Graphic Rule


3.41. The principle of tolerance

Since the State is necessary but diminishes freedom, the question arises whether any good principles can be found to guide us in deciding how far the State should go in infringing individual liberty. John Stuart Mill's great essay On Liberty is an attempt to answer this question, a search for 'a principle by which the propriety or impropriety of government interference' may be tested (p. 72, Everyman).

Government interference may be divided into two great departments, interference with our freedom to do wrong and interference with the rest of our freedom. By 'wrong' here is meant something wrong independently of the State's laws, not just legally wrong in that it has in fact been forbidden by the State. A law against murder is an interference with our freedom to do the moral wrong of murder; but a law establishing income tax is an interference with our freedom to spend our own income in morally innocent ways. Only the last twentieth of Mill's Essay deals with the latter kind of government activity, where he takes an excessively individualistic view, while at the same time wrongly regarding socialism as not involving infringement of liberty (p.164, Everyman). The bulk of the Essay deals with State interference with acts supposed to be wrong in some extralegal way. This is the sphere of tolerance, though Mill does not call it so. Tolerance is non-interference with wrong or harmful activities. It is not mere non-interference. Or, at any rate, the only kind of tolerance that needs to be defended and upheld is non-interference with the harmful. There is no need to argue that we ought not to interfere with the good and the harmless.

To demand toleration for someone is thus not merely to assert that we should leave him free. It is to reassert this, or very nearly this, after someone has interjected 'except to do evil'. To demand toleration is to demand that people shall be left free even to do evil in many cases. When Pravda countered Herbert Morrison's article by saying that 'there is free speech in Russia for everyone except enemies of the people', it betrayed that it does not understand what freedom and tolerance are. To demand toleration is nothing so obvious and selfevident as to demand free speech for friends of the people. It is precisely to demand free speech for enemies of the people.

Part of Mill's principle of toleration is that ' own good, either physical or moral, is not a sufficient warrant . He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right' (op. cit., p. 73, Everyman). I think Mill meant more than he wrote here. I think he meant that a man's own good is not merely not a sufficient warrant for interfering with him, but no warrant whatever, and does nothing to strengthen any case for interference, so that it ought never to be mentioned in any argument for interference. This, at any rate, is the form in which I myself hold this principle. A State, in other words, should not behave like a school, which compels each child for his own good much more than for the goods of the other children. In a school the child is compelled to learn mathematics for his own sake, and compelled to be peaceable for the sake of the other children. In the State the citizen should be compelled only for the sake of the other citizens, and not at all for his own sake. He may be compelled to avoid tuberculosis, because tuberculous persons are a danger to others; but he may not be compelled to avoid cancer, because this disease is no danger to others. It is wrong to say that 'no man has a right to anything save to that which is really good for him', and that 'the individual is often a very bad judge of his own happiness', if from this you are going to infer that the State is a good judge of the citizen's happiness and has a right to compel him for the sake of it (F. C. Montague, The Limits of Individual Liberty, pp. 183, 189).

It has been thought that Mill's principle fails because it depends on a distinction that cannot be made in practice, the distinction between harmful actions that harm only the agent and those that harm others too. But, even if this distinction never can be rightly made, many authorities do in fact appeal to it, for they claim to be restraining a person 'for his own good'; and Mill's principle says that this is an improper claim in any case. Mill's principle involves that, whether or not it is possible to find actions that harm the agent without harming anyone else, the claim that the action harms the agent is never a good reason for the State to forbid it.

Those who adopt the principle that the State may compel the individual for his own good probably feel it to be selfevident; but no practical principle is selfevident. The great reasons against it are, first, that compulsion is an evil which it takes much good to outweigh, and, second, that it is usually improbable that the State is a better judge of the man's good than he is himself. When you fill out a form of application for a passport, on which you are asked to say what is the purpose or good of your journey, you realize vividly how restricted and blind is the State's conception of possible individual goods. The State is not a god who knows my good better than I do; it is a tyrannical fool who cannot see most of the goods there are.

Another negative principle may be added: the State may not interfere with the individual merely on the ground that his action is morally wrong. That an act is contrary to the moral law is no good reason for suppressing it. Neither the government nor any other body or person has a right to enforce all moral rules all the time. Neither the State nor any church has a right to prevent men from doing what they ought not to do as such. The view that 'the State has a right to punish all moral delinquency' (Montague, op. cit., p. 192) is false; and is probably held only by confusion with the view that the State has a right to compel a man to be moral when by so doing it can prevent great harm to others. What gives the State a right here is the possible harm to others, not the immorality of the act. If all morally wrong acts were legally forbidden by the State, there would be no difference between morality and legality, and the duty to obey the government would be man's only duty, and no one could ever do the right thing in spite of there being no compulsion to do it. That is, no one could ever do right 'of his own free will' as we say.

A third true negative principle is that the State may not forbid acts on the ground that they are contrary to the will of a god. No one has ever produced, or ever will produce, good and reasonable evidence for any statement that the will of some god is so and so. But even if we did know what the will of some god was, we ought not to follow it unless we found that following it lessened human misery; and we ought to determine, whether following it did lessen human misery, by empirical investigation without reference to its being ordained by a god. We ought not to say: 'It is commanded by a god, and therefore it must make people less miserable, no matter what the appearances are.'

