SUMMARY
- British police forces recorded at least 62,199 arrests under three communications offences between 1 January 2021 and 31 December 2025, according to a report from the UK organiation Big Brother Watch.
- The same research found 18,520 charges, while separate Ministry of Justice data recorded 12,292 convictions in England and Wales; the datasets cannot be treated as a single case-by-case pipeline.
- Enforcement varied wildly by location: Cumbria recorded 25.7 arrests per 10,000 residents, compared with 1.9 in neighbouring Northumbria.
- The report says Britain’s vague communications laws, combined with the Online Safety Act and wider controls on lawful online content, are producing a chilling effect on speech.
British police forces made at least 62,199 arrests under communications laws in five years, an average of roughly 34 every day, according to a new 104-page report examining the country’s expanding restrictions on speech online and offline.
The civil-liberties group Big Brother Watch counted arrests between 1 January 2021 and 31 December 2025 under Section 1 of the Malicious Communications Act 1988, Section 127 of the Communications Act 2003 and, after it came into force, Section 179 of the Online Safety Act 2023. Several of the United Kingdom’s 45 territorial police forces supplied incomplete data or none at all. If the missing forces arrested people at the national average rate, the report estimates that the five-year total would rise to roughly 73,000.
The findings turn last year’s row over more than 12,000 arrests in 2023 into a longer and more complete picture. Big Brother Watch recorded 18,520 charges over the same five-year period, meaning police made approximately 3.4 arrests for every charge. Separate Ministry of Justice figures showed 12,292 convictions in England and Wales for the three offences. Those numbers are not a tracked funnel from arrest to conviction: they come from different sources, cover different territorial scopes and are affected by the time cases take to reach court. Even so, the large gap between arrests and charges raises an obvious question about how often detention was genuinely necessary.
The gap was especially pronounced in some areas. Sussex Police made 10.7 arrests for every charge, while Thames Valley Police made approximately 9.4. Northamptonshire recorded 23.4 arrests per 10,000 residents but only 2.7 charges per 10,000. At the other end of the table, Staffordshire reported just 0.6 arrests per 10,000.
The regional divide was equally striking. Cumbria Constabulary recorded 25.7 arrests per 10,000 residents, compared with 1.9 in neighbouring Northumbria. Big Brother Watch called the result a “postcode lottery” in which the practical boundary of lawful expression depends heavily on the police force receiving the complaint.
That matters because British police are not required to arrest somebody merely because an allegation appears credible. PACE Code G says a lawful arrest requires reasonable grounds to suspect involvement in an offence and reasonable grounds to believe arrest is necessary. Officers are also directed to consider less intrusive means, including a voluntary interview.
The laws captured by the report are not limited to social-media opinions. They can cover genuine threats, obscene material, persistent nuisance communications and conduct connected to harassment or domestic abuse. That distinction is important. The figure does not mean that every arrest involved an otherwise harmless political post. It does, however, measure police use of statutes whose central terms leave unusually broad room for interpretation.
Section 127 of the Communications Act 2003 criminalizes messages sent over a public electronic communications network that are “grossly offensive,” indecent, obscene or menacing, as well as persistent use of a network intended to cause annoyance, inconvenience or needless anxiety. Section 1 of the Malicious Communications Act 1988 covers indecent or grossly offensive letters, electronic communications and other articles sent with the purpose of causing distress or anxiety.
The Online Safety Act added a false communications offence from 31 January 2024. It applies when somebody sends information they know to be false without reasonable excuse, intending to cause non-trivial psychological or physical harm to a likely audience. The Act simultaneously repealed older provisions dealing with false and threatening messages, while leaving the “grossly offensive” limbs of the earlier statutes in place.
Even the Crown Prosecution Service’s current guidance acknowledges the danger. There is no statutory definition of “grossly offensive,” and prosecutors must consider context and the right to freedom of expression. The guidance says a case should amount to more than material that is merely offensive, shocking, satirical, rude, unpopular or ill-thought-out. In practice, however, those safeguards often arrive after police have already knocked on a door, seized a device or made an arrest.
