Chapter Two of On Liberty – “Of the Liberty of Thought and Discussion” – is the longest chapter in the essay. There is good reason for this: Mill sees his defence of free speech as a model of the broader defence of liberty. His analysis in this chapter is:
... of much wider application than to only one division of the subject, and a thorough consideration of this part of the question will be found the best introduction to the remainder. 1.16
The Harm Principle defines the limits of free expression. Of course, since speech is rarely harmful in Mill’s sense of damaging a person’s rights, the principle therefore allows for a great deal of latitude in this area. And Mill’s defence is robust. He offers four arguments for unlimited freedom of discussion, even when the prevailing opinion is correct. There is more at stake here than merely truth.
... the peculiar evil of silencing the expression of an opinion is, that it 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. 2.1
Mill’s first and strongest argument against censorship, and the one given most space in the chapter, is the argument from fallibility:
... if any opinion is compelled to silence, that opinion may, for aught we can certainly know, be true. To deny this is to assume our own infallibility. 2.41
History tells us repeatedly that feeling certain is no guarantee of being right. Many mistakes have been made because authorities could not accept their own fallibility. Mill cites the powerful examples of Socrates and Jesus Christ, both of whom were silenced because their ideas and opinions – which are now accorded great respect – were, at the time, held to be wrong or blasphemous. He also mentions the Stoic Emperor Marcus Aurelius who, despite being an unusually virtuous and wise ruler, nevertheless persecuted Christians because of the strength of his pagan beliefs. Mill uses this example to suggest that any argument from conviction that allows for the suppression of non-Christians in this era will equally legitimise Marcus Aurelius’ actions against Christians then. You can’t have it both ways, argues Mill.
People tend to trust their own opinions more than they should, given how little thought, research and discussion goes into them; and if they are not sure of their own thoughts, then they rely on the received opinions of their own small world:
And the world, to each individual, means the part of it with which he comes in contact; his party, his sect, his church, his class of society... 2.4
But Mill points out that other parties, sects, churches and classes think very differently. In fact, our values are wholly contingent: being a Christian in London or a Buddhist in Beijing is merely an accident of birth. Mill’s implication here is that public morality – in this case, Victorian Christianity – is equally contingent and dependent upon historical context, and not the timeless truth it is thought to be.
As usual, his primary concern is the liberty of individuals to make up their own minds, since intellectual freedom is at the heart of Mill’s conception of the good life.
... it is not the feeling sure of a doctrine (be it what it may) which I call an assumption of infallibility. It is the undertaking to decide that question for others, without allowing them to hear what can be said on the contrary side. 2.11
Although such freedom of expression is now mostly guaranteed by law, Mill nevertheless gives examples of some ‘rags and remnants of persecution’. One is that, in his day, atheists were not considered trustworthy enough to give evidence in court (if that sounds ridiculous, consider how unlikely it is to see a US President – or even a British Prime Minister – who is a professed atheist).
Mill acknowledges some objections to the argument from fallibility that he then goes on to address. Firstly, that the fear of being wrong should not prevent people acting in good faith on their convictions.
There is no such thing as absolute certainty, but there is assurance sufficient for the purposes of human life. 2.5
But, Mill replies, that assurance only comes when an opinion has been properly contested and no conclusive argument against it has been produced. Censorship, by removing the possibility of contesting the opinion, removes the grounds for that assurance. Good judgement involves learning from mistakes and giving due consideration to all sides of a question. For an opinion to be trustworthy it must first be challenged:
The most intolerant of churches, the Roman Catholic Church, even at the canonisation of a saint, admits, and listens patiently to, a ‘devil’s advocate’. The holiest of men, it appears, cannot be admitted to posthumous honours, until all that the devil could say against him is known and weighed. 2.8
The second objection Mill addresses is that certain beliefs may be essential to the security and integrity of society, and contradiction of those beliefs may legitimately be censored. Mill is thinking here of religious beliefs, notably the existence of God and the rewards of Heaven. But, argues Mill, who is to say for certain that any belief is essential? People are often wrong about the truth, and they can be equally wrong about the value of a belief. But the value of a belief also seems tied to its truthfulness:
... no belief which is contrary to truth can be really useful. 2.10
– and the best path to truth is a free contest of ideas. That contest, of course, can be quite bloody and some have argued that persecution is a good test for truth to pass (for example, the martyrdom of the early Christians). If so, then savage repression of true ideas will, in the end, lead to their triumph – just as Christianity triumphed after its persecution by the Roman Emperors. But Mill insists that this is an appalling price to pay, and remarks also that persecution tends to delay the acceptance of truth. The Reformation, for instance, ‘broke out twenty times before Luther’ (§17).