Mill intended to offer also a positive principle, embracing all cases where the State may rightly abridge the freedom of an individual to do wrong. He said that all cases where the State may interfere are cases of the 'selfprotection' of mankind, cases of 'preventing harm' to persons other than the agent interfered with, cases of conduct 'calculated to produce evil to someone else' (op. cit., p. 73, Everyman).

Did Mill intend to say the converse also, namely that whenever the individual does harm to others he should be restrained by the State? No, he said that, while 'damage, or probability of damage, to the interests of others, can alone justify the interference of society', it does not always do so (op. cit., p. 150, Everyman). Therefore his principle is not a complete positive guide to State action in this matter. However much we accepted it, we should still have to use other considerations also in deciding when the State should intervene. This incompleteness on the positive side is no doubt correct. Any true principle for the direction of State interference must be incomplete and leave a great deal to be decided by other principles or by judgement. No principle can relieve us from the need for continuous judgement here, because human circumstances alter and because different peoples may properly make different choices.

The best expression of the principle of tolerance seems to be this: we must not suppress the evil behaviour of other men until reasonable examination has made it very probable that trying to suppress the evil would greatly lessen human misery upon the whole. I will enlarge on each phrase of this principle in turn.

'Reasonable examination.' There must be careful and thorough examination, according to the best methods of inquiry and canons of evidence that reason recommends, of the question whether efforts to suppress the evil would in fact greatly lessen human misery. That is practically to say, whether it really is an evil or only seems so to its would-be suppressors. This is a question of natural science, of the prediction of future events; and it is to be determined as reason indicates that questions of predicting the future are to be determined. No person or government has a right to suppress any activity, if it has not taken reasonable care to ascertain that both the activity does cause much human misery, and efforts to suppress it would be successful and greatly lessen human misery on the whole.

'Very probable.' The great goodness of freedom, and the great fallibility of man, demand that the government shall not suppress any freedom until it has ascertained by the methods of reason that it is very probable that the attempt at suppression will greatly lessen human misery.

'Trying to suppress the evil.' For it is essential to take into account the possibility that one's activities of suppression may be ineffective, in which case one would merely have added a second evil to the existing one. It is no good predicting the consequences of the total disappearance of the evil without at the same time predicting to what extent one's proposed measures of suppression are going to succeed. That the suppression of the evil would enormously increase human happiness is perfectly irrelevant, and no ground at all for action, if also the measures proposed will not in fact suppress it.

'On the whole.' What has to be found is the net gain or loss, which depends on all the gains and all the losses; and therefore we must look for every significant result. For example, the act of suppression is itself an evil, since all loss of freedom is an evil; and this evil must be reckoned in the accounts on the debit side, instead of being omitted as it often is by the censorious and the tyrannous. 'On the whole' is a fundamental principle of all practical wisdom, all good judgement about what to do.

'Lessen human misery.' This is the only criterion by which men's freedom to do wrong may be taken away. All suppressions and interferences not justifiable by this criterion are wrong. I have earlier noticed and rejected the two other criteria which are often held to justify intolerance, namely the will of a god and the moral law. They are both wrong and to be abandoned.

The principle is, then, that we must not suppress the evil behaviour of other men unless reasonable examination has made it very probable that our attempts to suppress it will greatly lessen human misery upon the whole. Lawmakers should be tolerant in making laws. Officers should be tolerant in exercising their powers under the laws. Electors should be tolerant in electing. Citizens should be tolerant in talking.

3.42. Free speech

The application of the principle of tolerance to speech and publication is as follows. Many publications are blasphemous; but this gives no one any right to suppress them, for the will of a god is not a proper criterion of what may be suppressed. Many publications are false; but falsehood also is not a proper criterion of what may be suppressed. Many publications are immoral; for example, lying is usually immoral, and many publications are lies. But this by itself gives no one any right to suppress or punish them, for the moral law is not a proper criterion of what may be suppressed. It is a true moral principle that no man has a moral right to publish what he himself believes to be false, and no man has a moral right to publish statements without taking reasonable care to ascertain that they are true. But these true moral principles do not by themselves give any right of suppression or punishment, for nobody has a right to enforce moral laws as such. A government may suppress a publication only if it has ascertained by reasonable methods that its attempt to suppress the publication would probably greatly decrease human misery or prevent its greatly increasing.

Above all, of course, it is essential for a government to tolerate criticism of itself. That criticisms of the government should freely circulate, including those which the government itself thinks to be mere abuse or grossly false or otherwise grossly unfair, is a very great safeguard indeed against those diminutions of human happiness which governments are liable to cause. In democracies the government is usually more tolerant of criticisms of itself than in autocracies; but it is by no means always tolerant enough. If, for example, you read Erskine May's account of the British Parliament's privilege rules, I think you will judge, as I have, that they amount to not tolerating reasonable criticism of the Parliament by the citizens who have to suffer from Parliament's doings. Fortunately, they are rarely applied. But there is nothing in the constitution to prevent their being applied; and the Parliament of 1945-50 contained a number of unusually selfrighteous and touchy politicians who invoked these rules against reasonable criticisms of their doings.

