Liberty — Mill and Berlin

Few ideas are invoked more often, or analysed more loosely, than liberty. This note pairs the two texts that frame the Oxford debate: John Stuart Mill's On Liberty (1859) and Isaiah Berlin's lecture 'Two Concepts of Liberty' (1958). Between them they ask the two questions that organise the subject — where are the limits of legitimate coercion? (Mill) and what do we even mean by freedom? (Berlin).

Mill defends a single principle: that the only justification for coercing an individual, against their will, is to prevent harm to others — the harm principle. A person's own good, physical or moral, is never a sufficient warrant. From this Mill builds a powerful case for freedom of thought and discussion, for 'experiments of living', and against the 'tyranny of the majority' in both its legal and its social forms.

Berlin steps back to ask what 'freedom' means, and distinguishes two senses. Negative liberty is freedom from interference — the area within which one is left to act unobstructed. Positive liberty is freedom to be one's own master — self-direction, self-realisation. Berlin's warning is that the positive concept, however attractive, has historically been twisted into a justification for coercion: once a 'higher' or 'real' self is posited, people can be 'forced to be free'. Beneath this lies his deeper thesis of value pluralism — that liberty is one genuine value among several that can conflict and cannot all be maximised together.

This note reconstructs the harm principle and Mill's defence of it, sets out Berlin's two concepts and the pluralism behind them, and surveys the major criticisms and refinements — Stephen's attack on Mill, MacCallum's claim that the negative/positive dichotomy is confused, and the republican 'third concept' of liberty as non-domination. It connects directly to your jurisprudence note on the Hart–Devlin debate, where Mill's harm principle is the liberal pole, and to Rawls's priority of liberty.

Mill writes from within, but also against, the utilitarian tradition he inherited from Bentham and his father James Mill. On Liberty grounds freedom not on any natural or abstract right — Mill is explicit that he forgoes 'any advantage which could be derived to my argument from the idea of abstract right, as a thing independent of utility' — but on 'utility in the largest sense, grounded on the permanent interests of man as a progressive being'. The phrase matters: Mill's utilitarianism is enriched by a Romantic concern, drawn from Coleridge, Wilhelm von Humboldt, and his own near-breakdown, with individuality and self-development as constituents of a flourishing life. His immediate target is less the state than society: the 'tyranny of the majority' — a phrase he takes from Tocqueville — which coerces through custom and opinion as much as through law.

Berlin writes a century later, in a very different climate. 'Two Concepts of Liberty' was his inaugural lecture as Chichele Professor of Social and Political Theory at Oxford in 1958, delivered at the height of the Cold War and published in Four Essays on Liberty (1969). Its animating worry is the way emancipatory ideals — Rousseau's general will, Hegelian and Marxist notions of true freedom — had, in Berlin's lifetime, underwritten totalitarian coercion in the name of liberation. Berlin's response is not to dismiss positive liberty but to expose how it can be perverted, and to set against all such monisms his pluralist conviction that there is no single harmonious system of values into which every genuine good can be fitted. Mill and Berlin thus approach liberty from opposite ends — Mill drawing a boundary around coercion, Berlin clarifying the concept itself — but both are liberals warning against the same enemy: the confident majority, or the confident reformer, who knows better than you what is good for you.

Mill's harm principle

The core of On Liberty is what Mill calls 'one very simple principle':

'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.'

Two consequences follow. The principle is anti-paternalist: the state may not coerce me for my own good. And it is anti-moralist: the mere fact that others find my conduct immoral or distasteful is no ground for interference. Coercion needs a victim, not an offended onlooker. As Mill puts it, 'Over himself, over his own body and mind, the individual is sovereign.'

Self-regarding and other-regarding conduct

The principle turns on a distinction between conduct that concerns only the agent ('self-regarding') and conduct that affects others ('other-regarding'). Only the latter falls within society's legitimate authority. Mill does not naïvely deny that almost any action may have some remote effect on others; his claim is that where conduct affects others' interests only through the agent's effect on themselves — or through others' mere disapproval — it remains self-regarding and immune from coercion. The integrity of this line is the single most contested feature of the whole argument.

