29 September 2026 · Ethics and the State

Mill and the Limits of Liberty

John Stuart Mill published On Liberty in 1859. He had been raised from childhood by his father and by Jeremy Bentham himself to think in strictly utilitarian terms, and by the time he wrote this book he still called himself a utilitarian, but a notably more careful one. Where Bentham wanted one simple measure, pleasure against pain, Mill insists throughout this essay that the real question is not just how much happiness an action produces, but who has the right to decide how a person lives their own life in the first place.

Two maxims and the one principle behind them

Mill opens by naming the actual subject of the book: not the old philosophical dispute about free will, but civil liberty, meaning the limits society can rightly place on an individual. He traces a long history behind this question, from the old struggle between subjects and rulers, where liberty meant protection against a ruler's power, through the rise of elected government, which many people assumed would end the problem since rulers and ruled would finally share one interest. Mill pushes back on that comfortable assumption. Even in a democracy, he points out, the will of the people in practice just means the will of the majority, and a majority can be just as harsh on a minority as a king can, sometimes more so, because social pressure reaches further into private life than any law does. He calls this the tyranny of the majority, and argues that protecting individuals from it is at least as important as protecting them from government officials.

From this concern, Mill states the entire argument of the book in two maxims. First, a person is not accountable to society for actions that concern only themselves. Second, a person is accountable, and may rightly face social or legal punishment, for actions that harm others. He sharpens this into what is usually called the harm principle: the only purpose for which power can rightfully be used against a member of a civilized community, against their will, is to prevent harm to others. Their own good, physical or moral, is not sufficient grounds for compelling them, even though it is good reason to reason with them, persuade them, or argue with them. Mill is explicit that he is not grounding this in some separate idea of natural rights, the way Locke does. He grounds it in utility, but, he says, utility understood in the largest sense, tied to the permanent interests of a human being as a growing, developing creature, not just whatever produces the most pleasure in the moment.

Why silencing an opinion is always a loss

Chapter two turns to one specific application Mill treats as foundational: freedom of thought and of speaking or publishing that thought. He argues that if the entire human race agreed on one opinion except for a single dissenter, the world would have no more right to silence that one person than that person, if they had the power, would have to silence everyone else. His case rests on walking through every possibility for what a silenced opinion could actually be.

If the opinion being suppressed might be true, then silencing it assumes the people doing the silencing cannot be wrong, when history shows governments and churches have confidently crushed opinions later recognized as correct. Mill reaches for two extreme examples to make this point land: the trial and execution of Socrates, condemned by his own society as a corruptor of youth, and the death of Jesus, executed as a blasphemer by officials who were not unusually cruel people but ordinary men enforcing what they sincerely believed was right. He even uses the Roman emperor Marcus Aurelius, a genuinely thoughtful and humane ruler by Mill's own account, who nonetheless allowed the persecution of Christians because he sincerely judged the new religion a threat to the empire's stability. If someone that careful and that good could get it so wrong, Mill argues, no one alive today should feel too confident in their own certainty.

If the opinion is false, suppressing it is still a loss, because received opinions are rarely the complete truth, and only a clash with an opposing view keeps the fuller truth alive. And even if a received opinion is entirely true, letting it go unchallenged turns it into a dead formula rather than a living conviction. Mill argues most people who inherit a true belief without ever having to defend it do not actually understand the reasons behind it, so the belief sits in their mind as an unexamined custom rather than something they could explain or apply with real judgment. He compares this to knowing geometry only by memorizing conclusions rather than understanding proofs. A truth not actively defended, in his phrase, risks surviving only as the husk of an idea with the living meaning gone out of it.

Working the principle through real cases

By the time Mill reaches chapter five, titled Applications, he treats the two maxims as established, and spends the chapter testing how they play out in specific, often tricky situations, less to extend the theory than to show its limits in practice. Trade, he says, is fundamentally a social act, since selling something to the public affects other people, so restrictions on trade are not restrictions on liberty in the strict sense. They are simply judged by whether they actually work, which is the separate question of free trade economics.

He works through a long series of edge cases. Selling poison raises a real question, since poison has legitimate uses as well as lethal ones, so Mill proposes requiring clear labeling or a witnessed purchase record rather than banning the substance outright, a middle path he borrows the phrase preappointed evidence for, from Bentham. On drunkenness, he argues ordinary drinking is nobody's business but the drinker's, but once a person has already committed a violent act while drunk, society may reasonably restrict their future drinking specifically, since by then it has shown itself connected to harm done to others. He takes a similar line on idleness: being lazy is not punishable in itself, but failing to support one's own children because of that laziness is a real harm to others and can rightly be compelled.

He spends real care on harder cases like gambling houses, where the act itself, gambling between consenting adults, is tolerated, but someone making a living by encouraging others into it introduces a profit motive Mill admits troubles the clean logic of his own principle, and he says plainly that he will not try to resolve exactly where the line falls. He treats taxing stimulants like alcohol as legitimate, since government needs tax revenue from somewhere and taxing something already judged harmful in excess is more defensible than most alternatives, so long as the tax is not so steep it amounts to a disguised prohibition.

Two applications get more extended treatment. One is education. Mill calls it close to self evident that the state should require education for every child, since bringing a human being into the world without any provision for training their mind is, in his words, a moral crime against both the child and society. But he draws a sharp line between the state requiring that children be educated, which he fully supports, and the state itself controlling what gets taught, which he opposes just as firmly, because a single state run curriculum becomes a tool for molding everyone to one pattern, whoever happens to hold power. His proposed solution is compulsory public examinations testing basic, factual knowledge, leaving parents and private institutions free to decide how children actually get there, specifically avoiding any test of opinion on religion or politics.

The other is the broader question of when government should actively do something for people rather than just leave them alone. Mill lists three distinct objections to government stepping in, even where no question of liberty is involved. First, individuals usually handle their own business better than officials would. Second, letting people run their own local affairs, juries, municipal government, voluntary associations, is valuable training in itself, the practical education of a free citizen, even when a government department could technically do the job as well. Third, and most serious to him, every function added to government swells its power and turns ambitious people into hangers on of the state, risking the kind of entrenched, self perpetuating bureaucracy he describes in Russia, where he says even the Czar cannot truly govern against the bureaucracy's quiet resistance. He contrasts this with what he sees in America, where ordinary citizens are so practiced at organizing their own affairs that they could rebuild a functioning government from scratch if the existing one vanished overnight. His proposed balance is to centralize information and let local bodies keep the actual power to act, so a government informs and coordinates without absorbing the work and the judgment that free citizens need to practice for themselves.

Why this still matters, and how it bends Bentham's idea

What makes Mill worth reading right after Bentham is watching him work from inside the same utilitarian framework and still end up somewhere noticeably different. Bentham wanted a single calculation applied uniformly to any action. Mill keeps utility as his ultimate standard, but insists it has to be judged over the whole, long run development of a person as a thinking, growing being, not simply whichever choice produces the most immediate pleasure. That shift is what lets him draw a firm line between conduct that only concerns oneself, where he thinks society has no real business interfering even if it disapproves, and conduct that harms others, where interference is justified. It is a distinction Bentham's original framework does not obviously require, and it is the argument later readers return to most, whenever the question comes up of how far a government, or a majority of one's own neighbors, should be allowed to reach into a person's private life.

Source: John Stuart Mill, On Liberty (1859), Chapter 1 (Introductory), Chapter 2 (Of the Liberty of Thought and Discussion), and Chapter 5 (Applications).