The greatest enemy of free speech in Britain is our laws about libel and slander, or rather the way in which our lawyers interpret whatever laws our Parliament makes about libel and slander. We have a great tenderness for people's reputation; and our lawyers make it very hard for us to publish the errors and shortcomings of living persons here. For this reason some important news about the United Kingdom is to be found only in foreign newspapers, and it is therefore wise to get the habit of reading some foreign newspaper regularly.

This doctrine about free speech is inconsistent with the doctrine expressed in Pope Leo XIII's encyclical letter of 20 June 1888, as translated by John A. Ryan and Francis J. Boland in Catholic Principles of Politics (New York, 1948, p. 174). Leo XIII there declared that public authorities ought diligently to repress the publication of 'lying opinions'. I propose to give reasons for free speech and against the doctrine of Leo XIII.

It is not a good reason for free speech to remark that 'people cannot help what they believe'. They can help publishing what they believe, for they can keep their thoughts to themselves. But, further, they can help what they believe to a large extent; for they can choose whether or not to seek and listen to evidence and argument on both sides of the question, whether or not to try to judge equably on the basis of all available evidence and argument, whether to be reasonable, in short. And their choice in this matter will largely determine what they believe.

There are two great and good reasons for free speech. One of them is simply that freedom is a great good, and any suppression of freedom is consequently an evil. And this is a very great and strong reason though it is short to say.

The other strong reason for free speech is that the toleration of free speech is far more likely to produce a general spread of true opinion than is the suppression of it; and truth and the general spread of truth are very great goods.

One of the premisses of this second argument, however, is disbelieved by many people. They hold that, if all views are allowed to be expressed, false views will be generally adopted and their true contradictories generally rejected. They hold that the false is more easily believed than the true, so that, if a man hears both a proposition and its contradictory freely asserted, he will usually adopt as true that one of the pair which is in fact false. Pope Leo XIII expressed this view in his encyclical already referred to, when he wrote:

If unbridled licence of speech and writing be granted to all, nothing will remain sacred and inviolate; even the highest and truest mandates of nature, justly held to be the common and noblest heritage of the human race, will not be spared. Thus, truth being gradually obscured by darkness, pernicious and manifold error, as too often happens, will easily prevail.

In these words the phrase 'mandates of nature' was probably intended to include moral rules; and it is probably a consideration of moral rules that chiefly leads people to adopt this view. They think that, if the contradictory of a moral rule is allowed to be preached as freely as the rule (e.g. 'you may have sexual intercourse with whomever you wish'), most people will adopt the contradictory and not the rule.

To me this proposition seems ridiculous on its face. Leo seems to be saying that a great truth has only to be contradicted by somebody in public to be generally disbelieved, and that seems absurdly improbable. But I shall not leave it at that. I shall develop an argument in favour of my premiss that the toleration of free speech is far more likely to produce a general spread of true opinion than is the suppression of it. My argument is that all men are fallible in their opinions and reasonings as in everything else, and therefore they need to take all available means of lessening the chance of their believing falsehoods, and the strongest means available to this end is to be and remain exposed to free criticism and argument and contradiction from all sides. To paraphrase and slightly weaken a statement by John Stuart Mill, the complete liberty of all men to contradict and disprove my opinion is a necessary condition of my being justified in assuming its truth. People go mad if they live in a world of their own; and governments and popes go mad if they hear no independent voices criticizing them. The habit of rational discussion, of listening to argument and searching for evidence, is of enormous value in increasing the spread of true instead of false opinions; and this habit is starved and discouraged by all intolerance of free speech. If you suppress the contradictions and arguments of others against you, you are not doing all you could to lessen the risk of your being wrong. If you are a governor this is immensely serious to those you govern, and a grave breach of your duty towards them.

This is true even of moral rules, the case where the intolerant have the strongest argument. Moral rules go against the flesh, and the reason for going against strong desires of the flesh is sometimes obscure. The reason for a moral rule is more obvious to the experienced than to the inexperienced. Hence there is a case for saying that moral rules may not be discussed, because the reasons for them cannot yet be properly appreciated by those who need them most.

Yet here, too, the case for intolerance is bad. Every moral rule either has a good reason, or ought to be abandoned as a useless restriction on liberty. A moral rule, like all laws, is a restriction on liberty; and a restriction on liberty is always improper unless it can reasonably be shown to be very probably the cause of a great diminution of human misery. While we should emphatically and solemnly preach to the young such moral rules as we believe to be important, we should also right from the beginning offer them the reason which in our eyes justifies these rules; and there is no good reason except the appeal against man's misery. When moral rules are not allowed to be criticized, bad ones creep in, and good ones are held in a stupid and immoral way. The man who suppresses the contradictors of his moral rules implies that either there is no good reason for his rules or at least he is incapable of giving it. It is unreasonable for a grown person to hold a moral rule for which he can give no good reason. To do so is to be still in the prison of taboo.

