes killing with eternal fire', and 'there is a god who is our father and commands us not to kill'. None of these is an entailing reason for the law that 'thou shalt not kill'. The following, however, is an entailing reason for this law: God commands us not to kill and thou shalt do whatever God commands.' This is an entailing reason because it contains a 'thou shalt', and therefore is itself a moral law. A moral law can be entailed by a sentence about a god only if that sentence is itself a moral law. It cannot be entailed by a sentence which merely informs us that there is a god, and what his commands are. The moral laws as a whole are not and never will be entailed by anything. In other words, there is a good sense in which ethics has no basis and cannot have a basis. There is a good sense in which there is no such thing as 'the foundations of morality'. Mr. Hare stated this point very clearly, with special reference to Christian thought, in Philosophy for 1950, p. 376.
There could be a man who said: 'There is one and only one moral law, and it is that we ought to do whatever God commands.' If this man believed it possible to discover what his god did command he could then go on to discover it and do it; and he would then be acting morally because he would be acting under a moral law, a sentence with an 'ought' in it. There could be another man who said: 'I am determined to devote myself utterly to God, and to do whatever he commands.' If he also believed that it was possible to find out what his god commanded, he, like the former man, could go on to discover it and do it. But unlike the former man he would not be acting morally, because he would be acting from a principle which contained no 'ought' and no 'thou shalt', but merely said 'I will'. His procedure would be like that of a man who gives up trying to do right and tries instead to please his mistress. He would simply have taken a god for his mistress. To devote oneself utterly to a divine being, and decide to do everything he ordered, would not be to base ethics, but to abandon it and substitute another way of living. We get an entailing religious basis for ethics only if we adopt the moral law that we ought to do whatever our god commands.
Now let us turn to non-entailing reasons for moral laws. There could be, and no doubt there is, a logically minded theist who agrees that religion can provide no entailing reason for the moral laws, but says that it can provide a good non-entailing reason for them, and nothing else can do that. The only good non-entailing reason for the moral laws, he may say, is that there is a god who commands us to obey them and will burn us eternally if we do not. Contrast this with my view that the only good reason for a moral law is that its reign in a society substantially decreases misery in that society.
If it really were probable that we should burn eternally, or not burn eternally, according as we disobeyed or obeyed a certain set of moral laws, that would, indeed, be an excellent reason for obeying them. But, while it would be an excellent reason for obeying them, it would be a poor reason for respecting them, or regarding them as worthy of reverence and awe. On the contrary they, and the god who imposed them on us in this unbelievably brutal way, could only be regarded as beneath contempt. And in fact it is very improbable that hell-fire exists, or that an afterlife at all exists. And we have only very poor evidence indeed that there is a god of any kind, and yet enormously poorer evidence as to what his commands and threats are if he does exist. And it is always possible that, if he exists and ordains certain laws, these laws are in conflict with those that would do most to lessen misery on earth. Because of these facts the religious reason for obeying the moral laws is a very poor reason. On the other hand, we can hope to get good evidence whether obeying a certain set of laws does or does not substantially decrease misery in a given society; and hence, if we take the decrease of misery as the criterion of a right moral law, we are likely to be able to form a reasonably probable opinion as to what the right moral laws are.
People talk disapprovingly of 'moral anarchy'. This phrase 'moral anarchy' is always a muddle, and should be given up. If it refers to the fact that people often disobey the moral laws, it is a bad way of referring to it. Correct phrases for the purpose are 'immorality', 'wickedness', 'disobedience to the moral law', and many others. If it refers to the fact that people disagree to some extent as to what moral rules should be obeyed, that is not something for disapproval; two people can disagree without either of them being reprehensible. If it refers to the fact that no government on earth is trying to enforce legally all the recognized moral rules, it refers to something that is to be approved, not disapproved. If it refers to the fact that many people have given up deducing moral laws from statements about gods, it refers to something good, since moral laws are not in truth entailed by anything but moral laws. If it refers to the fact that a great many people today, in deciding what moral laws to adopt, do not accept the authority of any church or priest or god, but freely criticize them all, it refers to something highly approvable, to that coming of age of man's reason which is an encouraging feature of our time. The truly moral world is essentially anarchical, in that to be a really moral agent a man must judge and choose for himself. A world in which people follow authority and obey an 'archy' for their moral opinions and choices is an imperfectly moral world. The disapprovable thing is moral 'archy', not moral anarchy. Great harm is done to morality by its authoritarian connexions, by the notion that the parson is the depository of it, by the habit of turning to the parson when the question is what is morally right. The average parson is a worse judge of right and wrong than the average layman, and that is simply because he takes his moral laws on authority. Never go to church to learn how to behave unless the sermon is preached by a layman.
