Reversing Britain’s Free Speech Recession

BNP: British National Party

CPS: Crown Prosecution Service

DPP: Director of Public Prosecutions

ECHR: European Convention on Human Rights

ECtHR: European Court of Human Rights

EHRR: European Human Rights Report

EWCA: England and Wales Court of Appeal

EWHC: England and Wales High Court

MCA 1988: Malicious Communications Act 1988

OSA 2023: Online Safety Act 2023

QB: Queen’s Bench

UKHL: United Kingdom House of Lords

UKSC: United Kingdom Supreme Court

WLR: Weekly Law Reports

Jon Holbrook is a barrister, having been called to the bar in 1991. He previously practised at Garden Court and Cornerstone, coming to specialise in public law. In 2021, he was expelled from Cornerstone for expressing socially conservative beliefs on Twitter. He was cleared of wrongdoing in 2022 by an appeals panel independent of his regulator, the Bar Standards Board, and the High Court has affirmed the principle used to clear him as establishing the right for professionals to speak freely on political issues. He now uses his legal expertise and personal experiences to support others in fighting for free speech. Visit his website at www.jonholb.com.

When I was younger, I would often hear adult friends and relatives arguing about politics. Accepting that some disagreements could not be resolved, they would shrug their shoulders and say “well, it’s a free country”. You do not hear people saying that very often nowadays. For we all know in our hearts we are not in as free a country as we were then, and we have all learned the art of keeping our thoughts on certain matters to ourselves. Today any of us might say the wrong thing, in the wrong way, at the wrong moment, and find the police at the door.

How this happened, and what we can do about it, is the subject of this excellent and timely Prosperity Institute paper by the barrister Jon Holbrook. Its first great merit is that it documents exactly how the decline happened. It traces the story back to the 1965 Race Relations Act, which first established that the state could criminalise a particular kind of political belief, and then, crucially, to the Public Order Act 1986, which made it illegal to cause “harassment, alarm or distress”. This was the crucial point at which it became accepted that the state had the right to limit whole categories of speech—the intellectual turning point from which much else has flowed.

Of course, the world of 1986 was very different. Words considered to be dangerous were spoken or written by small numbers of people: the demonstrator with a megaphone, the cheaply photocopied pamphlet, the green-ink abusive letter, the obscene phone call. Illiberal legislation, bad as it was, was never likely to be turned against most British citizens.

Things are very different now. We all have in our pocket a gadget that can broadcast to the entire world our carelessly thought-through and casually expressed ideas and beliefs. Even if you do not use social media yourself, your spoken or written words in an entirely different context can be circulated and made available to anyone—the casual remark or question at work, the unhappy conversation with a defensive colleague, the banter in a pub among friends.

This is what makes the current situation so dangerous. Our legislation was simply not designed for this world, for a social media environment where everyone gets a say and with their own punchy style, where one person’s sharp comment is another’s abuse. In this new environment, the cumulative effect of the laws on the books is to hand government an extraordinarily broad power, with precious few checks and balances to restrain its use. Vague terms—”grossly offensive,” “abusive,” “insulting,” “stirring up”—are left to be given meaning by judges and prosecutors after the event, and the chilling effect falls on all of us. In a country with real problems of immigration and integration, is a comment on crime among particular communities fair political argument, or is it “stirring up racial hatred”? Nobody quite knows, and so people fall silent. Laws that once caught a handful of extremists can now be turned against any of us, at any moment.

Behind the laws lies a wider mood. Much of our political class has succumbed to a moral panic about “misinformation” and “disinformation”—words scarcely used until around 2016, when Brexit and the election of President Trump persuaded a startled establishment that the voters could no longer be trusted. Their reaction was to control and to clamp down. Yet they are as bad as anyone else. The most damaging misinformation of our age comes not from ordinary citizens but from governments. They are the last people who should be the arbiters of truth. Only free and open debate can give us the right answers to our problems and bring people along to support them.

Holbrook also shows why the European Convention on Human Rights gives us no meaningful protection. Article 10 has provided no real shield against the rise of hate speech law. Quite the contrary. The Strasbourg court has been comfortable in supporting hate speech restrictions and has steadily imported continental norms that sit uneasily with our own common law tradition of liberty. The argument about leaving the ECHR is often entirely framed in the context of immigration policy, but Holbrook reminds us there is much more to it than that. Once again, there is no alternative but to leave the ECHR altogether and to restore these questions to where they belong: to Parliament and our own laws.

The second great merit of Holbrook’s paper is that it does not just document what happened but shows how we can change it. Holbrook’s five specific proposals would hack back the undergrowth, that whole accretion of restrictions built up over sixty years: repealing the bad criminal law; replacing a tangle of overlapping communications offences with a single, tightly drawn provision; sweeping away hate speech law entirely; and giving free speech proper protection in the civil courts through a new tort and a sensible costs regime. Together they amount to a serious plan for getting a grip on the current unsatisfactory situation. They deserve full attention.

It will of course always be impossible to have perfect laws in this area. No legislation can foresee all circumstances or define all possible actions. The culture is also important. Britain was fortunate in having, until recently, a political class that genuinely believed in free speech, indeed saw it as a fundamental element of the British way of governance distinct from many others. Unfortunately, this is simply no longer true. Our establishment no longer regards free debate as important. Indeed, it generally sees it as an inconvenience, a threat, a danger to a complex social order that needs to be managed carefully if it is not to blow up in our rulers’ faces. They would like to be the Platonic guardians who know what is best for us and who can control what can be said and what can’t.

This is the attitude we must defeat. The first step is changing the laws. Doing so will not in itself rebuild the culture, but it is where the recovery must begin. If we allow the present drift to persist, the same statues may still be standing in the same tree-lined squares but we will be a different country. Indeed, we are already a long way down that road. The time is short and the situation is urgent. Jon Holbrook’s paper shows how we can change things before it is too late.

David Frost

The Rt Hon Lord Frost CMG

“This report provides a forensic explanation of why free speech is in such a bad way and how we can get back to where we should be, positing a vision of free speech that draws on a fundamentally English tradition from first principles. Until we implement these recommendations we cannot, in good faith, claim to live in a nation that values the freedom that precedes all others.”

— Jack Rankin MP

“Freedom of speech is the cornerstone of a free society. It is the right that protects all other rights. Without it, democracy becomes little more than managed consensus, with citizens discouraged from questioning prevailing orthodoxies or challenging those in power. For generations, Britain was a country that understood this. We believed that bad ideas should be defeated through debate, not censorship. We recognised that the answer to speech we dislike is more speech, not state intervention. Yet over recent decades, that confidence has been steadily eroded.”

