CAGE OF CONTROL
The Cage

Communication & Media Control

How the Cage of Control controls communication and media.

Control the conversation, control the people

Before a government can control the opposition, the courts or the economy, it has to control what people are allowed to say to each other. Communication is the oxygen of a free society. Cut it off, filter it, or make people afraid to use it, and everything else gets easier.

I'm not talking about the obvious dictatorships here. I'm talking about three countries that call themselves free democracies: Australia, the United Kingdom and France. Each one has built, piece by piece, a legal machine for watching, filtering and punishing speech. Each piece comes with a reasonable-sounding justification, usually children, safety or "misinformation." Put the pieces together and you can see the bars of the cage.

Australia: the eSafety state

Australia has gone further than almost any Western country in putting a single regulator in charge of what appears online.

The Online Safety Act 2021 and the eSafety Commissioner. The Act gives the eSafety Commissioner powers to order content removed from a wide range of online services, from social media platforms to games and websites.[1] In April 2024, after the stabbing of a religious leader at Wakeley in Sydney, eSafety issued a removal notice to X Corp over footage of the attack. eSafety sued in the Federal Court, then dropped that case in June 2024 and moved the fight to the Administrative Appeals Tribunal.[2] In October 2024 that case was settled by agreement, with no ruling on the merits.[3] In my view, the lesson is clear: an unelected Australian official tried to dictate what could be seen on a global platform, and it took a billionaire with deep pockets to push back.

The under-16 social media ban. Parliament passed the Online Safety Amendment (Social Media Minimum Age) Act in November 2024, and it received Royal Assent on 10 December 2024.[4] Since 10 December 2025, age-restricted platforms have had to take "reasonable steps" to stop Australians under 16 from creating or keeping accounts.[5] In September 2026, Parliament strengthened eSafety's enforcement powers again, including powers to compel documents from platforms and age-assurance providers and to require individuals to attend and answer questions.[5] Ask yourself how a platform proves someone isn't under 16. The answer, sooner or later, is that everyone has to prove who they are.

The misinformation bill. In 2024 the government introduced a bill to combat "misinformation and disinformation" that would have given the media regulator new powers over platforms. In November 2024 the Communications Minister announced it was being dropped because it had no path through the Senate.[6] It didn't fail because the government changed its mind. It failed because it ran out of votes, and ideas like that have a way of coming back.

Encryption and hacking powers. The Assistance and Access Act 2018 allows the Attorney-General to issue "technical capability notices" that require communications providers to build capabilities to help security and law enforcement agencies.[7] The Surveillance Legislation Amendment (Identify and Disrupt) Act 2021 went further, creating data disruption warrants, network activity warrants and account takeover warrants. Account takeover warrants let police take control of a person's online account.[8]

Metadata retention. Australia also requires telecommunications providers to keep customer metadata for law enforcement access. [Needs source: confirm the current statutory retention period and scope under the Telecommunications (Interception and Access) Act 1979 as amended in 2015.]

The United Kingdom: arrested for a post

If Australia is building the machine, Britain is already running it.

The laws

  • The Online Safety Act 2023. This law gave the regulator Ofcom sweeping duties to enforce against online platforms. From 25 July 2025, services likely to be used by children had to put "highly effective age assurance" in place for certain kinds of content.[9] In practice, millions of adults are now asked to prove their age, with ID or a face scan, to reach large parts of the internet.
  • Section 127 of the Communications Act 2003 and section 1 of the Malicious Communications Act 1988. These make it an offence to send messages that are "grossly offensive," "indecent, obscene or menacing," or sent to cause "annoyance, inconvenience or needless anxiety."[10]
  • The Public Order Act 1986. Its "stirring up" offences make it a crime to publish threatening, abusive or insulting material intended to stir up racial hatred. That is the offence behind several of the cases below.[11][12]
  • Non-crime hate incidents. For years police recorded "hate incidents" against people's names even when no crime had been committed. The think tank Policy Exchange estimated about 13,200 a year, consuming over 60,000 police hours.[13] London's Metropolitan Police said in October 2025 that it would stop investigating them.[14]
  • The Investigatory Powers Act 2016 ("the Snoopers' Charter"). In early 2025 the UK government reportedly used this law to serve Apple with a secret Technical Capability Notice demanding the ability to access users' encrypted iCloud data. Apple responded by withdrawing its Advanced Data Protection encryption from new UK users.[15][16] The original notice was later reported withdrawn and replaced with one aimed at British users, and Advanced Data Protection is still unavailable to new UK users.[16][17]