Why liberty matters: discussion and individuality

Mill defends two great applications. In freedom of thought and discussion (ch. 2), even a true belief held without the discipline of challenge degenerates into 'dead dogma'; suppressing an opinion robs the human race, 'those who dissent from the opinion, still more than those who hold it'. In individuality (ch. 3), he argues that people should be free to conduct 'experiments of living', because individuality is not a threat to well-being but an essential ingredient of it; the great danger of modern society is the flattening pressure of custom and conformity.

Berlin's two concepts of liberty

Berlin distinguishes two questions that 'liberty' can answer. The negative sense answers 'What is the area within which the subject is left to do or be what he is able to do or be, without interference by others?' — liberty as the absence of deliberate obstruction by other people. The positive sense answers 'What, or who, is the source of control that can determine someone to do this rather than that?' — liberty as self-mastery, deriving from 'the wish on the part of the individual to be his own master'.

The two are not mere restatements of one idea. Negative liberty is about the extent of the unobstructed area; positive liberty is about who is in charge. One could enjoy wide negative liberty under a benign autocrat who simply leaves people alone, and little of it under a democracy that regulates minutely — which is why Berlin insists that the negative question ('how far am I governed?') must not be confused with the question of self-government ('who governs me?').

Value pluralism

Beneath the two concepts lies Berlin's deepest commitment: value pluralism. There is, he holds, an irreducible plurality of genuine human values — liberty, equality, justice, security, solidarity — which can conflict, which are not reducible to a single super-value, and between which there is sometimes no rationally determinable trade-off. Liberty is therefore one good among many, not an absolute that always wins. The opposite view, monism — that all true values form one harmonious system with a single right answer to every question — is, for Berlin, the intellectual root of the coercive utopianism he most fears.

J.S. Mill, On Liberty (1859)

The single most influential statement of the liberal case for freedom, and a set text you should know chapter by chapter. Ch. 1 states the harm principle and the self-regarding/other-regarding distinction. Ch. 2, on liberty of thought and discussion, gives the famous four-part defence of free speech (the suppressed opinion may be true, so to silence it assumes our own infallibility; it may be partly true, the remainder supplied only by the 'collision of adverse opinions'; and, even where the received view is the whole truth, an uncontested belief decays into a prejudice held as 'dead dogma, not a living truth', its very meaning enfeebled). Ch. 3 defends individuality and 'experiments of living'. Ch. 4 marks the limits of society's authority over the individual. Ch. 5 works through applications — trade, the sale of poisons, and the striking case of a contract to sell oneself into slavery, which Mill holds society may not enforce, because 'the principle of freedom cannot require that he should be free not to be free'.

A point of scholarship worth a sentence: Mill records in his Autobiography that the essay was jointly conceived with Harriet Taylor Mill, and he dedicated it to her memory; she died in 1858, the year before publication.

Isaiah Berlin, 'Two Concepts of Liberty' (1958)

Delivered as Berlin's inaugural lecture at Oxford in 1958 and published in Four Essays on Liberty (Oxford University Press, 1969); now most accessible in the collection Liberty (ed. Henry Hardy, OUP, 2002). The lecture sets out the negative and positive concepts, traces the historical perversion of the positive concept into 'forcing men to be free', and closes with the statement of value pluralism. Read the closing pages for the pluralist conclusion as carefully as the famous central distinction — examiners reward students who see that the two concepts rest on the pluralism, not the other way round.

1. Mill on free discussion: the argument from fallibility and 'dead dogma'

Mill's defence of free speech runs through four claims. A silenced opinion may be true, and to deny this is to assume our own infallibility. It may be partly true, and since prevailing opinion is rarely the whole truth, the clash of views is how the missing part is supplied. Even a wholly false opinion is useful, because a true belief that is never contested is held 'in the manner of a prejudice', as 'dead dogma, not a living truth'. His most quoted formulation: 'If all mankind minus one were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind.'