“Too many people now fear speaking openly about issues that matter deeply to them. Whether the subject is immigration, religion, national identity, race, crime, or gender, there is a growing sense that expressing lawful opinions can carry serious professional, social and even legal consequences. That should concern everyone who values democracy, regardless of their political persuasion.”

“This paper provides a powerful and timely examination of how we arrived at this point. It argues persuasively that Britain has drifted away from its long tradition of robust free expression and that a growing body of legislation has contributed to a culture in which certain viewpoints are tolerated while others are increasingly suppressed. In particular, the paper raises important questions about the role of the Equality Act and the extent to which it has been stretched far beyond its original purpose. It is increasingly clear that Britain needs a new settlement, one that robustly protects equality before the law while placing freedom of speech firmly back at the centre of our democratic culture.”

“This paper is a serious, thoughtful and important contribution to that debate. I warmly welcome its publication and hope it helps to build the momentum needed for meaningful reform. Britain’s tradition of free speech is one of our greatest inheritances. We should be determined to defend it.”

— Rt Hon Suella Braverman KC MP

“From down in Australia, Anglophile friends of Britain look at the state of free speech up there in despair. The world is no longer praising the flood of British freedoms because they are no longer flowing with pomp of waters, unwithstood. Free speech is in a particularly bad way, if I may presume to say so from a former colony. This paper sets out exactly the proper needed responses.  It is a thoroughly British response, largely to repeal the various statutes that have created this censorious state of affairs. The author is correct to reject the European Convention-based approach. The state of play there today makes that plain. And the simple fact is that no country today is able to import a US First Amendment, plus its last century plus of jurisprudence, plus the sort of American judges who make it work. No, this paper is by far the best way to reinvigorate free speech in the United Kingdom. It needs to be read, and more importantly to be acted upon, by all political parties up there which purport to care about free speech.”

— James Allan, Garrick Professor of Law, University of Queensland

Numerous commentators have drawn attention to the demise of free speech in Britain in recent years.[1] Having once been a country which experienced a boom in free speech, we now find ourselves in a recession of it. This paper seeks to explain it and makes five proposals to remedy it, as outlined in a draft Free Speech Bill shown in boxes throughout.

Most legal problems have long tails and Britain’s free speech recession is no exception. The ground was laid by the passage of two Acts of Parliament in 1965 and 1986, each of which abridged legal norms that had hitherto protected free speech. The first, the Race Relations Act 1965 which introduced the offence of “stirring up racial hatred”, ended the notion that any citizen was free to express his political beliefs, no matter how unpleasant or extreme. Even the Public Order Act 1936, which was passed to control extremist political movements, particularly the British Union of Fascists led by Oswald Mosley, did not outlaw any form of political speech.

By making it an offence to “stir up” racial hatred in 1965 the government criminalised expressed beliefs on the basis that they were the “wrong” beliefs. With this new offence the government established that the state had a right to limit the expression of beliefs of a particular political kind. After 1965 any debate about race and immigration could extend without limits in a liberal direction (“immigration and multiculturalism are good”) but those who took it in the opposite direction (“immigration and multiculturalism are bad”) ran the risk of prosecution.

The second Act to abridge a legal norm was the Public Order Act 1986, which criminalised for the first time words that were “likely to cause harassment, alarm or distress”. This threshold was much lower than had existed in previous legislation that had required the prosecution to show that offending words were likely to “provoke a breach of the peace”. With this new offence of causing harassment, alarm or distress the government established that the state had a right to limit speech which threatened to cause neither a breach of the peace nor violence.

By the twenty-first century social media had given everyone a voice that could be heard beyond the citizen’s family and friends. This should have been viewed as an opportunity to engage the public in a free and wide debate on anything political. Instead, the state, drawing on ground prepared in 1965 and 1986, sought to restrict public debate with criminal laws that inappropriately targeted three types of speech:

• General speech: defined as speech that is not sent via a medium (electronic or traditional) is speech that a potential listener can usually avoid. Its curtailment should be limited to that which is either intended to, or might reasonably be expected to, provoke violence. Yet, since the passage of the Public Order Act 1986 the threshold for curtailing general speech has been too low, as section 5 criminalises speech that may cause harassment, alarm, or distress. It needs to be repealed.
• Communications speech: defined as speech sent via a medium (electronic or traditional) is speech that a potential listener cannot usually avoid. It should only be outlawed if it is either intended to be, or might reasonably be expected to be, menacing and targeted. This principle requires three existing offences to be repealed (under the Malicious Communications Act 1988, the Communications Act 2003, and the Online Safety Act 2023) and replaced with one offence of communication that is intentionally menacing and targeted.
• Hate speech: defined as political speech that the state has decided to criminalise. Hate speech laws have no place in a democracy, where undesirable political views should be defeated with better ones. Accordingly, all hate speech laws need to be repealed.

With laws that inappropriately criminalise speech, Parliament has engendered a cultural norm that those who express the “wrong” views may be imprisoned or visited with punishments in the workplace and beyond. Underpinned by bad criminal laws, the nation’s civil laws have proven inadequate to protect citizens from what has become known as “cancel culture”.

When it comes to speech, the criminal law has expanded too much and the civil law has expanded too little. Although the common law has a good tradition of protecting free speech it has so far failed to protect most victims from cancel culture in the workplace and beyond. Although many gender critical feminists[2], some Christians[3], and some progressives[4] have won important free speech victories by relying on the statutory protection afforded to “religion and belief” in the Equality Act 2010, others who have lost their freedom after speech have fared less well.[5] This problem can be addressed with three reforms:

• Take the religion and belief protection out of the Equality Act, where speech suppression is viewed through the lens of discrimination law, and place it in the proposed Free Speech Bill, where it is protected as a tort of “unlawful interference with the right of free speech”.
• Widen the protection by protecting speech in general, whether or not it qualifies as a “religion or belief”.
• Introduce a costs-capping regime for free speech cases in the civil courts which mirrors the costs protection currently afforded to “environmental claims” under the Aarhus Convention.

There are some specific speech issues which, while egregious, have fallen beyond the scope of this paper, such as the legal “buffer zones” for protestors around abortion clinics, an issue which also incorporates distinct questions of freedom of assembly and association.

Finally, the approach adopted in this paper is unashamedly English and British. The UK is not America and although there is much to admire with the American approach to free speech it cannot and should not be transported across the Atlantic. Britain, unlike America, does not have a written constitution, not because it is deficient but because the sovereignty of Parliament is at the apex of its constitutional framework. Problematic laws in this jurisdiction are Parliament’s responsibility to fix.