The numbers

In April 2025, The Times reported freedom-of-information data showing that police in England and Wales made 12,183 arrests in 2023 under section 127 and the Malicious Communications Act. That's about 33 a day, up from 7,734 in 2019. Not every force answered the requests, so the real figure is likely higher.[10][18] These laws cover a range of conduct, including menacing messages, so not every one of those arrests was over an opinion. But a system that lets you be arrested for being "grossly offensive" or causing "annoyance" is, in my view, a system built to punish speech.

UK arrests over online speech, by year

ARRESTS

Speech the government decides is criminal. Not threats. Not defamation.

UK arrests over online speech, by yearArrests under Communications Act 2003 s127 and Malicious Communications Act 1988 s1, England and Wales: 2017: 5,502; 2019: 7,734; 2023: 12,183. Other years not shown.04,0008,00012,0005,50220177,734201912,1832023Year (only years with published totals shown)Arrests

12,183

arrests in 2023 under s127 Communications Act 2003 + s1 Malicious Communications Act 1988 (England & Wales), up from 5,502 in 2017

Chart: Arrests by year. Figures from The Times (April 2025, freedom-of-information data): 5,502 in 2017, 7,734 in 2019 and 12,183 in 2023.[18][19] Only years with a published total are shown; other years are left out rather than estimated. Not all police forces responded (the 2023 total covers 37 forces), so the real numbers are likely higher. The figures cover communications offences broadly, including menacing messages, not only opinion.

In my view, these are arrests for speech the government decides is criminal. Not threats. Not defamation. The usual justification is "hate," but hate is a label the government defines for itself, and in practice I believe it gets used against speech that undermines the government's authority. The official figures do lump in genuine threats and menacing messages, and I don't defend those. But the cases below show how far beyond threats these laws reach.

And in my view, the arrest is the punishment, whether or not anyone is ever convicted. Being detained, having your phone taken, hearing the knock at the door and living with the fear that follows would shake anyone. Every one of those arrests also sends a message to everyone watching: be careful what you type. That is how speech gets chilled without a single trial.

How many investigations? How many knocks at the door?

Arrests are only the visible part. Here is what the published data shows about how often police get involved over speech, with each estimate labeled:

  • Online hate crimes. For the year to March 2026, the Home Office flagged 8,028 recorded hate crime offences, about 7% of the total, as having an online element. The Home Office itself says this is probably an undercount, and the figures come from 35 forces as "official statistics in development."[20]
  • Non-crime hate incidents, per year. The Times, using police data, reported more than 13,200 non-crime hate incidents recorded in the 12 months to June 2024.[21] Policy Exchange estimated that this took up more than 60,000 police hours a year.[13]
  • Non-crime hate incidents, all time. Freedom of information requests found 119,934 recorded by 34 forces between 2014 and 2019.[22] A 2024 Telegraph analysis put the total since 2014 at more than 133,000 across the 43 forces, about 13,000 a year. It estimated this at roughly 666,000 police hours, assuming up to five hours per incident (Policy Exchange's estimate).[23] The Free Speech Union, a campaign group, estimates more than 250,000 over the regime's 12-year history. That is the campaign's own estimate, not an official figure.[24] None of these figures separates online posts from offline incidents.
  • Home visits. There is no national data on how often police visit people's homes over what they've posted. No force or government body publishes it, and I found no freedom-of-information dataset that counts it. The best-known example is the Telegraph columnist Allison Pearson. Two Essex officers came to her door on Remembrance Sunday in November 2024 to invite her to a voluntary interview about a deleted post on X, which they were investigating under the Public Order Act.[25][21] A police review later concluded the force had been right to investigate.[25] Without official numbers, the real frequency of these visits is unknown.

