Censorship and free expression

Model introduction (~90 words):

Few principles are defended as passionately as the claim that censorship is never justified, yet almost everyone accepts that material inciting violence or harming children should be restrained. The word "never", then, deserves real scrutiny. By "censorship" I mean preventing books, films, art or online content from being published or shown, and the crucial test I will apply is whether the material causes genuine harm or merely offence. This essay argues that censorship is rarely justified — and never to suppress dissent or mere offence — but that it can be justified, narrowly, to prevent demonstrable harm.

Bullet-point plan:

• Body 1 (mostly AGAINST): State censorship is easily abused to hide truth and protect power — China's "Great Firewall" and the suppression of the 1989 Tiananmen Square events. Link: this is censorship at its most harmful.
• Body 2 (AGAINST, on art): Offence is no basis for a ban — Salman Rushdie's "The Satanic Verses" was banned and its author attacked; silencing art by threat rewards the violent. Link: artistic freedom must be broad.
• Body 3 (concession): Yet almost everyone accepts limits on incitement and material harming children, and classification (age ratings) guides rather than bans. Link: the issue is WHERE the line falls.
• Counterargument: Some say even offensive art should be restrained to protect communities — but offence is not harm, and there is no right not to be offended.
• Conclusion: Censorship is justified only to prevent genuine, demonstrable harm — so "never" is too strong, but the principle's warning is sound.

Far from making censorship impossible, the internet has in some respects given states more powerful tools to control what their citizens can see. (Point) China's "Great Firewall" is the clearest example: it blocks many foreign websites and platforms and filters search results, so that references to sensitive events such as the 1989 Tiananmen Square protests are systematically removed from the Chinese internet. (Example) This matters because it shows that the same technology which spreads information can also be turned into an instrument of control: rather than chasing individual books or newspapers, a government can filter an entire population's access at the source, quietly and at scale. (Analysis) It is true that determined users can sometimes circumvent such controls, and that information does leak across borders, so the control is never total. (Analysis) Nevertheless, the existence of a sophisticated national filtering system shows that the internet has not abolished censorship but transformed it — making it, if anything, more centralised and harder for ordinary citizens to escape. (Link)

Every day, a handful of private companies make decisions that shape what billions of people may say in public. The question of how far social-media platforms should decide what their users can post is therefore one of the defining free-expression debates of our age. By "decide" I mean the power to remove content or accounts — content moderation and "deplatforming". This essay argues that platforms must remove clearly illegal and genuinely harmful material, but that their power to police merely offensive speech should be limited, transparent and accountable.

There is a strong case for platforms exercising real control. They host material that can cause direct harm — incitement to violence, terrorist propaganda, child abuse imagery and coordinated disinformation — and a platform that refused to remove such content would be enabling serious damage. Because this material spreads instantly and globally, only the platforms themselves are positioned to act quickly enough. In this sense, moderation is not censorship of ideas but the removal of demonstrable harm, and most users accept that some such limits are necessary for the spaces to be usable at all.

However, the danger is that unelected companies become the arbiters of legitimate public debate. The controversy over "deplatforming" prominent political figures showed how a private decision to remove an account can shape national conversation, yet such decisions are made by firms answerable mainly to shareholders, under rules that are inconsistent and rarely open to appeal. When a platform removes lawful but unpopular speech, it is exercising a form of private censorship with none of the accountability we would demand of a government. The same power used to remove genuine harm can, in other hands or under commercial pressure, suppress dissent.

It might be objected that platforms are private property and may set whatever rules they like, just as a newspaper chooses what to print, so users who dislike the rules can simply leave. There is some force in this: no one is entitled to a megaphone owned by someone else. Yet the analogy understates how dominant the largest platforms have become; for many people they are the public square, not one option among many, so leaving is not a real choice. The greater their reach, the stronger the case that their power over speech should be bounded by transparent rules and independent appeal, rather than left to private discretion alone.

The banning of books has a long and troubling history, and the claim that no book should ever be banned has obvious appeal to anyone who values free thought. By "banned" I mean legally prohibited or forcibly removed from circulation, as opposed to merely criticised or age-restricted. This essay argues that the presumption against banning books should be very strong, and that offence can never justify a ban, though a narrow category of genuinely harmful material may be the rare exception.

The strongest argument against banning books is that the power to ban is so easily abused. Salman Rushdie's novel "The Satanic Verses", published in 1988, was banned in several countries and publicly burned, and Rushdie lived for decades under threat, being violently attacked in 2022. Whatever one's view of the book's content, the episode shows that once a book can be suppressed because it offends, the suppression tends to escalate, and a veto over literature passes to whoever is most offended — or most willing to use violence. Banning books also rarely works: prohibition often increases a work's notoriety and readership rather than erasing it.

A further danger is that book-banning is frequently driven by social or political pressure rather than any genuine harm. The school book-removal controversies in parts of the United States, where particular titles have been challenged and withdrawn from school libraries, show how communities can act as censors, restricting what young people may read on the basis of contested values. Removing a book because some find its themes uncomfortable is a long way from preventing real harm, and it narrows the range of ideas a society is willing to confront.

Against this, it could be argued that a few categories of material — detailed instructions for making weapons, or content that exploits children — are so dangerous that prohibition is plainly justified, and that an absolute rule of "never ban" would protect even these. This objection has real force, and it shows why "never" is too strong a word. Yet it actually supports the central argument rather than undermining it: the legitimate cases turn on demonstrable HARM, not on offence or discomfort. The line should be drawn at content that directly endangers people, not at books that merely challenge, provoke or upset — so the presumption against banning remains, with only the narrowest of exceptions.