The UK also does not have the traditions of a European state, with their civil law traditions that tend to command legal outcomes with top-down codes. The English bottom-up common law approach has served it well with a judiciary that has mostly recognised the importance of free speech. In recent years, this recognition has sometimes yielded in the face of an edifice of criminal laws that have induced a degree of judicial suspicion towards certain kinds of speech. But the remedy for that problem is to repeal the offending criminal laws; it is certainly not to empower the judiciary even more with a civil law code such as that provided by the European Convention on Human Rights, a code that has done nothing to protect free speech in the face of hate speech laws in Britain or Europe.

Main points

• The Race Relations Act 1965 inaugurated an era of bad laws by which Parliament restricted free speech. The shortcomings of these laws disastrously collided with the social media age.
• We must repeal inappropriate criminal laws, criminalising only speech intended or likely to cause violence in public order speech or that is intended to be menacing and targeted in communications speech.
• In civil law, we must remove religion and belief protections from the Equality Act and replace them with a new tort of unlawful interference with free speech and introduce a costs regime to protect those wrongfully accused and to deter interference in the first place.

“We may pride ourselves on the fact that freedom of speech has existed in this country perhaps as long as, if not longer than, it has existed in any other country in the world.”

— Lord Goff[6]

The problem

“At what point did we become North Korea?” asked Nigel Farage when questioned by a US congressional committee about free speech limitations in the UK. Farage is not alone in noting that the UK’s traditional respect for free speech has declined in recent years. Speaking in late 2025 former Liberal Democrat leader and Deputy Prime Minister, Nick Clegg, said the UK needed to “think long and hard” about “whether we’ve overdone it”. The former president of global affairs at social media giant Meta said “When I’ve looked at some of the examples, I thought to myself, ‘yeah, that’s really unpleasant speech or egregious speech.’ But really, surely part of the definition of being in a free society is people say ghastly things, offensive things, awful things, ugly things, and we don’t sweep them under the carpet.” He concluded by saying that “the balance [on free speech] is out of whack here.”[7]

Farage and Clegg may have had in mind the arrest at Heathrow of Father Ted co-creator Graham Linehan over his X-posts on transgenderism, or Lucy Connolly’s imprisonment following her intemperate X-post in response to the shocking murder of children at a Taylor Swift themed event. Her 31-month prison sentence, shocked many, including people who were appalled by what she had written. In recent years scarcely a week has passed without somebody being in the news for having said something for which they have been sanctioned or shunned. Indeed, the practice has become so commonplace that we have a now familiar expression to describe it: cancel culture.

The individual consequences of cancel culture vary from social shunning to the loss of employment, and in some cases even the ability to work, but its effects are generally ruinous to a democracy. If people are intimidated into not feeling able to speak their minds on political issues then political discourse is a charade and the politicians who treat it as representative of national sentiment are deluding themselves. Voters typically respond either by getting angry or by switching off and neither is healthy for the body politic. For good reason, free speech has often been described as the lifeblood of democracy.

Furthermore, cancel culture is pernicious because it is not even-handed in the views that it tends to target. As Steve Forbes—the editor-in-chief of US business magazine Forbes—noted: “the UK has, with increasing vigour, been curbing what one is allowed to say, all in the name of fighting racism, sexism, Islamophobia, transgenderism, climate-change denial, and whatever else the woke extremists conjure up.”[8] The censuring of anti-woke beliefs may explain why those on the political right tend to value free speech particularly highly. A YouGov poll in 2025 asked British adults: “When it comes to online behaviour in the form of comments and social media posts, which is more important?” 28% said: “That people are able to speak their minds freely” (as compared to 61% who said “That people are safe from abuse and threats”). But that 28% average rose to 64% for Reform voters (29% for Conservative) yet fell to 14% for Liberal Democrat voters (17% for Labour).

A similar disparity existed in response to Brexit. A 2019 study of over 500 university students in the UK found that fewer than 40% of Leave-supporting students felt they could express views on Brexit in front of their classmates, compared to nearly 90% of Remain supporters who felt comfortable expressing their views.[9] So, right-wing or Brexit supporting voters are far more likely to experience cancel culture than left-wing or Remain voters. Cancel culture degrades democracy by silencing the political views of a particular persuasion.

The problem’s genesis

The UK’s free speech recession of recent years has roots dating back to 1965 when the first Race Relations Act was passed. Aside from pre-modern issues like treason, sedition, and blasphemy this meant that for the first time in the UK’s history Parliament had enacted laws that could criminalise people on account of their particular political opinions on the issues of “colour, race or ethnic or national origins”. This was via a “stirring-up hatred” provision but hatred, of course, is a sentiment qualitatively different from say, an act of violence which the criminal law would appropriately target. Before 1965 speech had been outlawed on public order grounds, but after 1965 it could also be outlawed on political grounds, in circumstances that gave rise to no public order issues.

Having crossed the Rubicon, other hate speech laws followed that targeted views and epithets about religion, sexuality, disability, and gender. Within ten years, between 1998 and 2008, Labour governments passed five Acts that expanded the reach and scope of what are tendentiously called “hate speech” laws, but which could more neutrally be called simply “speech” laws. Starting in 1965 civil laws with the same targets expanded in step with criminal laws to challenge people with the “wrong” views on race, religion, sex, sexuality and gender. Although these civil laws aimed to and did outlaw improper discriminatory treatment in the workplace and beyond they also fuelled a culture in which expressions of certain beliefs, that fell short of being instances of discriminatory treatment, were unlawful.

So, when social media expanded in the twenty-first century and gave everyone a public voice, there was a panoply of criminal and civil laws already in place ready to address expressions of everything from the anodyne to the vile, even though these views fell short of inciting violence. For reasons explained below the people who fall foul of these laws are victims of legislative overreach. It is the objective of this paper to explain that overreach and to propose legislative reform to end it. But first it is necessary, at a time when free speech is seen by many as inherently bad, to explain why it is actually inherently good.