The government has defended the framework on the grounds that communications cases frequently overlap with threats, harassment, sexual offending, hate crime and domestic abuse. During a House of Lords debate in July 2025, ministers argued that police must investigate suspected offences independently and that freedom of speech does not excuse criminal conduct. That is a legitimate warning against treating every entry in the dataset as a free-speech martyr. It does not explain why neighboring forces use the same powers at radically different rates, or why some forces arrest so many more people than they charge.
The report places the statistics beside cases showing how elastic laws become punitive long before conviction. In the infamous “Twitter Joke Trial,” Paul Chambers was prosecuted after joking in 2010 that he would blow an airport “sky high” if it remained closed. Airport staff did not regard the post as a credible threat, but Chambers was convicted under Section 127. The High Court quashed his conviction in 2012.
More recently, wildlife presenter and animal-rights campaigner Anneka Svenska said Sussex Police visited her home in July 2026 over a video about a company breeding beagles for scientific research. Officers accused her of knowingly spreading false information under Section 179 and proposed a community resolution that included an “online thinking skills” course. According to Big Brother Watch, her solicitor supplied documents supporting the claim and the case was dropped shortly afterwards.
In another case that reached Parliament, six officers arrested Maxie Allen and Rosalind Levine after the parents criticized their daughter’s primary school in emails and a WhatsApp group. They were detained for hours and the investigation ended without charges. Hertfordshire’s chief constable later conceded that police could have achieved the same result differently. The incident is a stark example of arrest becoming the punishment even when no prosecution follows.
The report treats the arrest figures as one part of a broader British censorship system. Its separate examination of the Online Safety Act found lawful material being blocked or placed behind age checks, including political videos, discussions of BBC news reports, a complaint about a paper cut, 1970s punk music and reproductions of famous paintings. Small forums serving communities from single fathers to cycling enthusiasts have closed or restricted access rather than absorb the Act’s compliance burden.
The platform regime is legally distinct from the criminal offences in the arrest data, but both rely on the same precautionary logic: when the boundaries are vague and the penalties severe, the safest response is to restrict first. The government’s own Online Safety Act explainer states that Ofcom can fine companies up to £18 million or 10 percent of qualifying worldwide revenue and, with court approval, order payment providers, advertisers and internet providers to stop working with a service.
As FFO previously reported, those powers reach American platforms and their business partners whenever a service has relevant links with Britain. British speech regulation is therefore not a purely domestic matter. It can alter what US companies host, how they design their services and which users they require to identify themselves.
FFO has also documented how an “anti-trolling” campaign helped place identity-verification tools inside the Online Safety Act, allowing verified users to filter out anonymous accounts and potentially rewarding identified users with greater visibility. A separate FFO investigation showed how the government’s planned ban on social media for children under 16 would deepen demand for age assurance, digital identity and biometric estimation across ordinary online services.
US-Funded NGOs and Former Officials Shaped The UK’s Online Safety Act
The policy network is international as well. FFO found that US-funded NGOs and former American officials helped shape Britain’s Online Safety Act, while another FFO report examined a Welsh law requiring the prohibition of deliberate falsehoods in election speech. Britain’s communications arrests are therefore not an isolated quirk of old statutes. They sit inside a continuing political project to regulate who may speak online, under what identity, and within whose definition of truth or harm.
Big Brother Watch is now urging Prime Minister Andy Burnham’s government to commission an independent review of laws governing communications, online safety and protest. The call is backed by the Free Speech Union, Index on Censorship and other civil-liberties groups.
Nadine Dorries, the former culture secretary who led the Online Safety Act through the House of Commons, wrote the report’s foreword. She now says the law failed both to protect children adequately and to preserve free expression, describing Britain’s collection of online-safety laws as a “dog’s dinner.”
The case for review does not require pretending that threats, stalking or targeted abuse are protected speech. It begins with a simpler proposition: vague laws should not expose ordinary people to radically different risks of arrest depending on their postcode, and the state should not treat detention as a routine first response to words on a screen. Britain’s 62,199 recorded arrests show that this is no longer a theoretical argument about the edge cases of free expression. It is an industrial-scale policing practice.