It is true that a powerful preacher will sometimes sway people to the side of bad action. But this is done by religious or moralizing persons more often than by others. It is true that sometimes, of a pair of contradictory statements, the false one obtains belief more easily than the true one. But the best precaution we can take against that happening is always to let both sides argue, never to suppress one side. For there is no good reason to believe that, if we suppress one side, we the suppressors are exempt from the human tendency to believe the false, or from the need of contradictors to keep us straight. Every man, however wise, needs all the criticism and argument and opposition he can get to keep him nearer truth than falsehood. As Mill put it in his great chapter on liberty of thought and discussion:

Silencing the expression of an opinion is ... robbing the human race; posterity as well as the existing generation; those who dissent from the opinion, still more than those who hold it. If the opinion is right, they are deprived of the opportunity of exchanging error for truth: if wrong, they lose, what is almost as great a benefit the clearer perception and livelier impression of truth, produced by its collision with error (op. cit., p. 79, Everyman).

3.43. All men are fallible

I have based the demand for freedom of speech in part on the reason that freedom of speech is far more likely to produce a general spread of true opinion than is the suppression of any form of speech. And I have based this reason on the further reason that all men are fallible. I propose now to take the matter still another step further back, because many people deny or tend to deny the doctrine that all men are fallible. It is official Papist doctrine that the Pope, when speaking ex cathedra on faith or morals, is infallible; and many who are not Papists also feel that there is something wrong with the doctrine that all men are fallible.

First it is necessary to be clear what the word 'fallible' means. To say that anyone is fallible is simply to say that he sometimes makes a mistake. And to say that he is infallible is to say that, on the contradictory, he never makes a mistake. The notion is applicable not merely to persons, but to anything whatever that can in any sense habitually succeed or fail. For example, one could perfectly well call a cigarette-lighter infallible if it flamed every time one pressed the button, and never failed to flame. And this would be exactly the same sense of the word as when a Papist says that the Pope speaking ex cathedra on faith or morals is infallible.

Thus the notion of infallibility contains the notion of all or always or every time, and that of fallibility contains the notion of not all or not always or only sometimes. This involves that it makes nonsense to talk of infallibility or its contradictory in a context where no question of always arises. It is sense to say that the Pope speaking ex cathedra is infallible, but nonsense to say that when the Pope spoke at a certain moment he was infallible. It is sense to say that your lighter flames infallibly, but nonsense to say that it flamed infallibly when you used it just now. The words 'fallible' and 'infallible' have no application to particular cases and events. They apply only to a general class of cases or events. The Pope's utterances ex cathedra can in general be fallible or infallible; but one of them in particular cannot be either fallible or infallible; it can only be true or false. Anyone who calls some particular statement infallible is either talking nonsense or saying in an improper way that it is true.

Given that this statement made by this speaker is false, it follows that this speaker is fallible. But, given that this statement made by this speaker is true, nothing follows about whether this speaker is fallible or infallible. Given that a speaker is fallible, nothing follows about the truth or falsehood of any particular statement he makes. But, given that a speaker is infallible, it follows that any statement he makes is true. An infallible person would therefore never be required to give any reason or evidence for his statements. Or, rather, he would be able to give one and the same final reply to every challenge, no matter which of his statements he was challenged about. To the question How do you know that?, no matter which of his statements it referred to, he could always reply with truth and finality: 'It must be true because I say it and I am infallible.'

Thus to say that all men are fallible is to say that every man without exception makes at least one mistake in his life, and no man ever has gone or will go through his whole life without making a single mistake. And this is the proposition I wish to recommend now.

To say, or to deny, that all men are fallible, is to make an assertion about the course of events. The only justification for assertions about the course of events is experience. Direct or indirect experience, and experience extended or not extended by generalization and deduction; but in any case experience. Therefore this question whether all men are fallible is to be settled by nothing else but our experience of men.

I believe that all men are fallible on the ground of my experience. It seems to me an overwhelmingly probable induction from every day of my life. Every man with whom I have conversed for an hour or more has in my opinion evinced at least one error in that period. Every statement that I have read amounting to ten or more pages has appeared to me to contain at least one falsehood, if I was capable of judging it. My own past life, when I look back on it, appears to me full of mistaken opinions. Are you acquainted with any man who has in your opinion never made a mistake?

Ought I to go from the premiss, that all men I have met are fallible, to the conclusion that all men whatever are fallible? It seems to me that I ought, again on the ground of experience. The experience I have gathered of the general nature of man makes it immensely probable that all men will frequently entertain false opinions and utter false statements. That is, the fallibility which I have observed in all of my acquaintances and in myself seems clearly to depend on universal features of human nature. It depends in particular on the capacity of human language to make assertions about any matters whatever, whether we know anything about those matters or not, together with our frequent desire and need to know or believe propositions about all kinds of things, while at the same time our opportunities for really experiencing most things are extremely limited. Descartes was feeling for this when he explained human error as due to man's combining an infinite capacity to will with a finite capacity to judge. What he took for an infinite capacity to will is the capacity to construct a huge number of statements, which is inherent in human language owing to its huge and always extending vocabulary. The beasts, though also fallible, make far fewer mistakes than we do because they say far less.