Five cases

These cases are a matter of public record. In fairness, I've included what was actually said and what the courts actually found.

1. Lucy Connolly: 31 months. Hours after the Southport murders in July 2024, Connolly, a childminder, posted on X: "Mass deportation now, set fire to all the fucking hotels full of the bastards for all I care... If that makes me racist, so be it." She deleted it and apologized, saying she had acted on false information. She pleaded guilty to publishing material intended to stir up racial hatred and was sentenced to 31 months in prison.[26][11] The post was ugly. But it was words on a screen, and she went to prison for them. That is my point.

2. Tyler Kay: 38 months. Kay, a 26-year-old father of three, reshared Connolly's post and admitted publishing material intended to stir up racial hatred. He was sentenced to 38 months.[11] That is more than three years in prison, connected to sharing someone else's words.

3. Jordan Parlour: 20 months. During the August 2024 disorder, Parlour wrote on Facebook: "Every man and their dog should be smashing fuck out Britannia Hotel," a Leeds hotel housing asylum seekers. The judge accepted that he took no part in any violence, but found he was encouraging others. He pleaded guilty to stirring up racial hatred and got 20 months.[12][27] The courts called it incitement. I call it what it was: a post with six likes that led to nearly two years behind bars.

4. The Hertfordshire parents: six officers over a WhatsApp group. In January 2025, six officers arrived at the home of Maxie Allen and Rosalind Levine and arrested them in front of their young daughter. They had sent emails to their daughter's primary school and made critical comments in a parents' WhatsApp group. They were held for 11 hours on suspicion of harassment, malicious communications and causing a nuisance on school property. Police later took no further action, then admitted the arrest was unlawful and paid them £20,000.[28][29] Two parents criticized a school, and the state sent six officers.

5. Graham Linehan: arrested at Heathrow. In September 2025, the comedy writer was arrested by five airport officers as he stepped off a flight, over three posts on X. One said that if "a trans-identified male" was in a female-only space, women should "make a scene, call the cops and if all else fails, punch him in the balls." The Met said airport officers routinely carry firearms, and that none were drawn. He was bailed on condition that he stay off X, the case was downgraded to a non-crime hate incident, and no further action was taken. The Met later apologized and reached a settlement with him.[30][31][32] You can find his posts crude. But armed officers at an airport over three posts tells you what kind of country Britain is becoming.

Ordinary people, ordinary posts

The jail cases above involved posts the courts treated as incitement. These next cases are different. They are ordinary people who posted something mild, a joke or a meme, and ended up dealing with the police.

Julian Foulkes: a retired special constable, six officers. In October 2023, Foulkes, then 71, replied on X to an activist who was threatening to sue the Home Secretary for calling pro-Palestinian marches "hate marches." His reply read: "One step away from storming Heathrow looking for Jewish arrivals…" It was a warning about antisemitism. On 2 November 2023, six Kent officers came to his home, handcuffed him, searched the house, seized his devices and held him for eight hours on suspicion of malicious communications. He accepted a caution. Kent Police later concluded the caution "should not have been issued," removed it from his record, apologized, and paid him £20,000 plus legal costs.[33][34][35]

Darren Brady: an Army veteran and a meme. In 2022, Hampshire police arrested Brady, a 51-year-old Army veteran, at his home in Aldershot. He had reshared a meme showing Progress Pride flags arranged in the shape of a swastika. On video, an officer told him someone "has been caused anxiety based on your social media post." He was arrested on suspicion of an offence under section 127, released under investigation, and then told he faced no further action. The county's police and crime commissioner said she was concerned about the "proportionality and necessity" of the response.[36][37]

Harry Miller: "check your thinking." Miller, a former police officer, was never arrested, but his case shows how the system works. In January 2019, Humberside Police visited his workplace over his tweets about transgender issues, including a limerick he had retweeted, and recorded a non-crime hate incident. The High Court later ruled that "the claimant's tweets were lawful" and that the police response was "a disproportionate interference" with his freedom of expression because of its "potential chilling effect."[38][39]

The Hertfordshire parents. As described above, two parents were arrested in front of their child by six officers over emails and critical comments in a school WhatsApp group, and later received £20,000 for unlawful arrest.[28][29]

France: "pure bullshit"

France has a long record of trying to police speech online.