So much for religion as a reason for obeying moral laws.
2.85. Religion and causes of morality
Now let us turn from reasons to causes and consider the proposition that 'the only cause which in fact makes people obey moral laws is their believing some religious proposition'.
The first and most important point to make about this proposition is that, whether it is true or false, to use it as an argument in favour of religious belief is a disgraceful thing to do. To do that is to commit the pragmatic dishonesty of arguing that a creed is true because it is useful that people should believe it. I know that this argument is used extremely frequently, and in the most respected quarters. Nevertheless, it is selfevidently null both in logical effectiveness and in common decency. That it would be very convenient if people believed the Christian creed is nothing whatever to do with the question whether that creed is true. And to recommend a proposition on the ground of the convenience of having it believed is just as dishonest as telling a lie. The improper appeal to expediency, which religious people are fond of imputing to their opponents, is in truth made by the religious themselves enormously more often than by the irreligious. To say in a solemn and intimidating tone of voice that, in so far as atheism comes to be believed, murder and thrift and rape increase in the world, is part of an immense number of Christian arguments; and it is thoroughly immoral.
If it were true that a general belief in atheism caused a high frequency of murder and theft and rape, while a general belief in theism caused a low frequency of them, and at the same time atheism were true and theism were false, we should be compelled to make a painful choice. There would be nothing for it but either to preach the true doctrine and see murder and theft and rape increase in consequence, or to keep down the rate of crime by preaching the false doctrine. That would be a hateful dilemma indeed. Yet there is no doubt which course we ought to take if we ever were obliged to make this choice. We ought to preach the true doctrine and endure the consequent increase in crime. To preach a false doctrine, or to preach a doctrine without considering whether it is false or true, is base and beneath human dignity. It is an abandonment of the great good of truth, and a treachery towards our fellows worse than exposing them to a greater risk of crime.
That is the first and most important point about the proposition that 'the only cause which in fact makes people obey moral laws is their believing some religious proposition'. However true it may be, to use it as an argument in favour of religious belief is logically null and morally indecent.
The second aspect of this proposition which I wish to consider is how one would test it. To what branch of inquiry does it belong? Either to history or to anthropology or to sociology. It is a sociological sort of proposition, because it offers a general rule about a causal connexion between human beliefs and human behaviour. So we must ask: What is good method in sociology? How can one reach a justifiable judgement on a general proposition asserting that a certain kind of belief tends to be accompanied by a certain kind of behaviour?
To obtain a justifiable judgement on such a proposition is a matter of statistics and mass observation, and hence a very large undertaking. Sociologists and anthropologists have scarcely ever ventured, yet, to make any such sweeping statement about man in general. They have confined themselves almost wholly to saying things about particular groups of men. I think they have been wise to do so, and the apologists for theism are unwise when they offer us unrestricted generalizations about the effects of theological beliefs on human behaviour.
Even when we have abandoned talk about human behaviour in general, and confined ourselves to a particular society, the inquiry is still laborious in procedure and doubtful in result. Consider, for example, how you would test the proposition that 'from 1700 to 1900 moral and law-abiding behaviour declined in England because religious belief declined'. This involves three different questions: (1) First, did moral and law-abiding behaviour decline in England during this period? (2) Second, did religious belief decline in England during this period? (3) Third, if so, were the two causally connected? We can obtain a probable answer to only one of these three questions, the one about religious belief. The quantity of churchgoing affords a good rough measure of the quantity of religious belief; and we can find out something about that for the period in question.
We have no good way of answering the question whether moral and law-abiding behaviour declined in England during this period. Perhaps we know the population in 1700 well enough to compare with the population in 1900, which we do know. But police figures are not available to give any reliable comparison on the legal side; and on the moral side there is no record and no way of coming at the proportion of moral offences committed. If a centenarian tells us that people were much more moral in his young days, or much less so, that is very poor evidence indeed in view of the deceptiveness of memory, the bias of the old, and the fragmentariness of any one person's experience. Also the centenarian is most unlikely to be sociologically judicious and able to tell good evidence from bad in such matters.