Some persons tend to believe that the question whether all men are fallible is to be answered, not by experience and generalization, but by arguments from the Bible or some other book. A learned Papist once suggested to me that the infallibility of the Pope can perhaps be inferred from Matt. xxviii. 20 and Luke x. 16. But the question whether any man is infallible is not to be answered by pointing out that some book says that some man is infallible. For all books, including the Bible, are utterances by men; and our experience of men teaches us that their books, like their spoken words, are fallible. And therefore, if any book contains a sentence asserting that a certain man is infallible, it is extremely probable that the book shows its own fallibility by being false in this instance. The only respect in which the Bible is good evidence on the question whether all men are fallible is that, by the falsehoods which it contains, it enforces the generalization that all men are fallible. It would be just as unreasonable to believe that the Pope was infallible ex cathedra because the Bible said so (if it did) as it would be to believe that all men had blue eyes because the Bible said so (if it did). The statement that all men have blue eyes is disproved by looking at men until you see one whose eyes are not blue; and that is the end of that, no matter what any book may say. Similarly, the statement that the Pope's utterances ex cathedra are infallible is disproved by reading them until you come to one that is false; and that is the end of that, no matter what any book may say. In each case it is a question only of looking to see what happens.

'But', it is sometimes argued, 'the Bible and the Pope ex cathedra are not human utterances; they are utterances by God, and therefore infallible.' This is an assertion about what happens. Therefore the proper way to decide whether to accept or reject it is to appeal to experience. Experience overwhelmingly indicates that it is false and is to be rejected. We find the Bible and the Pope's pronouncements written or printed on ordinary human paper in ordinary human ways. We can observe a new papal pronouncement ex cathedra being composed with pen and paper by men in the Vatican. That they are fallible documents is abundantly shown by the falsehoods and horrors which they contain. Example of falsehood: Joshua made the sun stand still. Example of horror: thou shalt not suffer a witch to live. It is grossly bad judgement to claim for any document that it is true in every sentence because it is inspired by someone infallible.

The most serious and respectable objection to the doctrine that all men are fallible is the uneasy feeling that somehow or other this statement refutes itself. But it does not refute itself; and I will now try to remove the feeling that it does from the minds of any of you in which it may exist.

The most obvious kind of selfrefutation is selfcontradiction. To contradict oneself is to say and deny the same thing, or to entail one's own denial. Now 'all men are fallible' does not do this. It is not a selfcontradictory statement like 'all men have blue eyes but some do not', or like 'a father is not a parent'. It is a selfconsistent statement. It presents a possibility which, as far as logic tells us, could actually be realized in the world. Thus it is not selfrefuting in the obvious sense of selfcontradictory.

But there is another kind of selfrefutation besides selfcontradiction. If a man opens his mouth and says 'I am not speaking now', he makes a selfconsistent but false statement. The peculiarity of it is that the fact, to which one appeals to show that the statement is false, is the utterance of the statement itself. Precisely by uttering the statement he produces the state of affairs in virtue of which the statement is false. (Similarly, if a man says 'I am speaking now', he makes his statement true by uttering it.)

The statement that 'all men are fallible' is not selfrefuting in this way either, for you do not by uttering it produce an infallible man. (It would be remarkably convenient if you could make yourself infallible by declaring that 'all men are fallible'.)

These are the only two ways in which a statement can refute itself, so far as I can see. Either it contradicts itself, or by its utterance it provides a negative instance which disproves itself. Since 'all men are fallible' does neither of these, it is not selfrefuting.

In addition to selfrefutation there is perhaps such a thing as selfstultification. The statement that 'what I say is never worth saying' neither contradicts nor otherwise refutes itself; but it appears to stultify itself. A statement stultifies itself, we may define, if it entails that to assert it would be silly.

The statement that 'all men are fallible' does not stultify itself. On the contrary, if it is true it is very important, and a wise man will assert it from time to time.

I fear that, in spite of these explanations, the uneasy feeling may remain with some of you that the statement that 'all men are fallible' does after all somehow do away with itself. If that is so, I ask you to write down at your leisure exactly how it does this, and then to look for a flaw in what you have written. I think you will probably find a flaw; but, if you do not, bring it to me and I will try to find a flaw in it.

I will give now two examples of finding a flaw in such attempts. People sometimes say that 'those who argue against infallible authority claim infallibility for themselves'. The flaw here is that this is simply false. We do not claim infallibility for ourselves. Every man who utters a statement thereby implicitly claims that that statement is true. But he does not thereby claim that all the statements he ever utters are true. That is, he does not claim that he is infallible. Whenever a man makes a sincere statement he thinks it true; but no sensible man has ever thought that all the statements he had ever uttered or would ever utter were true. The statement that 'all men are fallible' is the same in this respect as the statement that 'all men are mortal'. The speaker of either of them claims to be telling a truth but does not claim to be infallible. Every statement equally claims truth for itself, and every statement equally refrains from claiming that its utterer is infallible.