The solution: recognising that free speech is inherently good

Many philosophers have recognised that free speech is inherently good. In 1859 the English philosopher, John Stuart Mill, published On Liberty, explaining why. Mill’s essential point was noted by House of Lords judge, Lord Steyn, when in 1999 he observed that “In a democracy [freedom of expression] is the primary right: without it an effective rule of law is not possible.”[10] In the following quote Lord Steyn restates the point and then summarised the three reasons given by Mill, which were then echoed in America by Justice Oliver Wendell Holmes:

Freedom of expression is, of course, intrinsically important: it is valued for its own sake. But it is well recognised that it is also instrumentally important. It serves a number of broad objectives. First, it promotes the self-fulfilment of individuals in society. Secondly, in the famous words of Holmes J. (echoing John Stuart Mill), “the best test of truth is the power of the thought to get itself accepted in the competition of the market:” Abrams v. United States (1919) 250 U.S. 616, 630, per Holmes J. (dissenting). Thirdly, freedom of speech is the lifeblood of democracy. The free flow of information and ideas informs political debate. It is a safety valve: people are more ready to accept decisions that go against them if they can in principle seek to influence them. It acts as a brake on the abuse of power by public officials. It facilitates the exposure of errors in the governance and administration of justice of the country…[11]

Nearly thirty years after Lord Steyn described free speech as the lifeblood of democracy, it can be seen that his three reasons for supporting freedom of speech were apposite. First, those cancelled, or fearing cancellation, do lack self-fulfilment and are likely to feel alienated and frustrated, if not angry and bitter. Secondly, bad ideas are more likely to be adopted where the marketplace of ideas is skewed by an unfair competition that rigs outcomes in favour of some ideas at the expense of others. Thirdly, democratic engagement has waned as those who experience or fear cancellation are less likely to accept decisions that go against them as they are denied their right to influence them. The “brake on the abuse of power by public officials” that Lord Steyn referred to is unavailable. Just as free speech is the lifeblood of democracy, cancellation is its cancer.

A Free Speech Bill to reverse the free speech recession

Few people are free speech absolutists. It is taken for granted in this paper that the state must have laws that may curtail speech in the interests of protecting others and other interests. Issues such as terrorism, defamation, confidentiality, and the protection of children from online harms are beyond the remit of this paper. The issues addressed below are aimed at liberating Britain from the tyranny of cancel culture so that people of all descriptions can engage in political speech without fear or favour. The central issue is therefore to explain where and why to draw the line between speech that should be lawful and speech that shouldn’t.

A Free Speech Bill with five key proposals, outlined in boxes below, is explained in this paper as being necessary to reverse the UK’s free speech recession. The Bill’s very existence will alert the public and judiciary to the importance of free speech and to its role as being the lifeblood of democracy.

The Bill will address criminal and civil laws with provisions that will:

1. Repeal inappropriate criminal laws. These reforms will be based on the principle that speech should only be criminalised where it was either intended or likely to:

(a)   cause violence – in the context of public order

(b)   be menacing and targeted – in the context of communications speech

Accordingly, “hate crime” laws which criminalise speech that falls short of having either of the above two effects will be repealed.

2. Protect speech in the workplace and beyond. Belief discrimination laws will be recast as protections for speech in general, whether or not it constitutes a “belief”. These civil law protections will be taken out of the Equality Act and put into the proposed Free Speech Bill because free speech is not a discrimination issue. Claimants will also be given cost protection so that these rights can actually be enforced by an individual with modest resources.

Main points

• Repeal section 5 of the Public Order Act 1986, which criminalises speech merely because it may cause harassment, alarm, or distress.
• Repeal three existing communications offences under the Malicious Communications Act 1988, the Communications Act 2003, and the Online Safety Act 2023, and replace them with one offence of communication that is intentionally menacing and targeted.
• Repeal all so-called “hate speech” laws.

Criminal law is concerned with harm to the public, as opposed to harm to a private or particular interest, which is the concern of civil law. This is why criminal proceedings are brought by the Crown and funded by the public. The criminal justice system in general has four stages: (i) prevention and investigation (by the police), (ii) prosecution (by the Crown Prosecution Service, CPS), (iii) trial (by judges) and (iv) punishment (such as by the prison service). Each of these four stages can have a profound effect on speech, often referred to as “the chilling effect”. Clearly, a conviction and sentence for saying something criminal has the greatest impact. But the chilling effect also applies where a citizen is brought within the first stage, such as by receiving the policeman’s knock on the door, even if the case is not prosecuted. Campaigners like Harry Miller[12], journalists like Allison Pearson[13] and social media activists like Pete North[14] can testify to the chilling effect of coming within the purview of criminal justice.

Accordingly, criminal laws that impact on speech need to be tightly framed so that only those who truly deserve to be investigated can ever expect to receive the policeman’s knock on the door. This gives rise to three issues, which are explained in more detail below:

• General speech in a public place, which a potential listener can generally choose to avoid. This speech should only trouble the criminal law if it is likely to cause violence.
• Communications speech, which relies on a communications network, which may be traditional (post or telephone) or electronic but which is addressed to a specific person or persons so as to deny the potential listener the option to avoid it. The targeted and unavoidable nature of this speech requires the criminal law to outlaw the communication if it is likely to be menacing.
• Hate speech is speech with hateful content about a section of society (defined by their race or religion, etc.) but which is criminalised even though it is not likely to cause violence and is not likely to be a menacing form of communications speech. Hate speech should only trouble the criminal law if it crosses either of the above two thresholds.

General speech

We saw above with reference to the words of John Stuart Mill and Lord Steyn that speech is inherently good and hence that any curtailments of it must be narrowly defined. In fact, in a public order context that restriction should be confined to instances where speech is likely to cause violence. And in this regard the problem is not necessarily the message itself, but the circumstances pertaining to it. As Mill observed in 1859:

An opinion that corn-dealers are starvers of the poor, or that private property is robbery, ought to be unmolested when simply circulated through the press, but may justly incur punishment when delivered orally to an excited mob assembled before the house of a corn-dealer, or when handed about among the same mob in the form of a placard.[15]

Mill’s point is that an otherwise unobjectionable message could become objectionable if it incited violence, an outcome that would depend on how, where and when it was delivered. The point was echoed recently when a Koran burner was acquitted by the Crown Court after having regard to the fact that his protest took place outside the Turkish Consulate, whereas he might have been convicted had he burnt the Koran outside a mosque. The former was a legitimate form of expressive protest whereas the latter may have incited violence and hence properly been the subject of criminal law.[16]

In 1919 Justice Holmes concurred with Justice Brandeis in delivering what law professor Adam Tomkins describes as “probably the most brilliant defence of free speech ever presented from the bench”.[17] The defendant had been convicted of a crime for supporting and helping to organise the California Communist Party at a time when, in the aftermath of the Russian Revolution, there was much concern in America about the “Red Scare”. Brandeis and Holmes JJ delivered a judgment addressing three issues. First, they restated the importance of free speech by drawing on principles outlined by Mill and others, such as America’s founding fathers, who “believed that freedom to think as you will and to speak as you think are means indispensable to the discovery and spread of political truth”. They noted “that the greatest menace to freedom is an inert people; that public discussion is a political duty; and that this should be a fundamental principle of the American government.”[18]