In practice, disagreement is rarely a matter of outright truth pitted against outright falsity. Mill acknowledges that the situation is usually more complicated than that:
... though the silenced opinion be an error, it may, and very commonly does, contain a portion of truth; and since the general or prevailing opinion on any subject is rarely or never the whole truth, it is only by the collision of adverse opinions, that the remainder of the truth has any chance of being supplied. 2.42
Mill’s second argument is that most claims contain at least partial truth, and to suppress them is therefore to lose a ‘fragment of wisdom’ (§34). An analogy can be drawn with politics. In most democracies there are at least two major parties, one representing conservative views and one advocating reform, who alternate in government. Both are necessary and both represent part of the solution – and often, Mill argues, the claims of the party not in government deserve the most attention as they reflect interests that are currently being neglected.
An example from the world of ideas is Jean Jacques Rousseau, who forced the rational enlightenment to confront its darker side: Rousseau’s ideas on the corrupting effect of civilisation and the superiority of natural man provided a healthy antidote to the optimism of the Philosophes. Mill admits that Rousseau’s claims were partly false, but his ideas are at least considered worthy of attention.
It might be objected that in one particular case, the truth is not partial but complete: the Christian morality is a perfect system that needs no supplement. Mill’s response is cool and devastating:
... unless ideas and feelings, not sanctioned by it, had contributed to the formation of European life and character, human affairs would have been in a worse condition than they now are. 2.37
He goes on to add that the ‘exclusively religious type’ is, in fact, so abject and servile that it offers a powerful argument for the value of more diverse influences upon character and morality.
It is worth noting a strategy that Mill does not adopt in his defence of free speech. As these two arguments show, Mill is not a relativist about truth: he believes that truth exists and that it is valuable to human progress. A relativist might argue that moral truths in particular do not exist, and therefore all competing opinions are equally valuable. But Mill’s utilitarianism gives him the grounds to insist that moral claims are based in facts about utility, and therefore can be true or false.
This appeal to utility raises another question. Mill is optimistic about the success of ‘the collision of adverse opinions’ in reaching the truth. In the short term, however, that does not seem likely: hate speech, for instance, can have immediate and terrible effects. But Mill makes a good case that in the long term, false opinions tend to be replaced by true ones:
Wrong opinions and practices gradually yield to fact and argument: but facts and arguments, to produce any effect on the mind, must be brought before it. 2.7
The two preceding arguments make a good utilitarian case for not censoring opinions that are at least partially true. But Mill has not yet given an argument against successful censorship – that is, suppression of only false beliefs. His other two arguments address this.
Even if a prevailing opinion is certainly true, that is still not a sufficient reason to censor debate. Mill is not only concerned with the truth of a belief, but also its utility. True beliefs can lose their value if not properly cultivated.