This is a mistake that has been made by the assailants of infallibility as well as by its defenders. Mill wrote that 'all silencing of discussion is an assumption of infallibility' (op. cit., p. 79, Everyman). This, I regret to have to admit, is false. To silence a discussion is not to assume that one is infallible. The editor who declares that 'this correspondence must now cease', the chairman who forbids the raising of a certain topic, the headmaster who forbids the boys to debate birthcontrol, are none of them assuming themselves infallible. They are merely assuming themselves to be right in thinking that they ought to silence this particular discussion now. Silencing a discussion is an act of government. Are we to say that all acts of government assume the infallibility of the governor, or that only this special kind of act of government assumes the infallibility of the governor? Both are obviously false, but Mill's sentence implies that one of them is true. However, it is only Mill's expression that is wrong here. What he had in mind was the truth that only a belief in his own infallibility could morally justify a governor in permanently forbidding adult persons to express a certain view (cf. p. 85). But he failed to say clearly that it is a matter of moral justification, not of logical assumption.

Here is a second example of finding a flaw in an attempt to show that the doctrine that all men are fallible disposes of itself. People sometimes think that the proposition that 'we are fallible' entails its own contradictory in the following way: 'Assume that we are fallible; it follows that we may be wrong in saying that we are fallible; and from this in turn it follows that we are infallible.'

The flaw here is that it is false that the second consequence follows. From 'we may be wrong in saying that we are fallible' it does not follow that 'we are infallible'. 'Are' never follows from 'may be'. From possibilities alone one cannot rightly conclude to facts. We may call this fallacy the illicit process from possibility to actuality.

These two examples must suffice to illustrate the endless task of pointing out the flaw in fallacious arguments against the doctrine that all men are fallible. With them I conclude my recommendation of this doctrine, which is one of the premisses for one of my arguments for free speech. But I want before leaving the topic to warn you against a certain misuse of this doctrine, a misuse which was perhaps committed by Oliver Cromwell on a famous occasion. In his letter of 3 August 1650 to the General Assembly of the Kirk of Scotland, Cromwell wrote: 'I beseech you, in the bowels of Christ, think it possible you may be mistaken.' What was his purpose in thus reminding his opponents of the fallibility of man? One good reason for doing this is to persuade your adversary not to suppress the publication of views opposed to his own; and that Cromwell had this purpose in mind is suggested by his mentioning, elsewhere in the letter, that the Kirk had been suppressing his papers whereas he had been publishing the Kirk's letters to him. On the other hand, the letter never unequivocally says that this is the purpose of the famous reminder; and some parts of it suggest that he was using it in another way, namely to insinuate that 'since you may be wrong, you are wrong and I am right'. One part of it implies that he knows he is right because he feels the grace of God upon him, and that way of thinking seems more typical of the man. Anyhow, the argument that 'since you may be wrong, you are wrong and I am right', whether or not Cromwell was guilty of it, is utterly fallacious and a damnable misuse of the doctrine of human fallibility. It is another case of the illicit process from the 'may be' to the 'are'. The undoubted fact that you may be wrong is an excellent reason for your allowing free speech to your opponents. But it is no reason for them to claim that you are wrong; for, from the truth that all men are fallible, nothing follows about who is in the right in any particular controversy. It would be better to say: 'I beseech Christ to make me think it possible that I am mistaken.' The fallibility principle should make us not merely patient of criticism, but eager for it.

This fallacy is often committed by lawyers crossexamining witnesses. 'But you could be mistaken, could you not?', they ask. Of course he could be mistaken, because he is fallible; but it does not follow that he is mistaken. It is a hard question to reply to. If you say 'No, I could not be mistaken', you appear to claim infallibility. But if you say 'Yes, I could be', you appear to withdraw your statement, or at least some of the force of it. Perhaps the best reply is 'One always can be mistaken, but one sometimes is not', or simply 'I could be, but I'm not'.

This completes my defence of my principle for the toleration of publication, against Leo XIII's position that authorities ought diligently to repress the publication of 'lying opinions'. Now that I have given the positive arguments for my principle, which are independent of any errors Leo XIII may have made in the statement of his view, I may remark that his letter makes his position more attractive by means of three confusions which are not detected by most of its readers. It confuses what is false with what the public authority thinks false, tacitly assuming that the public authority is infallibly right about what is false. It also confuses falsehood with lying, tacitly assuming that whoever utters a falsehood knows that it is a falsehood and so is guilty of the moral wrong of lying. And thirdly it confuses the true moral law, that men ought not to lie, with the false moral law, that public authorities ought to prevent men from lying.

3.44. The limits of tolerance

There are limits to tolerance; and this is implied by the principle of tolerance as I have formulated it. We may interfere with an evil where we have good reason to believe that our interference will greatly lessen human misery on the whole. I wish to point out certain departments in which the limits of tolerance come sooner than liberals have been inclined to think.

