Whenever liberalism is attacked today, John Stuart Mill's name will almost certainly be mentioned. Often indeed the conservative and radical critics of liberalism have seen in Mill's essay On Liberty [On Liberty in Utilitarism, Liberty, Representative Government (Everyman edn). All subsequent references to On Liberty and Utilitarism are to this edition] the embodiment of all the liberal errors and vices they wish to expose. Thus when Wilmoore Kendall speaks of the fallacies of the open society, it is on Mill's alleged fallacies that he dwells. Similarly, when Robert Paul Wolff attacks the poverty of liberalism, he devotes a great deal of his time to laying bare the purported confusions, inconsistencies, and gross inadequacies of Mill's essay. Like Mill's critics, I too regard this essay as the most eloquent expression of the liberal theory of the open society. But unlike them I am generally sympathetic to his values and I have tried to expound his case for liberty as clearly and fully as I can. The foundations Mill provides for his liberal theory have some faults, but a careful study of the essay will reveal that these are often quite different from those which conservative and radical critics of his have been inclined to stress.

Different interpretations of the essay have been given. Many of these are conflicting and yet each is, taken on its own, often quite plausible. It is therefore necessary not just to give a plausible account but also to test it against alternative interpretations. I have built my exposition of Mill on an examination of various views of his work. This approach would have won the approval of Mill himself. In the essay he claims that in complicated subjects like morals, religion, and politics, “three-fourths of the arguments for every disputed opinion consist in dispelling the appearances which favour some opinion different from it.” (96.) But there is also the danger of cluttering one's text with too many details and of being side-tracked by the arguments of others. I have therefore had to ignore some interesting papers, although wherever possible I have at the appropriate places referred [1/2] to them in footnotes, and sometimes also added brief comments there.


The essay On Liberty is concerned with the limits of the coercive power which the state and society may legitimately exercise over the individual. In the past, Mill says, the stuggle for individual liberty was waged against tyrannical governments. This battle was won with the establishment of democracy where the government was responsible to, and removable by, the people. It was then believed that the government could be trusted with power as it was the people who dictated the use to which such power was to be put. But this belief proved to be false. In a democracy the people who had power were the majority, and the liberty of the rest of society was still not secure against “the tyranny of the majority” and against the power which a government, subservient to current majority view, might exercise. Individual liberty in such a situation was threatened both by oppressive laws and by the use of extra-legal means to impose the prevailing views and practices on everyone. This extra-legal coercion was far more pervasive and insidious than the use of legal penalities because “it leaves fewer means of escape, penetrating much more deeply into the details of life, and enslaving the soul itself.” (68.)

To defend individual liberty a limit must be set to the extent of legitimate social intervention both by the state and by society. This limit should not depend on the “likings and dislikings” of the dominant group in society. In this respect most of those who held enlightened views had been at fault even when they were defending the freedom of dissentients. For they did not question the right of society to impose what it liked, and to suppress what it disliked. Instead they merely addressed themselves to the question of what society ought to like or dislike. They were interested in substituting a set of more enlightened views and practices for less enlightened ones, rather than in challenging the principle that one group in society had the right to impose its tastes and standards, however cultivated, on all the rest.

There is no recognized principle defining the proper limits [2/3] of social intervention with the freedom of individuals, and Mill put forward his “one very simple principle” to remedy this.

That principle is, that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His 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. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him, or visiting him with any evil in case he do otherwise. To justify that, the conduct from which it is desired to deter him must be calculated to produce evil to some one else. The only part of the conduct of any one, for which he is amenable to society, is that which concerns others. In the part which merely concerns himself, his independence is, of right, absolute. Over himself, over his own body and mind, the individual is sovereign. [72-73]

Mill explains the scope of his liberty principle. It applies only to “human beings in the maturity of their faculties” and not to children, or to “those backward states of society in which the race itself may be considered in its nonage”. In these “backward states” people are not “capable of being improved by free and equal discussion”. The test, then, is one of capability rather than willingness. It is also a test that is applied to ascertain the general level of society and not to particular individuals. This is presumably because Mill believes that where people in a society generally have the required capacity, those who lack it can aquire it through their interaction with others in an atmosphere of freedom. it is also evident that the level of the capacity is not pitched very high because Mill asserts that it has been reached long ago by “all nations with whom we need here concern ourselves”.

Mill's liberty principle specifies that where the individual is “sovereign”, the type of intervention in his conduct that is unacceptable involves the use of punishment and compulsion. He is careful to point out that argument, persuasion and other methods of non-coercive influence are permitted. Indeed later in the essay he pleads for an increase in the exercise of such non-coercive influences, and repeats his condemnation [3/4] of the use of “whips and scourges, either of the literal or metaphorical sort” (132). Throughout this book I shall use the terms “intervention” and “interference” to mean “coercive intervention” and “coercive interference”.