Another difficulty in any such inquiry is as follows. When we speak of the morality or immorality of Englishmen in 1700, do we mean the extent to which they obeyed their own moral laws, or the extent to which they obeyed the moral laws in which we believe now? Our decision on this point may make a big difference, for the reigning body of moral laws may have changed somewhat between then and now. For example, perhaps there was a double standard of sexual morality then, and perhaps there is not now. If so, James Boswell, when he got an illegitimate child upon a girl and then rebuked her for her behaviour, was perhaps obeying his own moral laws though he was certainly disobeying ours. It is possible that by the code of 1700 Englishmen of 1700 were more moral than those of 1900, while at the same time by the code of 1900 Englishmen of 1700 less moral than those of 1900. If we compare our fidelity to our code with the fidelity of an earlier society to our code, we are likely to come out nearly always with the conclusion that we are more moral than they were. But, if we try to compare our fidelity to our code with their fidelity to their code, the conclusion may be different; or there may be no conclusion because we cannot be sure what their code was.
For my part I think that we in England today both have a better code of morals, and obey the reigning code of morals better, than Englishmen did in the eighteenth century. But I also think that this is a precarious judgement, worth mentioning only as a counterweight to the low estimates of present morals which we often hear from bishops and writers to the press. In any case I have no doubt that much of what is said from pulpits today about the immorality of the present generation may properly be described as irresponsible abuse. It is a bad habit that Christians have contracted from certain verses of the gospels.
But suppose we had reliable answers to both the first and the second question. That is, suppose we could say with confidence just how religious belief had changed in these two centuries and just how morality had changed. We should still have no reliable means of knowing whether they were causally connected. A single case of concomitant variation is no evidence of causal connexion in matters so delicate and complicated as religious belief and moral fidelity. Many other possible causes would require to be excluded; and this would require similar knowledge of the variations in moral fidelity and other possible causes at other times and places. I judge that no reliable conclusion is to be obtained here.
I judge also that, when people think they can reach a reliable conclusion here, this is often because they are improperly thinking of the obvious proposition, which needs no statistics for its support, that, 'if the only cause of a man's acting morally is his fear of hell-fire, he will cease to act morally when he ceases to believe in hell-fire'. That is true by the mere meaning of 'only cause'. But it has no application to our question, because fear of hell-fire is not the only cause of moral action. We know that, because we are acquainted with at least one man who acts morally but disbelieves in hell-fire.
There certainly are some people who obey moral laws only from fear of a god or that god's hell-fire; and it certainly does happen sometimes that a person of this sort ceases to believe that there is a god and thereupon ceases to obey moral laws. There is truth in the story of a Papist priest saying to a pair of well-behaved atheists: 'I can't understand you boys; if I didn't believe in God I should be having a high old time.' Nothing that I have said denies the possibility of this kind of human nature. My view is that indefinitely many kinds of human nature are possible, and hence we must not make sweeping statements to the effect that only one kind of human nature is possible and all others are impossible. The proposition that 'the only cause which in fact makes people obey moral laws is their believing some religious proposition' is in effect an assertion that it is impossible for there to be a kind of human nature which obeys moral laws from a non-religious cause. Such assertions of impossibility are injudicious. We do not know all that human nature may do, and we do know that it may do a great many different things, and react in a great many different ways.
The false assumption, that nothing but religious belief will ever make people obey the moral laws, is widespread in the occidental world and it has harmful effects. It leads people into all kinds of intellectual dishonesties in their frantic efforts to save religious belief for the sake of saving morality. It leads many of those who reject religious belief into immorality, because their idea of obeying the moral laws was bound up with their idea of a god. It leads everyone who holds it to neglect the attempt to base morality on something not religious, and to neglect to teach their children a morality independent of religious opinion, that will remain in being however often they may change from theism to atheism or back again.