In the first place, a man's official position may diminish his right to free publication. For example, a teacher in a public institution has less right to publish his thoughts than a man who lives by mining. For it may be the case that the utterance of a certain opinion by a miner does not greatly increase human misery, but the utterance of the same opinion by a teacher does so. Perhaps it can be shown that a teacher who preaches suicide, or one who preaches communism, is probably greatly increasing human misery. If so, we are justified in depriving him of that job. But it is unlikely that it can be shown that a miner who preaches either of these things is doing much harm.

In the second place, the tolerance that should be extended to bad religions is a good deal less than is often claimed nowadays. While every religion should have freedom to publish and to preach, religion gives no right to disobey the ordinary civil laws made for the good of the people in this life. It gives no right to avoid military service, though a government is often wise to grant exemption from military service as a grace. A pastor's need to conduct a religious ceremony gives him no right to break a law that rations petrol or limits speed on the highway. Murder is still murder if someone holds a religion of human sacrifice. No civil crime becomes legal by being done out of religious beliefs or sentiments. Whatsoever is illegal in the commonwealth must be forbidden in the church. All religious practice must yield to, and be overruled by, the need to lessen the misery of man on earth, wherever there can reasonably be shown to be a conflict between the two. It is therefore too strong to say, as Locke did in his letter on toleration, that 'no man whatever ought ... to be deprived of his terrestrial enjoyments upon account of his religion'. A man's religion has led him to kill prostitutes before now; and a man who kills prostitutes ought to be deprived of some or all of his terrestrial enjoyments.

Still less does religion confer any right to make and enforce special laws incumbent on the whole population regardless of its religion. Locke wrote truly that 'whatsoever is lawful in the commonwealth cannot be prohibited by the magistrate in the church'. The religion of the English Nonconformists gives them no right to force the Sunday Observance Law upon the people; and the existence of this law is a gross tyranny. Of the freedoms which religion should not have, the most fundamental is that it should not have the freedom to control those who do not wish to obey it. That is, it should have no legal force; and no law should be made or unmade for the sake of any religion. In practice, unfortunately, it is often easy for a religious group to take away other persons' freedom merely by saying that this freedom is 'offensive to their religious feelings'. As Max Beerbohm has put it, 'the Nonconformist conscience makes cowards of us all'. Nothing ought to be made illegal because it is a sin (sin being a religious notion), but only because it is injurious to the earthly life of man. Nothing ought to be punished because it is a sin, but only because it is illegal, that is, a contravention of the existing law of the land. For example, the following ought not to be a law in any State: 'He that sacrificeth unto any god, save unto the Lord only, he shall be utterly destroyed' (Exod. xxii. 20).

The most important limit to toleration is the limit to be placed on the freedom of those who wish to take away freedom. A government may interfere when reasonable examination of the evidence has made it very probable that the interference would greatly diminish human misery; and one kind of harm which may justify such an interference with someone's liberty is that harm which tends to overthrow the general reign of liberty. We may interfere with the liberty of persons who are likely to interfere with everyone's liberty. As Dr. Popper has well said, we have the right not to tolerate the intolerant. We should tolerate even them whenever we can do so without running a great risk; but the risk may become so great that we cannot allow ourselves the luxury.

The most essential thing not to tolerate is any move that is very likely to give power to uncriticizable and irremovable governors. At the present time this means mainly the Communists, a body of thoroughly intolerant and very active persons, whose fellows are already in tyrannical control of about a third of the human race. Since they are thoroughly intolerant, we have a good right not to tolerate them. For the sake of freedom we do tolerate them a great deal; but we should watch them always, and suppress them whenever reasonable examination makes it very probable that their activities threaten to end the reign of tolerance altogether, and that our suppression would stop this threat. Fortunately, we are well aware of this threat.

There is another great threat to freedom of which we are not well aware. The second most intolerant and active body in the world today is the Papist Church. It is the policy of this Church, long fixed and declared, that 'no State is justified in supporting error or in according to error the same recognition as to truth' (p. 314), and that 'the fact that the individual may in good faith think that his false religion is true gives no right to propagate it' (p. 318), and that a Papist State 'could not permit to carry on general propaganda' (p. 320). Those are quotations from an official Papist book on politics, namely Catholic Principles of Politics, by John A. Ryan and Francis J. Boland, New York, 1948.

The danger of the Papist Church is not generally realized in England today. This is partly because the Communist Party is a greater danger, and the Papists are against the Communists. Partly also it is because most Papists do not know the political doctrines of their own Church. But mainly it is due to a third cause. It often happens that a body, which is fundamentally intolerant, turns tolerant while it is a minority among a tolerant majority. That is notably the case with the Papists in England now. They are tolerant men. But the doctrine of their Church is fundamentally and irretrievably intolerant; and whenever it comes into power in a particular place it turns intolerant in fact. The more liberal members, who often hold the higher places while the Church is a minority among a liberal majority, are now gradually replaced by illiberal leaders in greater harmony with the essential philosophy of the body. Hence the need to restrict the influence of Papists in England now is greater than it appears from the tolerant nature of the Papists with whom we are acquainted. Those are not the men who would be in power if the Church were in power. Englishmen have in the last 150 years gradually abolished their former safeguards against Papist control until there are almost none left, and no evil results have yet appeared, and so we are confident that all is well. But meanwhile the power, size, and prestige, of this body in the country have been steadily increasing, and its ancient principles of intolerance have been affirmed more explicitly than before. There are rocks ahead that must be seen to be avoided.