The proposition that censorship is never justified in a free society is attractive to anyone who values liberty, but its absolutism is also its weakness. By "censorship" I mean the prevention of books, films, art or online content from being published or shown, and by a "free society" one that protects open debate and dissent. The decisive question is not whether expression should be free — of course it should — but where the line falls between offence, which a free society must tolerate, and harm, which it may restrain. This essay argues that the word "never" cannot be defended: censorship is rarely justified and never to suppress dissent or mere offence, but it can be justified narrowly to prevent genuine, demonstrable harm.

There is a powerful case against almost all censorship, because the power to censor is so easily abused. China's "Great Firewall" blocks foreign platforms and filters search results, and references to the 1989 Tiananmen Square events are systematically erased from the Chinese internet. This is censorship in its most dangerous form: not the protection of citizens but the protection of power, achieved by hiding inconvenient truths from an entire population. The danger is structural — once a state may ban what it dislikes, the same machinery silences criticism — which is precisely why a free society treats censorship with deep suspicion.

The case is just as strong in the arts, where offence is no basis for suppression. Salman Rushdie's "The Satanic Verses" was banned in several countries and burned, and its author was violently attacked decades later. Whatever one thinks of the book, allowing it to be silenced by outrage and threat would hand a veto over all literature to whoever is most offended or most violent. In a free society there is no right not to be offended; ideas we dislike remain permitted, because the alternative is to let the most easily provoked decide what everyone else may read. On this reasoning, the proposition is largely wise.

Yet an absolute "never" is untenable, because not all restraint is the suppression of ideas. Almost everyone accepts that incitement to violence, targeted abuse, and material that exploits children should be restricted, and few would call such limits an attack on freedom. Crucially, classification is not censorship at all: age ratings on films and games GUIDE audiences and protect children without preventing adults from choosing freely. To say a free society may "never" restrict any expression would forbid even these narrow, widely accepted measures — a position almost no one truly holds.

It could be objected that any exception is a dangerous opening, since governments will always claim their censorship targets "harm" rather than dissent, exactly as authoritarian states dress up control as protection. This is a serious point, and the slide of states towards ever-broader online filtering shows how readily "safety" becomes a pretext. But the answer is not to abandon the distinction between harm and offence; it is to police it strictly — through transparent laws, independent courts and a free press — so that restraint is confined to demonstrable harm and never extended to mere disagreement. A free society defends itself not by pretending no limit is ever legitimate, but by guarding the few that are.

In conclusion, I agree with the proposition only in its underlying warning, not in its wording. Censorship is never justified to suppress dissent or to spare people from offence, for that path leads to the very control that destroys a free society. But it can be justified, narrowly and accountably, to prevent genuine harm such as incitement or the exploitation of children. The right principle is therefore not "never", but "only to prevent demonstrable harm, and never merely because ideas offend".

Art has always provoked, and the question of how far artistic freedom should be limited is as old as art itself. By "artistic freedom" I mean the liberty of writers, film-makers, musicians, painters and satirists to create and publish their work, and by "limits" the restriction of that work by law, by platforms or by social pressure. The central difficulty is that art is meant to disturb, so the test cannot be whether it offends. This essay argues that artistic freedom should be very broad — including the freedom to offend — and limited only where the work causes genuine, demonstrable harm rather than mere offence.

The case for broad artistic freedom is strong, because art that cannot offend cannot challenge. Much great work has shocked its first audience, and satire in particular exists precisely to puncture the powerful and the sacred. Salman Rushdie's "The Satanic Verses" was banned in several countries and burned, and its author was violently attacked decades after publication. The lesson is sobering: if a novel can be silenced by outrage and threat, then a veto over all art passes to whoever is most willing to be violent. A society that limits art whenever someone is offended will end by permitting only the bland and the safe, which is no freedom at all.

This is reinforced by the distinction between offence and harm. To be offended is to be upset by an idea or image; to be harmed is to be materially damaged — through incitement to violence, defamation or targeted abuse. In a free society there is no right not to be offended, because ideas we dislike are still ideas we must be free to encounter. The 2015 attack on the satirical magazine Charlie Hebdo, in which staff were murdered over cartoons, shows the catastrophic cost of treating offence as something that may be answered with force; the proper response to art one hates is argument or avoidance, not suppression.

Nevertheless, artistic freedom cannot be wholly without limits. Few would defend art that consists of genuine incitement to violence against a group, or that exploits children, and almost no society treats these as protected expression. Here the restraint is justified not because the work offends but because it harms. It is also worth distinguishing limits from classification: age ratings on films guide audiences and protect children while leaving adults free to choose, and this is guidance, not a ban — so much of what is loosely called "limiting art" is in fact a mild and defensible form of regulation.

Against this, it could be argued that some offence shades into real harm: that deliberately vilifying a vulnerable minority can entrench prejudice and even provoke violence, so that protecting communities sometimes justifies restraining offensive art. This objection deserves to be taken seriously, and it explains why the line between offence and harm is genuinely contested rather than obvious. Yet the remedy is to apply the harm test rigorously — asking whether a work truly incites damage, not merely whether it upsets — rather than to lower the bar to offence. If discomfort alone justified limits, almost all challenging art would be vulnerable, and the most easily provoked would govern what everyone may create.

In conclusion, artistic freedom should be limited only to a small extent. It must remain broad enough to include the freedom to offend, because offence is the price of art that genuinely challenges, and because silencing work by threat rewards the violent. The legitimate limits are narrow and turn on demonstrable harm — incitement, abuse, the exploitation of the vulnerable — supplemented by classification that guides rather than forbids. The right answer to the question, then, is "only slightly, and only at the point where expression stops being offensive and starts being genuinely harmful".