Britain’s War on Free Expression

The Daily Telegraph on October 18 reported that since 2018, dozens of British civil servants had been seconded to the BBC for periods of three to six months, during which they wrote news stories and advised on “high-impact, sensitive topics” such as Channel migrant crossings in small boats from France and the Covid pandemic. Their work appeared in stories published on the BBC News website, sometimes under the auspices of BBC Verify, supposedly an exemplar of rigorous standards whose stated mission is to “counter disinformation” and fact-check untruths. Their salaries were paid by the government whose record they were charged with checking. Nobody told the audience. 

Within two days of the story breaking, the BBC suspended the scheme, insisting that all editorial decisions are made by its own journalists. It remains the case that for eight years, the department responsible for immigration policy had staff inside the national broadcaster’s newsroom, “verifying” its coverage of immigration.

Another recent report from the Telegraph: One fine day last July, two police officers knocked at a house in the northeast of England. The woman who answered, Holly Passmore, is 44, the mother of two children, and a graduate student of Holocaust history at the University of Haifa. She had recently returned from a press trip to Israel, and it was this that the officers wished to discuss. They did not mention her health, although the document that had sent them to her door described her as having mental health problems and being “vulnerable to negative influence.” They asked why she was interested in “Jewish issues,” given that her neighborhood “is not a Jewish area,” and whether she had ever been to Auschwitz. They told her she might be “at risk of manipulation.”

Passmore had been referred to Prevent, the counterterrorism program under which the British state identifies people supposedly at risk of being drawn into extremist activity. The police record, which Passmore later obtained, stated that she “over-identified with the conflict between Israel and Palestine” and was “supportive of Israel to the point that she has traveled to Israel.” The document also conceded, in a sentence that ought to have ended the matter before it began, that “there is no CT [counterterrorist] concern with an individual being supportive of Israel as it’s a democracy.” When she complained about this incursion into her personal freedom by the state, the police advised her not to discuss the case online.

This state scrutiny of legal speech is now all too common in the UK, though this sort of intrusion would be deemed unconstitutional in the United States, thanks to the development of First Amendment protections, specifically the “chilling effect doctrine.”

Both incidents illustrate an attitude toward freedom of expression that would surely astonish most Americans: a baseline acceptance of state interference with free expression and state scrutiny of legal speech. In Passmore’s case, machinery built to disrupt jihadists was turned on a woman whose offense was to hold a lawful opinion with what a police officer determined was too much warmth. In the BBC’s case, the state did not merely censor the news, but helped to write it. Both, it should be noted, were supposedly justified by a bureaucratic language of protection: the safeguarding of a “vulnerable” adult; the fight against “disinformation.” Unfortunately, they exemplify a trend that is intensifying, with that vocabulary now being used to underpin other current and future measures, which threaten to install the most draconian regime of censorship seen in my country for almost 400 years. 

I work for the Free Speech Union, a nonprofit founded in London in 2020 to defend the speech rights of its members and to campaign for free expression more broadly. When the current Labour government took office in July 2024 we had 14,000 members. We now have almost 51,000, an increase of more than 250 percent in a little over two years. Some of that growth reflects nothing more sinister than a growing awareness that my organization exists. But it also reflects a perception, now widely shared, that free speech in this country is increasingly under assault.

Since opening, we’ve taken more than 7,000 cases involving individuals, including a few celebrities. Graham Linehan, the writer of some of the best-loved television comedies Britain has produced, stepped off a transatlantic flight at Heathrow in September 2025 to find five armed officers waiting to arrest him over three posts on X about transgender rights; the shock sent his blood pressure so high that he was taken to hospital in a blue-light ambulance. (The Crown Prosecution Service did not charge him, and, thanks to the FSU, he has been paid compensation for his unlawful ordeal.) A couple in Hertfordshire, one a radio producer, were arrested in front of their children and held for eleven hours over messages criticizing their daughter’s school in a parents’ WhatsApp chat group. No charges followed; the police apologized and, again thanks to help from the FSU, paid them £20,000. Most of those we help are not famous at all: nurses, teachers, students, people reported to human resources in corporations or public sector organizations for the heresy of the month.

It’s widely known that Britain lacks a written constitution, but it does possess a free press older than the United States. It also has Article 10 of the European Convention on Human Rights, incorporated into UK law by the Human Rights Act 1998. But this is nothing like as strong a free-speech shield as America’s First Amendment. It is hedged with caveats and exceptions, and the way the law has evolved means the courts must “balance” it against other, supposedly competing rights when it is invoked (for example, the right to privacy).