2. Experiments of living

Against the tyranny of custom, Mill argues that individuals must be free to test different ways of life. Just as there is no one true opinion fixed for all time, there is no one mode of living suited to all; 'experiments of living' let human beings discover and develop the forms of flourishing that fit them. The argument is broadly consequentialist — diversity of character is a condition of social progress — but it carries a Romantic charge about self-development that sits in some tension with Mill's official utilitarianism.

3. Berlin: the perversion of positive liberty

Berlin's central argument is historical and conceptual. The positive idea of self-mastery invites a division of the self into a 'higher', rational, 'real' self and a 'lower', empirical, appetitive self. Once that move is made, it becomes possible to say that coercing a person's empirical self — overriding their actual wishes — liberates their real self, which 'truly' wills what is rational and good. Rousseau's formula that a citizen who is compelled to obey the general will is thereby 'forced to be free' is the hinge; Hegelian and Marxist variants follow. The danger is that a vanguard who claim to know the real self's true ends can present tyranny as emancipation. Berlin does not say positive liberty is this perversion — only that it is peculiarly liable to it.

4. 'Freedom for the wolves'

Berlin is equally clear that negative liberty is not a sufficient social ideal on its own. Unlimited freedom from interference, in conditions of gross inequality, is freedom for the strong to dominate the weak: 'Freedom for the wolves has often meant death to the sheep.' Negative liberty must therefore be weighed against equality and welfare — a concession that defeats any reading of Berlin as a simple libertarian, and reconnects his argument to the pluralism that frames it.

MacCallum: freedom is triadic

The most important refinement of Berlin came from Gerald MacCallum, 'Negative and Positive Freedom' ((1967) 76 Philosophical Review 312). MacCallum argues that Berlin's dichotomy is confused, because all freedom is a single triadic relation: 'x is free from y to do or become z' — an agent is free from some constraint to do or be something. Negative and positive 'concepts' are not two kinds of freedom but two emphases within one concept: disputes apparently about which freedom matters are really disputes about what counts as an agent (the empirical self or a 'true' self), what counts as a constraint (only external obstacles, or also internal ones such as ignorance or addiction), and what range of actions is in view. MacCallum's analysis is now the standard starting point, and a good answer uses it to reframe the Berlin debate rather than simply repeating the two-concepts story.

Taylor: the value of an 'exercise' concept

Charles Taylor, 'What's Wrong with Negative Liberty' (1979), defends what the positive tradition was getting at. A purely negative, 'opportunity' concept of freedom — freedom as the sheer absence of external obstacles — cannot explain why we think a person mastered by fear, addiction, or false consciousness is unfree in an important sense. Freedom, Taylor argues, is partly an 'exercise concept': it involves actually directing one's life according to one's significant purposes, so that internal obstacles can diminish freedom as surely as external ones. The point is not to license coercion but to show that Berlin's suspicion of positive liberty throws out something real.

Republican liberty: a third concept

A further tradition, revived by Quentin Skinner ('A Third Concept of Liberty', 2002) and Philip Pettit (Republicanism, 1997), argues that both Berlinian concepts miss the older republican idea of freedom as non-domination — not merely non-interference, but the absence of any power of arbitrary interference over you. A slave with a kind master suffers no interference yet is plainly unfree, because they remain subject to the master's will. On this view negative liberty is too thin: what matters is not whether you are actually interfered with, but whether you stand in anyone's power.

Stephen against Mill

The classic contemporary attack on On Liberty came from the judge James Fitzjames Stephen in Liberty, Equality, Fraternity (1873). Stephen argued that the harm principle is far too permissive: coercion for moral and paternalistic ends is both legitimate and indispensable, since law inevitably rests on a shared morality and rightly enforces it. The dispute prefigures the twentieth-century Hart–Devlin debate almost exactly, with Devlin in Stephen's role and Hart defending a modified Millian line — which is why this note pairs with your jurisprudence reading on the enforcement of morals.

The self-regarding / other-regarding line

The most pressing internal difficulty is whether Mill's central distinction can bear its weight. Critics from Stephen onwards object that scarcely any conduct is purely self-regarding: the drunkard's family, the drug-user's dependants, the example set to others, the cumulative effect of private vices on the public culture — all seem to give society a foothold. Defenders (notably C.L. Ten and John Gray) reply that Mill's line is drawn not by causal effect but by the kind of interest affected: mere offence, moral disapproval, or harm mediated through the agent's effect on themselves does not count as the relevant 'harm to others'. Whether this reconstruction rescues the principle or quietly narrows it is a live question.