Mill's principle invokes a distinction between conduct which “merely concerns” the individual himself and conduct which “concerns others”. This distinction is commonly restated as that between self-regarding and other-regarding actions. The term “self-regarding” is used by Mill himself although he does not employ the term “other-regarding”. His claim is that society is never justified in interfering with self-regarding conduct. But society may be justified in interfering with other-regarding actions which fall within the limits of social intervention. From this it does not follow that intervention in such acts is always justified. For example, the harm caused by intervention may sometimes be greater than the harm to others inflicted by other-regarding actions.


How does Mill defend his liberty principle? He states that his case for liberty is not based on “the idea of abstract right, as a thing independent of utility”. “I regard utility as the ultimate appeal on all ethical questions; but it must be utility in the largest sense grounded on the permanent interests of man as a progressive being.” [On Liberty, 74] The nature of Mill's ultimate value, “utility in the largest sense”, is the subject of much dispute. The main bone of contention is over whether, or to what extent, Mill is a consistent utilitarian.

Utilitarianism is a consequentialist doctrine in that it judges the rightness or wrongness of an act solely in terms of its good or bad consequences when compared with those of alternative acts. The desirable or good consequences of the act consist in the production of utility or happiness. The happiness here is that of all those affected by the act. A right act is thus that which among all the available alternatives maximizes utility or happiness. This standard is often referred to as the principle of utility. Classical utilitarianism, as propounded by Jeremy Bentham and others, conceives of happiness hedonistically as pleasure or the balance of pleasure over pain, and pleasure is regarded as a psychological or [4/5] mental state. Classical utilitarianism can be distinguished from preference utilitarianism in which the notion of maximizing utility or happiness is reinterpreted as, or replaced by, the notion of maximizing the satisfaction of preferences or desires5. Maximizing the satisfaction of desires involves taking into account both the number of desires satisfied and their intensity. When a person gets whatever he wants his desires are satisfied. But it does not follow that he experiences a pleasurable sensation. He may want others to inflict some pain or discomfort on him. The notion of desire is also used in a wide sense such that a person's desire can be satisfied without his knowledge or even after his death. If I desire to be cremated when I die, then my desire is satisfied if I am cremated even though I shall not have any experience when it happens [further discussion of the differences and similarities between classical utilitarianism and preference utilitarianism in Chapter 4]

Both classical and preference utilitarianism are versions of act utilitarianism in which each act is assessed by the utilitarian standard of maximizing happiness or utility. They are to be distinguished from rule utilitarianism in which the utilitarian standard is not applied directly to particular acts but to rules or institutions. In rule utilitarianism the right act is that which falls under the best rule. A particular right act may not maximize utility, but it belongs to a class of acts, which, if generally performed, will maximize utility. Unless otherwise stated I shall use the term 'utilitarianism' throughout this book to refer to act utilitarianism in both the classical and preference versions.

The question then is whether Mill's defence of liberty is consistently utilitarian. I shall argue that there are significant non-utilitarian elements in his case for liberty, and I shall try to identify some of these. A crude, and somewhat inaccurate, way of describing Mill's position is to say that he was a straightforward utilitarian at the level of other-regarding actions, but that he was not a utilitarian at the level of self-regarding conduct. This is inaccurate for several reasons. First, the distinction between self- and other-regarding conduct is problematic, and I shall argue that Mill's defence of liberty does not depend on there being two different areas of a person's conduct which have very different effects. His case depends on distinguishing between different reasons for interfering with the individual's conduct in any area. Certain [5/6] reasons are always ruled out as irrelevant, but there is one reason, the prevention of harm to others, which is always relevant [further discussion of this point in Chapter 2 and Chapter 4]. Secondly, even when interference is designed to prevent harm to others, Mill's concern is not simply to minimize the harm (or maximize the prevention of harm). He is also interested in the way the harm is spread out or distributed between different people, and the distributive principle he gestures towards seems to be independent of utilitarianism. A utilitarian is ultimately only interested in maximizing happiness or, where appropriate, minimizing harm and suffering. The way in which the happiness or suffering is distributed is only of secondary importance: it is important only in so far as it affects the total. An unequal distribution is not intrinsically worse than an equal one.

So there is no simple and accurate description of Mill's position. But the crude account will serve as a rough guide which will be qualified and refined as we proceed.


In Chapter 2 I discuss Mill's notion of self-regarding conduct. According to the traditional interpretation, self-regarding conduct does not affect others at all except with their consent. But this interpretation is mistaken for Mill readily concedes that self-regarding conduct has certain adverse effects on others. His argument is that a principled defence of individual liberty will lead us to discount these effects on others. For example, other people may be affected by my conduct because they dislike it, find it disgusting, or regard it as immoral. These effects, taken in themselves, are never good reasons for interfering with my conduct. In thus ignoring certain effects, Mill's defence of liberty is not utilitarian. A utilitarian cannot disregard any of the effects of my conduct since they are all part of its consequences, and help to determine whether the suppression of my conduct or leaving me free will maximize happiness. Various attempts to show that a sophisticated utilitarian can consistently discount certain pleasures and pains, or the satisfaction or frustration of certain desires, are considered and rejected.