  • The 2018 law against the manipulation of information lets courts order the rapid removal of false information spread massively and deliberately during election campaigns. The Constitutional Council upheld it with limits.[40]
  • The Avia law (2020) would have forced platforms to remove certain content within 24 hours. The Constitutional Council struck down that core regime in June 2020 as a disproportionate restriction on freedom of expression.[41] That's one of the few times the brakes worked.
  • Pavel Durov. In August 2024, the founder of the messaging app Telegram was arrested at Le Bourget airport outside Paris. French judges placed him under formal investigation for alleged complicity in crimes committed through the app and for refusing to cooperate with authorities, and released him on €5 million bail with a ban on leaving France.[42][43] For the first time, a Western democracy had locked up the head of a major communications platform over what his users did.
  • An under-15 ban. In July 2026, the French parliament adopted a law banning children under 15 from social media. In August 2026 the Constitutional Council struck down the blanket ban as a disproportionate restriction on freedom of expression, and the government is pursuing a revised version.[44][45]

Then there is the president himself. Speaking in New Delhi on February 18, 2026, Emmanuel Macron attacked social media companies that defend themselves in the name of free speech: "Some of them claim to be in favor of free speech. We are in favor of free algorithms, totally transparent. Free speech is pure bullshit if nobody knows how you are guided through this."[46] To be accurate, his target was opaque platform algorithms, not free speech as such. But I think the words matter. When the leader of a major Western democracy puts "free speech" and "pure bullshit" in the same sentence, he is telling you how much weight he gives it. In his framing, speech is only "free" once the state is satisfied with the system that carries it.

What it means for the United States

Look at the pattern. A regulator with the power to order content removed. Age checks that end anonymity. Encryption backdoors. Laws that make "offensive" speech a crime. Police at the door over a WhatsApp group. A president who calls the free speech defense "pure bullshit."

None of these countries announced that they were ending free speech. They just kept adding bars, one reasonable-sounding law at a time. The First Amendment is the strongest protection on Earth, but it is only as strong as the people willing to defend it. The same arguments about safety, children, misinformation and hate are already being made here. Once the cage closes, it's closed forever.