Is negative liberty enough? Berlin's monism worry

The debate around Berlin runs in two directions. Some (Taylor, the republicans) press that negative liberty is too thin — that it cannot account for internal unfreedom or for domination without interference. Others ask whether Berlin's own preference for negative liberty is consistent with his pluralism: if values are genuinely plural and incommensurable, on what basis does Berlin lean toward the negative concept? Berlin's answer is that negative liberty, precisely because it makes no grand claims about the self's true ends, is the safer bet against the coercive utopian — but this is a prudential, pluralist argument, not a proof that negative liberty is the highest value, and critics press the tension.

Pluralism versus monism

Finally, Berlin's value pluralism is itself contested. Monists (and some liberals, such as Ronald Dworkin in his later work) argue that genuine values cannot ultimately conflict, and that Berlin's pluralism risks collapsing into relativism. Berlin denies the charge: plural values are objectively real, and choice between them is tragic but not arbitrary. The status of pluralism — liberating insight or self-undermining doctrine — remains one of the richest questions the lecture leaves open.

Mill and Berlin. Mill is, in Berlin's terms, largely a theorist of negative liberty — the harm principle draws a protected area around the individual — yet his stress on individuality and self-development has a clear positive colouring, which is why Berlin admired Mill while gently noting that Mill's case for liberty rests on a particular (developmental) ideal of the person. The sharper contrast is methodological: Mill is a monist about value (everything ultimately answers to utility, broadly construed), whereas Berlin's whole position rests on pluralism. A Millian can ask 'how much liberty maximises well-being?'; for Berlin that question mis-describes a tragic choice between incommensurable goods.

Against Rawls. Rawls's first principle gives lexical priority to a scheme of basic liberties, which looks Millian, but Rawls grounds it in the original position rather than in utility, and the basic liberties are a specified list rather than Mill's open-ended sphere of self-regarding conduct. Where Mill would protect any self-regarding act, Rawls protects the particular liberties needed for citizens to develop and exercise their moral powers.

Against republicanism. The republican concept of freedom as non-domination (Skinner, Pettit) cuts across both Mill and Berlin: it agrees with Mill that the enemy is arbitrary power, but locates unfreedom in subjection to a master's will rather than in actual interference, and so can count the contented slave or the indulged dependant as unfree where a purely negative account cannot.

Against Devlin. In the jurisprudence debate, Mill's harm principle is the liberal benchmark against which Devlin's legal moralism is measured; Stephen's Victorian attack on Mill and Devlin's twentieth-century attack on the Wolfenden Report are, in substance, the same objection a century apart.

Question: 'Positive liberty is more dangerous than negative liberty.' Discuss.

Introduction

The claim is Berlin's, and it is more guarded than the slogan suggests. Berlin does not say that positive liberty is bad, nor that negative liberty is innocent; he says that the positive concept has historically proved more liable to perversion into a justification for coercion. This essay reconstructs why Berlin thinks so, tests the claim against the abuses of negative liberty and against MacCallum's argument that the dichotomy is confused, and concludes that Berlin identifies a real pathology but mislocates it: the danger lies not in positive liberty as such but in its combination with monism and a doctrine of the 'real self'. Properly understood, neither concept is intrinsically more dangerous; what is dangerous is the conviction that one knows another person's true ends.

Berlin's argument

Positive liberty answers the question 'who is my own master?'. Berlin's worry is the move it invites: to be self-governed is to be governed by one's rational or true self, which can then be opposed to one's actual, empirical desires. Once that split is made, coercing a person's empirical self can be redescribed as liberating their real self — Rousseau's citizen 'forced to be free', and its Hegelian and Marxist descendants. A party or planner who claims privileged knowledge of the real self's ends can thus present domination as emancipation. Negative liberty, by contrast, makes no claims about the self's true ends: it simply marks out an area to be left alone, and so offers the would-be tyrant no comparable foothold. That asymmetry is the heart of Berlin's case.