This means that criminal offenses whose definitions would be struck down in America, such as the vague and flexible offense of “stirring up” racial hatred, are enshrined in UK law. And according to figures released in September by the advocacy group Big Brother Watch, based on Freedom of Information Act disclosures, more than 62,000 people have been arrested here for speech and communications offenses over the past five years.

A dense thicket of statutes makes this possible. Section 127 of the Communications Act 2003 makes it an offense to send a “grossly offensive” message over a public electronic network. Drafted with threatening or abusive phone calls in mind, it is now often used to try to criminalize posts on social media. The Malicious Communications Act 1988 covers “indecent” or “grossly offensive” messages sent with alleged intent to cause distress. Arrests and prosecutions under these laws do not require evidence of an immediate physical threat, incitement, or a falsehood. 

The Public Order Act 1986 is the statute that outlaws “stirring up” hate, an offense punishable by up to seven years in prison. It was under this Act that Lucy Connolly, a childminder and mother, was charged for a post on X on the day in July 2024 when three little girls were murdered at a dance class in Southport. Mistakenly believing that the killer was a recent illegal migrant, she suggested that the hotels housing such individuals at public expense should be set alight. She went nowhere near a hotel, was not alleged to have incited a successful, real-world attack, and deleted her post within hours—but was still sentenced to 31 months in prison. 

Hamit Coşkun, a Turkish-born refugee opposed to the Islamist government of Recep Tayyip Erdoğan, burned a copy of the Quran outside the Turkish consulate in London as an act of political protest. He was stabbed by a passer-by who objected to his actions, but it was Coşkun who was charged and then convicted of a “religiously aggravated” public order offense. The FSU supported him, and eventually an appellate court quashed the conviction, observing that the criminal law does not exist to protect people from being upset. This was a victory. But there have since been other attempts to recreate a British blasphemy law, 18 years after Parliament abolished it.

Then there is the most distinctively British invention of all: the Non-Crime Hate Incident (NCHI), a police record, held in your name on a national database, of something you said that broke no law whatsoever, but was logged because someone perceived it to be motivated by hostility toward a protected group. More than a quarter of a million such incidents have been recorded in England and Wales over the past 10 years. One was recorded against a man for whistling the theme tune of a children’s cartoon, because a neighbor decided it sounded racist; another against a former home secretary for giving a speech. NCHIs may not send you to prison, but they can wreck your career, because they may surface in background checks when you apply for a job. In January, Home Secretary Shabana Mahmood announced that they would be scrapped in England and Wales. But they are, she said, to be replaced by a “new legal framework,” and the fate of the existing quarter-million records is unresolved. Meanwhile, the police in Scotland and Northern Ireland, who do not fall within the UK Home Secretary’s jurisdiction, intend to carry on recording them.

The Prevent program, too, has strayed well beyond its purpose. Its training materials have listed “cultural nationalism,” the belief that Western culture is threatened by mass migration, as a warning sign of extremism: a mainstream opinion, held by millions, cataloged as a step on the road to terror. We at the FSU are assisting a health-service employee referred for attending a lawful rally alongside 120,000 others, and a schoolboy referred for saying he wanted a proscribed terrorist group destroyed.

Speech at protests has narrowed too. In July 2025 the government proscribed Palestine Action, a direct-action group that had caused millions of pounds of damage to military aircraft, as a terrorist organization, although it appears to have little in common with more obvious terror groups such as Al Qaeda or Hamas, and it has never launched a fatal attack. The Free Speech Union does not defend vandalism. But proscription makes it a terrorism offense merely to express support for the group, and at one London demonstration some 890 people were arrested in a day, most for doing nothing more threatening than standing peacefully in the street holding placards. Some were elderly; others members of the clergy. To criminalize the expression of support, as distinct from the acts supported, is to punish opinion.

For the online world, a larger machine is now being built. The Online Safety Act 2023 hands sweeping powers to Ofcom, the government communications regulator. Platforms must not only remove content Ofcom identifies as illegal; they must also go looking for content they themselves judge might be unlawful, on pain of high fines. The Act does contain sections that are supposed to protect free expression of democratic importance and legitimate journalism, yet they have not been brought into force. Meanwhile, the government is about to bar everyone under the age of 16 from social media. This means that everyone else will have to provide proof of their identity and age—bringing state surveillance into every smartphone. 