Secondly, Brandeis and Holmes JJ stressed that it was dangerous to silence speech without particularly good reason because “silence coerced by law” was the true enemy of resolving grievances by “the power of reason”. Thus, they noted that their founding fathers:

knew that order cannot be secured merely through fear of punishment for its infraction; that it is hazardous to discourage thought, hope and imagination; that fear breeds repression; that repression breeds hate; that hate menaces stable government; that the path of safety lies in the opportunity to discuss freely supposed grievances and proposed remedies; and that the fitting remedy for evil counsels is good ones. Believing in the power of reason as applied through public discussion, they eschewed silence coerced by law – the argument of force in its worst form. Recognising the occasional tyrannies of governing majorities, they amended the Constitution so that free speech and assembly should be guaranteed.[19]

Thirdly, words are not in and of themselves harmful. They have the potential, in a minority of cases and depending on circumstances, to cause harm as Mill recognised with his image of words addressed “to an excited mob assembled before the house of a corn-dealer”. Brandeis and Holmes JJ noted that advocacy of many things, such as law-breaking, could be “reprehensible morally” but it was:

not a justification for denying free speech where the advocacy falls short of incitement and there is nothing to indicate that the advocacy would be immediately acted on. The wide difference between advocacy and incitement, between preparation and attempt, between assembling and conspiracy, must be borne in mind.[20]

Hence their conclusion that they described as the “clear and present danger” test:

To justify suppression of free speech there must be reasonable ground to fear that serious evil will result if free speech is practised. There must be reasonable ground to believe that the danger apprehended is imminent. There must be reasonable ground to believe that the evil to be prevented is a serious one…. In order to support a finding of clear and present danger it must be shown either that immediate serious violence was to be expected or was advocated, or that the past conduct furnished reason to believe that such advocacy was then contemplated.[21]

In the twenty-first century our thinking on speech has been so inured by a prevailing cancel culture that the boundary outlined by two of America’s finest judges may seem preposterous. Surely, people say, it must be possible to criminalise speech that does not incite violence, such as speech that is vile, hateful, or offensive?

Britain before the 1986 Public Order Act

In fact, the notion that speech should generally only be criminalised if it posed a clear and present danger of violence mostly prevailed in Britain until the Public Order Act of 1986.

Since time immemorial and by statute dating back to 1361 speech that was likely to occasion a breach of the peace had been unlawful.[22] Moreover, a breach of the peace was closely connected to violence as the High Court observed in 1982 when defining it as:

whenever harm is actually done or is likely to be done to a person or in his presence to his property or a person is in fear of being so harmed through an assault, an affray, a riot, unlawful assembly or other disturbance. It is for this breach of the peace when done in his presence or the reasonable apprehension of it taking place that a constable, or anyone else, may arrest an offender without warrant.[23]

Dating back to 1839 the Metropolitan Police had a power to penalise those who “use any threatening, abusive, or insulting words or behaviour with intent to provoke a breach of the peace, or whereby a breach of the peace may be occasioned”.[24] This provision stood the test of time without undermining free speech rights.

Box 1: Public Order Act 1936

Nearly 100 years later the police power was codified into a nationwide criminal law as the Public Order Act 1936, section 5, as shown above. The Act was passed in response to marches by Oswald Mosely’s fascist Blackshirts and general disorder, much of it directed at Jews. Labour MP Herbert Morrison supported the Bill on the grounds that “It has, however, always been said that freedom of speech and of expression must take reasonable account of other people’s freedom as well” and hence he concluded that the Act was necessary “with a view to checking action which is calculated to destroy the liberty that we wish to preserve”.[25] The provision did not unreasonably fetter free speech and yet it proved to be adequate to the disorder the country then faced, as noted by two legal commentators: “Despite the moderate nature of the penalties, section 5 achieved its immediate object.”[26]

Britain after the 1986 Public Order Act

Section 5 of the 1936 Act was updated and replaced by section 4 of the 1986 Act by substituting the threshold of a “breach of the peace” with one of “unlawful violence”. In other words the new public order offence of “fear or provocation of violence” was entirely consistent with the approach towards speech that had existed in Britain for centuries and which had been supported by Mill in Britain and Holmes in America.

Box 2: Public Order Act 1986, as enacted

However, what was new and problematic with the Public Order Act 1986, was the creation of a new offence in section 5 which set the thresholds for criminal intervention much lower. Although the speaker’s words still had to be either “threatening, insulting or abusive” the criminal threshold was crossed without any connection to violence. Indeed, the threshold under section 5 is crossed providing the speech is “within the hearing or sight of a person likely to be caused harassment, alarm or distress”.

The breadth of the speech impugned by section 5 is subject to a defence on the basis that the “conduct was reasonable” (s5(3)).[27] It was as if the section had been designed to ensure that police, prosecutors and judges could read into the new offence whatever they wanted: behaviour they wished to ignore could be excused as reasonable, and behaviour they wished to sanction could be actioned on the grounds that it was likely that somebody would have been caused harassment, alarm or distress.

The broad and subjective nature of the offence explains why in recent years it has been used on numerous occasions to penalise speech that should not have troubled the criminal justice system (see box 3).

Box 3: Public Order Act 1986, s5: prosecutions that infringed speech

“Harassment, alarm and distress”: its enduring effect on speech

Two developments in recent years have reduced the impact of section 5 of the Public Order Act. First, the courts have tended to give greater weight to the importance of free speech, a welcome approach that enabled the Koran burner, Hamut Coskun, to be acquitted on appeal. Although even here the fact that the magistrates court convicted Coskun and that the CPS then appealed the Crown Court’s acquittal[34] to the High Court[35] shows that the message about “the cardinal importance of freedom of expression in a democratic society” (see Hammond case in box 3) has not resonated with the entire criminal justice system.

Secondly, in many circumstances prosecutors have been able to rely on hate speech laws, rather than section 5, a burgeoning problem addressed below.

However, section 5 remains a menace to free speech in those circumstances where hate speech laws either do not apply or are not relied on. The problem is highlighted by the recent conviction under s5 of COVID lockdown protestor, Debbie Hicks, summarised in the box above.