We have the right to see that neither of these intolerant bodies gets much influence in the government, or in any other powerful body, such as a trade union; and we ought to do so. We ought sometimes to see that individual members of these intolerant bodies are kept out of influential professions like the foreign service, the civil service, and the service of the elementary public schools. I do not say that you should never recommend a Papist for a post in the civil service; but I do say that you are to consider carefully that he is a member of a body which is always thoroughly intolerant when it has the power to be so, that he probably does not know himself how intolerant his church is, and that his presence in the civil service must tend to increase the influence of his church. I do say that you are to disregard all accusations that you are intolerant, or that you are persecuting religious minorities, or that you are unjust to an innocent man, in considering his religion; for your intolerance is only being intolerant of intolerant bodies; but the intolerance of his church is unlimited; and it is far more important that men in general should be shielded from that, than that this individual should be shielded from all disabilities arising out of his unfortunate allegiance.

Third on the list of dangerous intolerant bodies in England today come the trade unions. Their intentions are much less bad than those of either the Communist Party or the Papist Church; but their legal powers are much greater and much too great. The policy, to which they tend, that a man may not practice a trade without belonging to a union, is a great interference with freedom, and is protected by extraordinary legal exemption from accountability to the courts. Trade unions can break promises with impunity; and they can force a man to change his way of living with near impunity.

The principle of intolerance does not, however, indicate any denial of free publication to the intolerant. They should be allowed to publish their views as much as they please; and our defence against that should only be to publish our replies, never to suppress their publications. If we did otherwise we should offend against the principle that the free publication of all opinions is far more conducive to the general reign of truth than any suppression of any opinion whatever. If we judge that the voice of the intolerant is being heard too much and the voice of the tolerant too little, the right way to redress the balance is always to increase the voice of the tolerant. Government force and money may not be used to stifle the voice of the intolerant, but they may be used to increase the voice of the tolerant; and it is a pity that such use is sometimes condemned as 'propaganda'. It is not propaganda in the sense of lying libel, such as telling unproved stories of rape and murder about the enemy. It is propaganda only in the neutral sense in which all practical speech is propaganda, including these lectures of mine.

Censorship is not one of the legitimate forms of intolerance. That is, the government should not require proposed publications to be submitted beforehand to a censor for approval. By so doing it would prevent that clash of opinions from which truth is most likely to emerge. Freedom of expression should be absolute both in politics and in religion. Everyone should be allowed to express every opinion about gods and morality and ritual and man, whether blasphemous or pious, immoral or moral. Free speech in religion includes the freedom to preach and proselytize and try to make converts, which is denied to Protestants at the present time in Spain. It is a sad thing that one of our great defenders of toleration, John Locke, believed that atheism was not to be tolerated. It is a sad thing that atheists in U.S.A. and U.K. are still under serious disabilities in fact, though not I think by law. That is, the frank atheists are. But, in view of this intolerant attitude, there are probably many who conceal their atheism. As Disraeli made his characters say, 'Sensible men are all of the same religion'. 'And, pray, what is that?' 'Sensible men never tell.' There ought to be complete liberty of conscience, in the sense that anyone may say what he thinks true about gods and the moral law, as opposed to what any authority thinks true about them; and Leo XIII was grossly distorting this when he wrote that the only true liberty of conscience is liberty to follow the will of God. 'The moral decisions of others should be treated with respect, as long as such decisions do not conflict with the principle of tolerance' (K. R. Popper, The Open Society, U.S. ed., p. 508).

Some people give the name 'censorship' to all action whatever in restraint or punishment of publication. But that is a misuse of the word. A censor is an officer who examines each proposed publication because no publication may be made without his permission. In England now there are censors for the performance of plays and the showing of films, but not for the publication of books or periodicals or newspapers. It is not censorship if a publication is prosecuted after it has been made; and it can be prosecuted for other crimes besides not having obtained permission from an official censor. Nor is it censorship if some unofficial body declines to handle a publication because of its contents. It is not censorship if W. H. Smith & Son refuse to distribute the New York Times in this country. To condemn censorship is not to condemn all prosecutions for libel, nor is it to say that a newsagent is obliged to handle all publications however much he thereby exposes himself to prosecution for libel. Some libel law is certainly desirable. It is not good that an honest man should have no redress if a newspaper publicly calls him a 'hired liar'.

If any further restraint on newspapers were required, beyond the existing law of libel, it would not be to prevent them from publishing certain things, but to compel them to publish certain things. For instance, their constant refusal to publish the bad acts of newspapers is a serious harm to the community. I do not see, however, that any law could remedy it. A government office to decide what newspapers must publish would inevitably be staffed by men of bad judgement, and would do much more harm than good. All that a government can do in this respect is, apparently, to allow and encourage private enterprise in the publication of newspapers, in the hope that sometimes one newspaper will make known the wickednesses of another.

Graphic Rule