... even if the received opinion be not only true, but the whole truth; unless it is suffered to be, and actually is, vigorously and earnestly contested, it will, by most of those who receive it, be held in the manner of a prejudice, with little comprehension or feeling of its rational grounds. 2.43
Mill is attacking belief derived solely from authority. He is concerned less about whether it is true than why and how it is believed. Opinions that are held thoughtlessly, like a prejudice, are not properly understood and cannot easily be defended. Such an opinion becomes ‘but one superstition the more’, and this is ‘not the way in which truth ought to be held by a rational being’ (§22). The best remedy for this is debate and disagreement:
... however true it may be, if it is not fully, frequently, and fearlessly discussed, it will be held as a dead dogma, not a living truth. 2.21
Understanding why a belief is correct involves understanding why other, competing theories are wrong. This is plausible, at least: the best debaters always seek to understand their opponent’s arguments as well as their own. Mill insists that it is not even enough to be taught the reasons for a belief in the way mathematicians are taught the proof of a theorem. Only by hearing opposite views passionately and genuinely expressed do we really grasp the grounds of our beliefs – and if there is nobody to challenge us, we must do the job ourselves:
... if opponents of all important truths do not exist, it is indispensable to imagine them, and supply them with the strongest arguments which the most skilful devil’s advocate can conjure up. 2.23
Mill acknowledges the objection that this may be setting the bar too high. Ordinary people cannot be expected to explore the grounds of their beliefs so thoroughly. It is enough for them to have the right beliefs and ‘trust to authority for the rest’ (§24). However, even if this were sufficient, at least the aforementioned authority needs access to debate and discussion: everyone benefits from knowing that ideas have been accepted only after active and robust disagreement.
Mill’s final argument for free discussion is similar to his concern about ‘dead dogma’ and addresses again the utility of uncontested beliefs. Even if a belief is correct:
... the meaning of the doctrine itself will be in danger of being lost, or enfeebled, and deprived of its vital effect on the character and conduct: the dogma becoming a mere formal profession, inefficacious for good, but cumbering the ground, and preventing the growth of any real and heartfelt conviction, from reason or personal experience. 2.43
If an idea is never challenged we lose the ‘vivid conception’ of that idea and retain only ‘the shell and husk’ of its meaning (§26). This is particularly true of ethical and religious doctrines: once the struggle to establish the doctrine is over, beliefs lose their power to change feelings and conduct:
... until it almost ceases to connect itself at all with the inner life of the human being. 2.27
This is important, since the utility of any belief is measured by its tendency to have a practical effect on people’s lives. Once that belief has lost its power, then it has also lost much of its utility. Mill offers as an example the Christians of his own time who seem unmoved and unchanged by the words they speak. In this respect they are far removed from the martyrs who suffered persecution for their beliefs.
Mill acknowledges an objection to this argument: it seems to follow that somebody must always be advocating error for the truth to remain effective. Mill takes the point and recognises that as time goes by, there may be fewer contested beliefs. But he insists that the value of disagreement is undeniable, and where there is no longer debate, teachers should find a ‘substitute’ to keep beliefs vital. Mill suggests, by way of example, the Platonic dialogues; and, to a lesser extent, the disputations of medieval scholasticism.
A more serious objection was raised by James Fitzjames Stephen who took issue with Mill’s scaremongering – for example:
Both teachers and learners go to sleep at their post, as soon as there is no enemy in the field. 2.29
Stephen argued with the logic of Mill’s claim that liberty of discussion is the best antidote:
The facts that whilst a doctrine is struggling for ascendency it is full of meaning, and that when it has become a received opinion its living power begins to decline, surely prove that coercion and not liberty is favourable to its appreciation. A ‘struggle for ascendency’ does not mean mere argument. It means reiterated and varied assertion persisted in, in the face of the wheel, the stake, and the gallows, as well as in the face of contradiction. ch 2
It is absurd to recommend persecution and torture for the sake of healthy opinions – but it seems to follow from Mill’s claim. Stephen is at least partly correct here: rationality is often not the strongest spring of action and free discussion may well, in this respect, not serve its purpose. Moreover, views that are not challenged can still be held passionately and with active conviction: think of fundamentalist beliefs in countries where heterodoxy is forbidden.
For Mill, there is never a case for censorship – even when the prevailing opinion is correct and any opposition is in error. Throughout his discussion it is clear that liberty, not truth, is the central concern: truth can easily become ‘dead dogma’ or ‘the shell and husk’ of meaning, and liberty is the only mechanism for recovering old truths and discovering new ones. Society must therefore promote free speech even, perhaps, at the expense of truth. Liberty, as it is throughout the essay, is the best means of serving ‘the permanent interests of man as a progressive being’ (1.11).