1^st proposal: Repeal s5 of the Public Order Act – causing harassment, alarm or distress

Communications speech

The provision that is now found in the Malicious Communications Act 1988 (MCA), s1, can be traced back to the Post Office (Amendment) Act 1935[36], which was then aimed at hoax or other menacing phone calls, particularly those aimed at abusing telephone operators. However, with the development of electronic communication the MCA 1988 has been supplemented first by the Communications Act 2003 and more recently with the Online Safety Act 2023 (OSA), which for our purposes, now contains offences regarding sent communications that were either false or threatening. Accordingly, this paper is concerned with four offences arising from communications speech. They are set out more fully in box 4, but may be summarised as:

• Intending to cause distress etc (Malicious Communications Act 1988, s1)
• Improper use of a network (Communications Act 2003, s127)
• False communication (Online Safety Act 2023, s179)
• Threatening communication (Online Safety Act 2023, s181)

Although altered in scope the two OSA 2023 offences were previously found in the Communications Act as false speech (s1(a)(iii)) and threatening speech (s(1)(a)(ii)).

Communications speech relates to messages that are sent via a medium, traditionally by letter or telephone, but nowadays typically via the internet. This speech differs from the general speech identified above where the speech is not necessarily “sent” but is merely received. This distinction is material because when speech is sent it cannot readily, if at all, be avoided. Moreover, since it is sent it can be targeted, done repeatedly, and may even be done anonymously, and when done electronically it is also possible to display the communication to the public. Each of these factors can add to the menacing nature of communications speech, particularly if that is the sender’s intention.

Box 4: Existing communications offences

A different type of harm

The type of harm that can arise from communications speech is demonstrated by a case involving online grossly offensive abuse of a Jewish MP, Luciana Berger. The harm to her did not arise from there being a risk of violence, as can happen with general speech, but there was nevertheless harm, which was significant enough to warrant the criminal law’s attention.

• In 2014 Garron Helm, 21, posted a tweet with a photograph of Jewish MP Luciana Berger superimposed with a Holocaust-era yellow star on her forehead and the hashtag “Hitler was right”. Ms Berger was then bombarded with messages from trolls. This speech may not have incited violence but the defendant pleaded guilty to sending a grossly offensive message with the “purpose of causing annoyance, inconvenience or needless anxiety” under the Communications Act 2003, s127. He was sentenced to four weeks in jail.[37]

Thresholds for criminalisation: the problem

The first three communications offences are problematic because they do not identify a purpose that justifies criminal intervention. They rely on a hotch-potch of terms and provisions. The problem is most obvious with the “improper use” offence established by the Communications Act 2003, s127, where the notion of a communication being “improper” establishes a low and subjective threshold, namely the sending of a message that “was grossly offensive or of an indecent, obscene or menacing character”. These thresholds are far too low and subjective for curtailing speech, especially as they do not require any targeting or menacing intent to be established. The offence does not even require the prosecution to prove that the sender intended or was aware that harm could result.

The expansive nature of the “improper use” offence under the MCA 1988 induced the House of Lords to explain Parliament’s purpose in enacting it. The court did so by introducing the notion of “basic standards of our society” a term that is also far too low and subjective to be a basis for chilling speech, the lifeblood of democracy:

The purpose of the legislation which culminates in s.127(1)(a) was to prohibit the use of a service, provided and funded by the public for the benefit of the public, for the transmission of communications which contravene the basic standards of our society.[38]

Box 5: Communications Act 2003, s127: prosecutions that infringed speech

Box 5 identifies some of the cases in which the “improper use” offence has been used to chill communications speech in circumstances where it should have been clear that the speech in question should not have been subject to the criminal law.[43] Some of the particular problems are:

• The 1988 Act offence requires the prosecution to prove that one of the sender’s “purposes” was to “cause distress or anxiety to the recipient”, but distress or anxiety are too low a threshold. It means for example that the person who primarily intends to deliver a punchy political message (about say a war) but who also intends to cause the recipient to feel distress or anxiety, because the subject matter necessary invokes such responses (such as photos of dead children in a war), would have the necessary “guilty mind” for the offence to be made out.
• The false communications offence of the OSA 2023 is equally problematic because the guilty mind can be established from an intention “to cause non-trivial psychological or physical harm to a likely audience”. Again, “non-trivial psychological harm” is too low a threshold especially as the relevant sufferer need not be the recipient or intended recipient but is any member of “a likely audience”.
• The improper use offence does not expressly require any mens rea. However, the House of Lords[44], as interpreted by the Court of Appeal[45], implied into the offence a requirement that the prosecution shows an intention or awareness of risk that the person “to whom the message relates” would be grossly offended. As above, this implied provision suffers from requiring an awareness of a harm that is much too low.
• A further problem with the implied mens rea is that the awareness of harm need not be confined to an awareness that harm might be caused to the intended recipients of a message. The awareness need only arise from harm to those referred to in the message, who may never see it. This was the issue that enabled the police officers in Cobban to be convicted because although there was no likelihood of the actual recipients (consenting police officers) being grossly offended, those whom the messages insulted (women, gays and disabled people etc) clearly could have been, even though they were never the intended or likely recipients of the messages.

Thresholds for criminalisation: the solution

The threatening communications offence of OSA 2023 s181 should be retained for cases of death threats and other serious harm.

But lesser forms of communication should only be criminalised if the prosecution can satisfy these three elements:

• an actus reus arising from the sending of a communication that has two qualities, namely the communication is:
  • menacing in content due to being indecent, grossly offensive, threatening or false (when known or believed to be so); [46]
  • targeted to a specific person or persons; and
• a mens rea that requires the prosecution to prove that the sender either intended the message to be menacing or was aware that it was likely to be menacing to the intended or likely recipient.

This solution is set out in a second proposal shown below.

2^nd proposal: Communications offence – intentionally menacing and targeted

With the offence set out above it can be seen that none of the prosecutions shown in Box 5 should have attracted the attention of the criminal law because even if the first two requirements were satisfied the third criteria would not have been because the required menacing intention to a likely recipient was absent.

Hate speech

As set out above, there are occasions where people say things that properly attract the criminal law’s attention, namely where:

• general speech is intended or likely to cause violence; or
• communications speech is intentionally menacing and targeted.

This paper contends that speech of any kind that falls short of these two thresholds should not be criminalised, including so called “hate speech”. Hate crimes seek to delegitimise certain views about issues such as race, religion, sexuality and gender—the issues to which hate crimes usually attach. In a democracy the criminal law should not outlaw beliefs of any description. Of course, sometimes people expressing hateful views will also incite violence or engage in intentionally menacing communications. But if they do, they should then be prosecuted for crossing those thresholds not for having particular beliefs.

Box 6: Racially aggravated prosecutions, Crime & Disorder Act 1998, s28

Four instances of hate crimes are noted in box 6, all cases where the defendants did not incite violence or target anyone with a menacing intent. They were all prosecuted and all except Griffin were convicted (although Stewart was cleared on appeal) for expressing the “wrong” ideas.