The case against the claim

Three replies tell against the strong version.

First, negative liberty has its own pathologies. Berlin himself concedes that 'freedom for the wolves has often meant death to the sheep': a regime of pure non-interference, against a background of gross inequality, licenses the domination of the weak by the strong and can rationalise indifference to suffering as respect for liberty. The laissez-faire cruelties of the nineteenth century were committed in the name of negative freedom. If we are counting historical abuses, negative liberty has a long charge sheet of its own.

Second, MacCallum's analysis dissolves the contrast. If all freedom is the triadic relation 'agent free from constraint to do or be something', then 'negative' and 'positive' are not two freedoms but two emphases — disagreements about what counts as a relevant agent and a relevant constraint. On this view it is incoherent to call one concept more dangerous than the other; what can be dangerous is a particular theory of the agent (the 'real self') or of constraint. Berlin's target, properly identified, is that theory, not the positive concept.

Third, the abuses trade on monism, not on positive liberty alone. The slide to 'forcing men to be free' requires not just the idea of self-mastery but the further, monist assumption that there is a single rational good that the true self must will. Berlin's own deeper diagnosis — that the root evil is monism — supports this: a pluralist positive liberty, which accepts that people may freely master themselves toward different and conflicting goods, generates no totalitarian conclusion. Taylor presses the complementary point that the positive tradition captures something real — that addiction, fear, and false consciousness are genuine internal obstacles to freedom — which a purely negative account cannot express, so that discarding positive liberty altogether would be a loss.

Judgment

Berlin is right that there is a characteristic route by which ideals of self-realisation have been turned into instruments of coercion, and a student should be able to trace it. But the claim that positive liberty is intrinsically more dangerous does not survive scrutiny. Negative liberty has underwritten its own abuses; MacCallum shows the two are aspects of one concept; and the totalitarian danger depends on monism and the 'real self' doctrine rather than on the positive idea itself. The defensible residue of Berlin's thesis is therefore a warning, not a ranking: any conception of freedom that licenses someone to override your actual choices in the name of your 'true' interests should be treated with the deepest suspicion — and that warning applies wherever the move is made, under whichever concept's banner.

1. Equating positive liberty with totalitarianism. Berlin says the positive concept is liable to perversion, not that it is coercion. He grants that self-mastery is a genuine and valuable idea; the danger lies in the 'real self' move, not in positive liberty as such.

2. Treating negative liberty as Berlin's unqualified ideal. He explicitly warns that 'freedom for the wolves has often meant death to the sheep' and that negative liberty must be weighed against equality and welfare. Berlin is a pluralist, not a libertarian.

3. Grounding Mill's harm principle in natural rights. Mill expressly forgoes 'any advantage which could be derived… from the idea of abstract right', resting the principle on 'utility in the largest sense'. Saying Mill defends a natural right to liberty gets him backwards.

4. Forgetting that Mill's chief target is social, not legal, tyranny. The 'tyranny of the majority' includes the coercion of custom and opinion, not only the law. Much of On Liberty is about social pressure.

5. Ignoring MacCallum. The strongest scripts use MacCallum's triadic analysis ('x is free from y to do z') to reframe the negative/positive debate as a dispute about agents and constraints, rather than retelling the two-concepts story uncritically.

6. Overlooking the republican 'third concept'. Skinner and Pettit's freedom as non-domination is a distinct option: the contented slave is unfree because subject to arbitrary power, though not interfered with. Mention it to show range.

7. Missing that Berlin's two concepts rest on value pluralism. The lecture's deepest claim is the plurality and incommensurability of values; the two concepts are an application of it. Reducing the essay to 'two kinds of freedom' misses its centre.

The practice questions for this topic — with model answers and mark schemes across foundation, standard, and challenge tiers — are set out below. Attempt each under timed conditions before revealing the model answer, then mark your work against the scheme.

The primary texts and the leading critical literature are listed below. Read On Liberty (chs. 1–3) and 'Two Concepts of Liberty' first, then MacCallum to reframe the debate and Pettit for the republican alternative.