In other words, what may well be called the capture of morality by religion, which was an achievement of the Jewish religion in particular has turned out harmful for the world now that religion is declining; and it is an urgent task to set morality free again and give it an independent position, dependent only on its serving the earthly good of man, which is his only good. We are not doing all we could in that matter. For example, atheistic parents usually do not try to find atheistic schools for their children. That is a wrong attitude. At theistic schools children are taught that the reason for morality is hell-fire or God's command. This false reason is likely to collapse in later life; and if it does so it leaves the person with no reason for morality unless he thinks one out for himself. Parents ought to demand a school which teaches that the reason for morality is the alleviation of suffering. If you are rich, you could do a great service by helping to maintain a school where morality was taught in this honest way. You would be doing a great deal to make our lives more sincere and our morality more enlightened.
Christianity in a sense captured common decency and made itself the guardian thereof. Nevertheless, it did not itself act in a decent way. According to the Christian historian Professor Herbert Butterfield, 'the Christian church began a cruel policy of persecution from the earliest moment when it was in a position ... to do so; while at the other end of the story both Catholic and Protestant churches fought to the last point of cruelty not merely to maintain their persecuting power -- but fought a separate war for each separate weapon of persecution that was being taken from them' (The Listener, xli. 582-3). The Roman Church is still officially against freedom. A book published in 1948 with the imprimatur of a Roman archbishop says that a Roman State 'could not permit
There may be another harmful aspect to the capture of morality by religion. It may be that the gods are inevitably immoral, that is, that any god that ever has been or will be conceived acts immorally in some ways. We can, of course, all easily see that other people's gods are immoral. Zeus behaved immorally; so did Moloch; and so on. I suggest that you will find that your own god is immoral too, if you can bring yourself to apply to him the moral standards that you apply to men. Surely it is immoral to condemn people to everlasting fire, or to blame them for the sins of their ancestors. Surely it is immoral to be omnipotent and yet allow the vast and continuing miseries of living things, or to demand that people believe without regard to evidence. All such conventional phrases as 'God's ways are inscrutable' are in use partly because they help to prevent us from seeing the immorality of the god we have conceived.
Now it seems likely that this immorality of all the gods so far invented is not an accident, but a necessary consequence of the religious impulse. It is probably connected with the element of worship in religion. One cannot abase oneself before a perfectly moral person, because a perfectly moral person treats one as an equal and as having a right to one's way of life.
If the gods are inevitably immoral, the idea that they are the guardians of morality is bound to do some harm. The rules of morality will be distorted from time to time by the bad behaviour of their guardians. This, then, seems to be a way in which the capture of morality by religion is always unfortunate.
It sometimes happens, however, that the captive eats the life out of his captor. The capture of morality by religion ends, sometimes, in religion's becoming nothing but a picturesque or mythical form of morality. This was the position of Matthew Arnold in Literature and Dogma, where he wrote that 'religion is ... morality touched by emotion' (p. 47), that 'the object of religion is conduct' (p. 144), and that God is the eternal Power, not ourselves, that makes for righteousness (p. 303). In this stage people make no independent effort to communicate with their god or find out his nature; they infer his nature from their own moral views. Plato knows that 'God likes a joke' (Cra. 406 c) simply by knowing that it is good to joke sometimes. Benjamin Franklin knows that 'God helps them that helps themselves', although this contradicts the New Testament, because as a sturdy Yankee trader he knows that you ought to help yourself. His famous phrase is simply a mythical form of the moral law 'help yourself'. To many people nowadays god is merely the myth to which they attach their valuations. Religion, which often used to be entirely independent of morality, has become in these people nothing but morality itself in mythical guise.
If a man's religion just is his morality, then certainly he loses his morality if he loses his religion. But in maintaining that we know of no causal connexion between religion and morality, I have, of course, been speaking of religion as the worship of the gods, a combined creed and attitude which has no necessary connexion with morality at all.
The general conditions which tend to make a man behave morally are two. (1) First, his circumstances must not make it very difficult for him to obey the moral law even if he wants to. He must not be exposed to strong temptation to break it. For example, he must not be starving and penniless among people who have food but will not give him any. (2) Second, the moral law must be generally obeyed, respected, and recommended, by those among whom he lives. Departures from it must be generally met with disapproval or something more painful. Where these conditions obtain, the moral law is mostly obeyed. Where they do not obtain, it is largely disobeyed. To destroy or diminish these conditions in a community is easy, for it can be done by a devastating war. To introduce them where they do not obtain is very much harder. Each of us, however, has from time to time opportunities to do a tiny bit towards maintaining or improving the reign of the moral law.