Last year, Sam Melia became one of the latest to be convicted for hate crimes namely stirring up racial hatred, contrary to s19(1) of the Public Order Act 1986, and for racially aggravated criminal damage. Aged 34, he had no previous convictions and in the words of the judge “there is much to admire about your determination to provide for your family and to succeed in business”. However, he was sentenced to two years’ imprisonment after the judge concluded that “your mindset is that of a racist and white supremacist. You hold Nazi sympathies and you are an antisemite.” All of which may have been true, but it should have been irrelevant to the criminal law which should not be criminalising thoughts and beliefs.

The evidence for which Melia was sentenced to a significant term of imprisonment was the words that were printed on his stickers, which he encouraged supporters to download and display to the public. These words were clearly statements of his political opinion and belief:

• “Mass immigration is white genocide”
• “Second generation, third, fourth, you have to go back”
• “Labour loves Muslim rape gangs”
• “Stop anti-white rape gangs”
• “They seek conquest not asylum”
• “Britain is under occupation” printed over a Star of David
• “Antisemitism is caused by semitism”
• “Small hat, big problems” beneath a drawing of a kippah with a Star of David
• Free speech is the lifeblood of democracy because in a democracy opinions and beliefs, even extreme ones, are to be challenged with speech, not with criminal convictions and terms of imprisonment.

The mission creep of hate speech laws

The history of Britain’s hate speech laws is summarised in table 1. From the outset in 1965 they were designed to deal with political behaviour, which is why the Attorney General’s consent was required for a prosecution. As the Home Secretary of the time said: this law is aimed at “propaganda campaigns” which “over a period time, engenders the hate which begets violence”.[54] But in truth, “hate” rarely begets violence. Most people “hate” lots of things, including political ideas, but this emotion does not normally cause them to cross the threshold into violence. In any event, if violence is the problem, the solution is simply to criminalise that, which the law is well equipped to do.

Table 1: The evolution of hate speech laws

Prosecutions brought under hate laws in Britain since 1965 highlight their political nature. The first person to be prosecuted for an alleged hate crime was Christopher Britton, a seventeen-year-old labourer, who had attached a political tract, Blacks Not Wanted Here, to the front door of Southall’s MP. He was convicted of inciting racial hatred, until the Court of Appeal quashed his conviction on a technicality.[66]

Britain’s first hate speech law had a fairly immediate chilling effect on political discourse. For example, the Racial Preservation Society went out of its way to stay on the right side of the law by expressing their views without overt threats, abuse or insults as they “sought a humane solution to the problem of anti-white immigration”. But this did not prevent four of its members being tried for incitement to racial hatred when their broadsheet referred to the dangers of “race mixing” and to politicians as “race levellers” unwilling to face problems such as, in their view, excessive immigration and genetic differences between races. The defendants were all acquitted in March 1968 but the chilling effect of the prosecution was clear and when in April 1968 the London dockers marched in support of Enoch Powell following his “Rivers of Blood” speech one of their placards stated: “Race Relations Act is a Gag on Free Speech”.[67]

This early history highlighted the threat to speech posed by laws that sought to criminalise political ideas. But British politicians learned the wrong lessons: they simply made the charge easier to prove when the Race Relations Act 1976 ditched the requirement for the prosecution to prove that the defendant had intended to stir up racial hatred, in favour of a much lower test of likelihood.[68] The Public Order Act 1986 brought back the requirement to prove intent but only as an alternative to the lower test of likelihood, so that in practice the issue of intent never has to be proved.[69]

Box 7: Existing hate crime offences (England & Wales)[70]

Since 1986 hate crime law has expanded significantly in terms of (i) which identities are protected and (ii) how they are protected (with aggravated offences and enhanced sentencing). These expansions were particularly pronounced under the Blair-Brown governments of 1997-2010, as shown in table 1. First, what started out in 1965 as a substantive offence based on race expanded to create substantive offences based on:

• religion in 2001
• sexual orientation and disability in 2003
• gender in 2012.[71]

Secondly, what started out as a specific offence of stirring up hatred expanded in scope to enable:

• several existing offences (assaults, criminal damage, public order and harassment) to be treated as far more serious, aggravated offences if the defendant had demonstrated a hostility or motivation based on the victim having a protected identity; and
• sentences to be enhanced if the offender demonstrated a hostility towards the protected identity. For example, plain old murder has a minimum term of 15 years, but if it is racially aggravated this increases to 30.[72]

All hate speech laws need to be repealed as shown below.

3^rd proposal: Hate speech laws to be repealed

Main points

• Free speech is not a matter of discrimination. We must remove the religion and belief protections from the Equality Act and replace them with a tort of unlawful interference with free speech.
• We must introduce a costs regime, based on the Aarhus Convention’s protections for environmental claims, which will protect those whose speech is interfered with and deter interference in the first place.

Whereas criminal law is concerned with harm to the public, the civil law is principally concerned with harm to a private or particular interest. The starting point for “workplace speech” is that individuals should be free to express their likes and dislikes, thoughts, opinions, and beliefs without restraint in their non-working lives. This, as above, is because free speech is the lifeblood of democracy.

But it is also true that there can be compelling countervailing rights that operate to restrict this presumption in favour of free speech. For example, if a full-time judge wrote an opinion column for a daily newspaper arguing for open borders we would expect him to face disciplinary consequences on grounds of harming the judiciary’s reputation for political impartiality. But the existence or extent of any such rule that might be imposed in the workplace and beyond is not straightforward. Most would agree that judges should not express controversial political opinions but that most employees should be free to do so in their private lives. But what about Gary Lineker, a contractor who, when he presented Match of the Day was widely seen as a public face of the BBC, an organisation struggling to retain its reputation for political impartiality? And what about the rights of a political party to expel those who openly challenge the policies of their chosen organisation?[73]

This is an area where the law needs to outline clear and easy to understand principles which then give judges latitude to apply them to any given set of circumstances.

Box 8: Existing Equality Act protection for the protected characteristic of belief

The problem with the law of “belief discrimination” as currently drafted in the Equality Act (see box 8) is fourfold.

Free speech is not an issue of either discrimination or equality

Philosophically, the right to speak without censure is not about discrimination at all; it is an independent right, rather than one based on unequal treatment. Of course, a discriminatory approach towards a person’s speech might be relevant, as evidencing unreasonableness. The employer who tolerated woke proselytising in the workplace whilst clamping down on any challenges by the anti-woke would in most instances be discriminating on an improper basis.[74] But discrimination is not always the gravamen of the issue with speech. If an employer bans all staff from expressing political views on social media it would not have been guilty of discrimination, but it would probably have emasculated without justification the free speech rights of all its staff.