Sources

  1. Australian Government (DITRDCSA), "Current legislation" (online safety) — https://www.infrastructure.gov.au/media-technology-communications/internet/online-safety/current-legislation
  2. eSafety Commissioner, "Statement re: Federal Court proceedings" (June 2024) — https://www.esafety.gov.au/newsroom/media-releases/statement-from-the-esafety-commissioner-re-federal-court-proceedings
  3. eSafety Commissioner, "Administrative Appeals Tribunal orders" (Oct 2024) — https://www.esafety.gov.au/newsroom/media-releases/esafety-statement-administrative-appeals-tribunal-orders
  4. Parliament of Australia, Online Safety Amendment (Social Media Minimum Age) Bill 2024 — https://www.aph.gov.au/Parliamentary_Business/Bills_Legislation/Bills_Search_Results/Result?bId=r7284
  5. eSafety Commissioner, "Social media age restrictions" — https://www.esafety.gov.au/about-us/industry-regulation/social-media-age-restrictions
  6. Minister for Communications, media release on the Combatting Misinformation and Disinformation Bill 2024 — https://minister.infrastructure.gov.au/rowland/media-release/communications-legislation-amendment-combatting-misinformation-and-disinformation-bill-2024
  7. Department of Home Affairs, "Frequently Asked Questions – The Assistance and Access Act" — https://www.homeaffairs.gov.au/nat-security/files/assistance-access-act-faq.pdf
  8. Australian Government response to the INSLM review of the Identify and Disrupt Act 2021 — https://www.homeaffairs.gov.au/reports-and-pubs/files/government-response-review-data-disruption-network-activity-account-takeover-powers.pdf
  9. Ofcom, "New rules for a safer generation of children online" — https://www.ofcom.org.uk/online-safety/protecting-children/new-rules-for-a-safer-generation-of-children-online
  10. The Standard, "Fact check: International data on 'online comments' arrests not comparable" — https://www.standard.co.uk/news/politics/russian-wales-england-germany-the-times-b1266428.html
  11. Crown Prosecution Service, "Childminder admits inciting racial hatred over social media post" (with sentence) — https://www.cps.gov.uk/cps/news/updated-sentence-childminder-admits-inciting-racial-hatred-over-social-media-post
  12. Judiciary of England and Wales, Sentencing remarks, R v Jordan Parlour (Aug 2024) — https://www.judiciary.uk/wp-content/uploads/2024/08/Jordan-Parlour.-Media-Posts.-Final.pdf
  13. Policy Exchange, "Non-Crime Hate Incidents" (Nov 2024) — https://policyexchange.org.uk/publication/non-crime-hate-incidents/
  14. The Guardian, "Met police to end non-crime hate investigations after Graham Linehan case" (20 Oct 2025) — https://www.theguardian.com/culture/2025/oct/20/father-ted-creator-graham-linehan-says-police-taking-no-further-action-over-trans-posts
  15. BBC News, "Apple pulls data protection tool after UK government security row" — https://www.bbc.com/news/articles/cgj54eq4vejo
  16. BBC News, "Court documents shed new light on UK-Apple row over user data" — https://www.bbc.co.uk/news/articles/cx293qg7z39o
  17. Apple Support, "Apple can no longer offer Advanced Data Protection in the United Kingdom to new users" — https://support.apple.com/en-ca/122234
  18. House of Lords Library, "Select communications offences and concerns over free speech" — https://lordslibrary.parliament.uk/select-communications-offences-and-concerns-over-free-speech/
  19. Reclaim The Net, "UK Arrests for 'Offensive' Online Posts Nearly Double in Seven Years" (9 Apr 2025, reporting The Times' FOI data) — https://reclaimthenet.org/uk-arrests-for-offensive-online-posts-nearly-double-in-seven-years
  20. Home Office, "Hate crime, England and Wales, year ending March 2026" — https://www.gov.uk/government/statistics/hate-crime-england-and-wales-year-ending-march-2026/hate-crime-england-and-wales-year-ending-march-2026
  21. The Standard, "Children as young as nine being probed by police for classroom non-crime hate incidents" (reporting Times FOI data) — https://www.standard.co.uk/news/crime/police-record-children-hate-incidents-telegraph-allison-pearson-essex-police-b1194223.html
  22. The Telegraph, "Police record 120,000 'non-crime' hate incidents that may stop accused getting jobs" (14 Feb 2020) — https://www.telegraph.co.uk/news/2020/02/14/police-record-120000-non-crime-incidents-may-stop-accused-getting/
  23. The Telegraph, "Non-crime hate incidents surge in half of police forces despite government crackdown" (23 Dec 2024) — https://www.telegraph.co.uk/news/2024/12/23/non-crime-hate-incidents-surge-half-of-police-forces/