Recent studies of Mill suggest that his version of the principle or utility is not a moral principle but a principle for [6/7] evaluating all conduct, both moral and non-moral. On this view, only conduct which breaches moral standards can be subjected to sanctions or punishment by society. On the other hand, a person's faults in the self-regarding sphere are not properly regarded as immoral, and so should not be interfered with. It is argued that Mill's moral theory, being only a branch of his principle of utility, does not always require us to act to maximize happiness. Some acts which fail to maximize happiness are not morally wrong. in Chapter 3 I provide some background to the understanding of these recent studies, and I then examine one such interpretation of the limited scope of Mill's moral theory. I argue that if there is a conflict between the maximization of happiness and some moral requirement, a consistent utilitarian would have to resolve it in favour of the former. He cannot allow moral considerations to override non-moral ones. On the other hand, there may be no possibility of a conflict between moral and non-moral appraisals because morally required acts are a sub-class of acts which maximize happiness. But assuming again that self-regarding conduct does not belong to the moral realm, it does not follow that such conduct is immune from the application of sanctions, For a utilitarian sanctions are applicable whenever it will maximize happiness to apply them, irrespective of whether the conduct to which sanctions are applied is morally right or wrong or even non-moral. So if Mill is still a utilitarian, even though his moral theory is restricted in scope, then the area of application of sanctions cannot be confined in principle to the moral sphere. Hiving off self-regarding conduct to a non-moral realm does not guarantee that on utilitarian grounds it should never be interfered with.

Mill's liberty principle states that the only purpose for which interference with the conduct of individuals is permitted is to “prevent harm to others”. In Chapter 4 I suggest that Mill's notion of harm is narrower than that which a consistent utilitarian would accept. The basis of it is the idea of infringing the rules which are necessary for the viability of any society, but Mill complicates the notion in several ways. I then discuss the range of conduct that Mill brings within the scope of his formula that justifiable interference with the [7/8] conduct of individuals must “prevent harm to others”.

In Chapter 5 I analyse Mill's ideal of individuality which lies at the basis of his defence of freedom of action and freedom of discussion. Freedom is an ingredient of individuality in that individuality consists partly in choosing freely and critically between alternative beliefs and activities or plans of life. But the content of the choice is also important. It should be such as to develop certain distinctive human capacities and powers, as well as each person's special potentialities. I also discuss the connection between happiness and individuality, and argue that Mill's plea for individuality cannot be usefully described as utilitarian. The chapter ends with an examination of the extent to which Mill's doctrine of individuality is consistent with a belief in objective truths.

In his liberty principle Mill highlights certain bad reasons for interfering with a person's conduct: “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.” Mill strongly rejects paternalism and the enforcement of the dominant shared values of a society. For this he has been severely criticized. In Chaper 6 I show that Mill recognized the function of shared values in promoting social stability, and that his view here is consistent with the importance he attached to individual liberty. Chapter 7 is devoted to a discussion of different versions of paternalism and different justifications of it. Particular emphasis is given to some of Mill's examples, and to the implications of changes in a person's attitude towards his past conduct.

In Chapter 8 I turn to Mill's defence of freedom of expression. This is the subject of the longest, and for many the finest, chapter in On Liberty. I distinguish between different types of argument presented by Mill, and show that he attaches the greatest significance to the manner in which a person holds his beliefs in the important areas of “morals, religion, politics, social relations, and the business of life”. The chapter also examines the sorts of restrictions on freedom of expression that are compatible with Mill's arguments. I claim that though there are grounds for restricting the circumstances in which blasphemous, obscene, and racist [8/9] remarks may be expressed, they should not be suppressed, simply because they offend some people, be they a small minority or even the overwhelming majority.

In the final chapter I consider some unusual attacks on Mill's essay. It is generally accepted, both by Mill's supporters and by his critics, that On Liberty is a passionate defence of the liberty of every group in society and not just of the privileged few. But recently it is claimed that Mill's liberalism is really a defence of the liberty of the “the superior few”, and that he was intolerant of the others towards whom he showed signs of “moral totalitarianism”. it is also argued from another quarter that although On Liberty is a liberal tract defending the liberty of individuals generally, it is inconsistent with most of Mill's other works. The views that are supposed to be expressed in these other works are then used to attack the Mill of On Liberty, so that we are presented with a picture of “the two Mills” with the Mill of On Liberty as the clear villain. Chapter 9 is a defence of Mill from both these attacks.

In this book I have tried to show that Mill's case for liberty is not wholly reconcilable with any consistent version of utilitarianism. I do not thereby suggest that Mill is an inconsistent thinker. On the whole On Liberty is both internally consistent and also consistent with nearly all of Mill's other writings. But it is as a consistent liberal, deeply committed to the cause of individual freedom for everyone, that Mill should be remembered.

Last modified 1 May 2015