All of this was in place before Andy Burnham became prime minister in July. His new government proposes to go considerably further, and here the vocabulary of protection reaches its apogee. Culture Secretary Lisa Nandy’s recent policy paper, Watch this space, recommends compelling platforms to prioritize showing “regulated public service media,” so that outlets such as the BBC are (in her words introducing the paper) “seen and heard in the fierce battle against mis- and disinformation.” The game is zero-sum: Promoting the BBC demotes everyone else. Nandy also stated recently that the social media ban for under-16s was only a “staging post” toward “a very active interventionist state, policing and regulating the online space.” Her colleague Lucy Powell, Labour’s deputy leader, has said she wants platforms brought under “some of the same requirements” of accuracy and balance as state-licensed broadcasters during election campaigns. And following amendments to the Online Safety Act earlier this year, Ofcom has told platforms to draw up protocols for restricting content during a “crisis” such as the protests in several cities after the child murders in Southport.

Underpinning these proposals is a 136-page “evidence review” of mis- and disinformation commissioned by the Government Office for Science from a team at Sheffield University and published in May. It concluded that online falsehood is “a systemic national risk,” recommended statutory “quality signals” in content ranking, a state-linked “database of false and misleading narratives,” and official “prebunking” in schools, warning that debate about what misinformation actually is “should not become a distraction from action.” Its authors did not attempt to measure how much falsehood exists online, or whether it had much impact; by their own account they relied on AI-generated summaries of 228 papers. Its “expert reviewers” included the chief executive of the Global Disinformation Index, a lobby group that pressures advertisers to defund publishers it dislikes. Among its victims is the website UnHerd, which dares to publish gender-critical writers, and  which drew this group’s ire during the Covid pandemic for granting space to what it claimed was a “conspiracy theory”—the notion that the virus may have emerged from a poorly guarded lab in Wuhan, China. (Needless to say, that conspiracy theory is now widely accepted as the likeliest explanation for the pandemic by scientists and intelligence agencies such as the CIA.) 

What the review omitted was the large body of research published in the world’s leading journals that contradicts it. For example, a 2024 paper in Nature found exposure to false or extreme content online to be low, and concentrated among a motivated fringe who were not fed it by malevolent social media algorithms but took steps to seek it out. According to this and other studies, “fake news” accounts for roughly 0.15 percent of the average media diet. A 2023 experiment in Science that gave tens of thousands of Facebook and Instagram users a chronological feed instead of an algorithmic one for three months found this made no difference to their political attitudes. 

On September 22, at the United Nations General Assembly, Burnham announced his intention to create a new National Centre for Information Defence. Its stated target would be hostile-state information warfare, especially from Russia, which he said is driving Britons to the far right. His new organ, he promised, would be a bulwark for “community cohesion” against a “distorted and untrue narrative about Britain.” 

A few weeks later, he blamed Russian misinformation for the result of the Brexit referendum—a claim made widely back in 2016, which has been comprehensively demolished, not least by an exhaustive study by the Oxford Internet Institute. Nobody in his government has yet said where hostile-state activity ends and domestic dissent begins, or what the government’s plan to “disrupt” it means when the narrative in question is being generated by British citizens. Put all this together—ranking rules that privilege approved outlets; a state database of forbidden narratives; emergency protocols for restricting content; and an intelligence-led center to disrupt “untrue” stories—and you have the most far-reaching apparatus for the control of peacetime speech that this country has seen since the English press ceased to be licensed in 1695. 

Americans, when they think of British tyranny, think of George III: stamps, tea, soldiers quartered in private houses. But that quarrel was about representation and taxation; nobody in Boston or Philadelphia had to get a license in order to print. For that, you have to go back a further century and a half. In July 1637 the Star Chamber of Charles I decreed that nothing might be published without a license from approved authorities, with enforcement delegated to a private body, the Stationers’ Company, much as Ofcom now delegates the policing of the internet to the platforms. The categories of the forbidden, “seditious,” “scandalous,” “offensive,” meant whatever the licensers, following orders from the king, wished them to mean. 

The proposals of 2026 are not an exact echo, but like the regime set up in 1637 they would police content by controlling the medium through which it is disseminated: exalted legal status for approved publishers; emergency powers to take down “harmful” narratives; official programs to instruct the people on which voices to trust; and a national center to disrupt unwelcome narratives. 

It’s worth recalling that Charles I’s decree did not work. Sedition spread, we had a civil war, and he was beheaded twelve years later. Meanwhile, in 1644 John Milton inaugurated England’s free speech tradition with the publication of his brilliant Areopagitica. His answer to the censors and licensers still stands: “Let her and Falsehood grapple; who ever knew Truth put to the worse, in a free and open encounter?” The founders of the United States wrote into the Bill of Rights the principles that Milton had championed. The country that produced him is rejecting them, in the name of preserving public safety and what it calls “social cohesion.” 

American readers should watch what Britain does next—not as a spectacle but as a warning. For the tools being built on these islands are portable, and the justifications being provided for them will sound, in every language, eminently reasonable.