  24. Free Speech Union, "Home Secretary calls time on NCHIs, but the battle for common sense policing continues" — https://freespeechunion.org/news/home-secretary-calls-time-on-nchis-but-the-battle-for-common-sense-pol
  25. The Guardian, "Police correctly investigated Allison Pearson's alleged racism, review finds" (6 Mar 2025) — https://www.theguardian.com/uk-news/2025/mar/06/police-correctly-investigated-allison-pearson-alleged-racism-review-finds
  26. The Guardian, "Wife of Tory councillor jailed for stirring racial hatred after Southport attack" (17 Oct 2024) — https://www.theguardian.com/uk-news/2024/oct/17/wife-tory-councillor-jailed-lucy-connolly-racial-hatred-southport-attack
  27. The Guardian, "Two men jailed for social media posts that stirred up far-right violence" (9 Aug 2024) — https://www.theguardian.com/politics/article/2024/aug/09/two-men-jailed-for-social-media-posts-that-stirred-up-far-right-violence
  28. The Guardian, "Parents arrested by Hertfordshire police for complaining about daughter's school" (29 Mar 2025) — https://www.theguardian.com/uk-news/2025/mar/29/parents-arrested-by-hertfordshire-police-for-complaining-about-daughters-school
  29. BBC News, "Hertfordshire Police admit WhatsApp arrest error with £20k payout" — https://www.bbc.co.uk/news/articles/c4gz1qy30v5o
  30. BBC News, "Graham Linehan arrested at Heathrow over his X posts" (2 Sep 2025) — https://www.bbc.co.uk/news/articles/c07p7v2nn8mo
  31. The Independent, "Graham Linehan claims he was arrested at Heathrow Airport for posts about trans people" — https://www.independent.co.uk/news/uk/crime/graham-linehan-arrested-twitter-tweets-heathrow-b2818446.html
  32. BBC News, "Graham Linehan receives settlement from Met over online post arrest" — https://www.bbc.com/news/articles/cvgdyjeqz8lo
  33. The Telegraph, "Pensioner receives police payout over 'thought crime' tweet" (31 May 2025) — https://www.telegraph.co.uk/news/2025/05/31/kent-police-20k-pensioner-thought-crime-tweet-fsu-foulkes/
  34. The Independent, "Police apologise for arresting former special constable over Palestine march social media post" — https://www.independent.co.uk/news/uk/crime/kent-police-apology-julian-foulkes-palestine-tweet-b2749128.html
  35. BBC News, "Kent PCC demands answers from force over X post case" (13 May 2025) — https://www.bbc.co.uk/news/articles/c0j718we6njo
  36. BBC News, "Hate awareness course scrapped by Hampshire police chief" — https://www.bbc.co.uk/news/uk-england-hampshire-62464983
  37. Dublin Live, "Dubliner arrested in UK for 'causing anxiety' by retweeting Pride flag swastika meme speaks out" (includes Hampshire Constabulary statement) — https://www.dublinlive.ie/news/dublin-news/dubliner-arrested-uk-causing-anxiety-24653444
  38. BBC News, "Harry Miller: Police probe into 'transphobic' tweets unlawful" (14 Feb 2020) — https://www.bbc.com/news/uk-england-lincolnshire-51501202
  39. Sky News, "Harry Miller: Police unlawfully interfered with freedom of expression over 'transphobic' tweets" (14 Feb 2020) — https://news.sky.com/story/harry-miller-police-unlawfully-interfered-with-freedom-of-expression-over-transphobic-tweets-11933809
  40. Conseil constitutionnel, Decision no. 2018-773 DC (20 Dec 2018) — https://www.conseil-constitutionnel.fr/en/decision/2018/2018773DC.htm
  41. Wisconsin International Law Journal, "The NetzDG and the Avia Law" — https://wilj.law.wisc.edu/wp-content/uploads/sites/1270/2023/08/491-518-Lewis_Proof_Final.pdf
  42. Tribunal judiciaire de Paris, press release (28 Aug 2024) — https://www.tribunal-de-paris.justice.fr/sites/default/files/2024-08/2024-08-28%20-%20CP%20TELEGRAM%20mise%20en%20examen.pdf
  43. Reuters, "French authorities charge Telegram's Durov in probe into organized crime on app" — https://www.reuters.com/world/europe/french-judges-decide-next-step-durov-probe-2024-08-28/
  44. Reuters, "French parliament backs social media ban for children under 15" (21 Jul 2026) — https://www.reuters.com/legal/litigation/french-lawmakers-vote-social-media-ban-children-2026-07-21/
  45. Conseil constitutionnel, Décision n° 2026-911 DC du 14 août 2026 — https://www.conseil-constitutionnel.fr/actualites/communique/decision-n-2026-911-dc-du-14-aout-2026-communique-de-presse
  46. POLITICO Europe, "'Pure bullshit': Macron slams tech giants' claim they are defending free speech" (18 Feb 2026) — https://www.politico.eu/article/emmanuelmacron-calls-social-medias-free-speech-defense-pure-bullshit/

Next: Keywords →