Due to its unique quality, free speech protection in civil proceedings does not belong in the Equality Act but should be transferred to the proposed Free Speech Bill. Judges in courts and the employment tribunal will therefore recognise that they are dealing with the constitutionally important issue of free speech, rather than the philosophically different issue of discrimination.

Free speech is not just about “beliefs”

The Equality Act currently protects a “religion or belief”. In other words, it does not protect speech in general that falls short of being a manifestation of either a religion or belief. This threshold is a consequence of Parliament giving effect to article 9 of the ECHR which is concerned to protect freedom of thought, conscience, and religion. Parliament did not, when passing the Equality Act, expressly seek to protect article 10.

The problem created is highlighted by the unsuccessful belief discrimination claim brought by Karen Sunderland in the employment tribunal. Her employer dismissed her in 2021 after discovering her Twitter posts from 2017/18 in which she had expressed views on religion, gender, free speech, and Islam. For example she tweeted: “The thing is, religion undermines the hard fought for values and tolerances of progressed countries… Suppresses free-speech and is autoimmune from criticism. This is toxic and where the hostility comes from. Religion is totalitarian. Islam has become the new Nazism”.

None of Ms Sunderland’s tweets had a direct relationship to her employment, which did not begin until some years later. But her belief discrimination claim failed to satisfy the threshold of being a “belief” because the tribunal concluded that she had expressed opinions or views rather than something capable of being a belief in conservatism.[75]

The proposed Free Speech Bill will overcome this limitation by protecting “the right to impart ideas, opinions or information by means of speech, writing or images”, whether or not the ideas etc are manifestations of a religion or belief.

Free speech needs protection with a tort of unlawful interference

The current suite of protections against belief discrimination is set out in box 8. They have evolved over decades, primarily to address issues of race and sex discrimination. They are impenetrable to the layperson who invariably has to instruct expensive lawyers to make them intelligible and actionable.

The proposed Free Speech Bill will create a new tort of unlawful interference with the right of free speech. As shown in the fourth proposal box this tort will be easy to understand and will presume an unlawful interference with the right of free speech unless the person doing the interference can show that it was prescribed and necessary. These two requirements already exist in workplace discrimination cases and the most recent judgment from the Court of Appeal shows that judges should be able to apply them so as to give adequate protection to those who would otherwise fall prey to progressives and others who do not appreciate the constitutional importance of free speech.[76]

4^th proposal: Create a tort of unlawful interference with the right of free speech

Claimants in free speech litigation need costs protection

The usual rule in litigation is that the winner recovers their legal costs from the losing party. This has the advantage of discouraging weak claims (where the loser will have to pay the winner’s legal costs as well as its own) and of incentivising the early settlement of strong claims (for the same reason). Thus, the principle of “cost-shifting” has its merits. However, it also acts to prevent most people from ever issuing a claim because no matter how strong the case is the risk of losing, no matter how small, and of then having to pay the winner’s costs, over which he had no control, is a powerful disincentive.

In the field of employment law this problem was originally intended to be mitigated by making employment tribunal claims a “no-costs” jurisdiction, meaning that absent unreasonable conduct a losing claimant would not have to pay the winner’s legal costs. The original intention was also that claimants in “industrial tribunals”, as they were called when created in 1964[77], would be able to secure funding for unfair dismissal claims from their trade unions to assist them with expertise and funding.[78]

These protections are not adequate for belief discrimination claims where any damages awarded can be modest (unless there has been a significant loss of future earnings); respondents can often afford expensive lawyers; and trade unions are invariably either not available or are uninterested in assisting. Furthermore, claimants who are the victims of free speech infringements beyond the workplace, namely by service providers, have to issue claims in the county court where they are subject to the usual rules of cost-shifting. Cost shifting also applies to judicial review claims, which may provide an appropriate remedy in a free speech case.

The proposed Free Speech Bill will overcome these problems with two reforms to facilitate legal redress for an unlawful interference with the right of free speech. In employment tribunal claims successful claimants will be able to recover their reasonable legal costs.[79] In the event of the claimant losing a free speech claim the tribunal will be bound by the existing no-costs provisions.

Secondly, for free speech claims in the civil courts a system of cost-capping will be introduced. The courts are already familiar with these provisions in claims involving environmental issues, which are covered by the 1998 Aarhus Convention.[80] It works by imposing limits on the amount of costs that a party may be ordered to pay where the claimant in an environmental claim is a member of the public. The most that a (losing) claimant can be ordered to pay is £5,000 and the most that a (losing) defendant can be ordered to pay is £35,000. However, the court has a discretion to vary these amounts so long as the altered amounts do not make the proceedings prohibitively expensive for the claimant”.[81]

5^th proposal: Free speech costs regimes

Main points

• Britain’s free speech approach is often unfairly criticised in comparison to the American first amendment or Article 10 of the ECHR.
• Britain has done more than almost any other nation to found and develop the tradition of free speech.
• American and European approaches are incompatible with Britain’s system of parliamentary sovereignty in which no Parliament can bind its successor.

There is a tendency, particularly within the academic and practitioner worlds, to belittle the British approach to free speech when compared to approaches in America with its first amendment and in Europe with the ECHR. Geoffrey Robertson KC is a standard bearer for knocking the British tradition. In his recent book Lawfare[82] he takes issue with former justice secretary Dominic Raab who described free speech as a “quintessentially UK right… a unique and precious liberty on which the UK has historically placed great emphasis in our traditions.”[83] This, says Robertson “is nonsense. Magna Carta was silent on the subject in 1215”. Robertson might as well have denounced Magna Carta for being silent on motor cars and the mini-skirt. Free speech is of far more recent vintage and moreover, the British did much to identify and advance it, as is shown in box 9.

Box 9: Free speech – a very English tradition

Judicial attitudes to free speech

There is force in the argument that the English judiciary were at times less keen on free speech than on other freedoms such as personal liberty, freedom of contract, and property rights, all of which were robustly protected in common law. The problem is well illustrated by Chief Justice Lord Mansfield who ruled in 1772 that in the absence of any “positive law” that deprived a man of his liberty, the slave before him was to be freed.[88] Yet in a series of other judgments Mansfield made clear his belief in the “vigorous enforcement of the seditious libel laws as a way of supporting the government and arresting what he saw as a decline in the moral condition of the country.”[89]

But in this regard Lord Mansfield and the judges who came before and after him were merely reflecting the temper of their age, in a way that reflects the common law approach of the English judiciary. In the twentieth century the common law’s approach to general speech was good. For example, they honed the common law offence of seditious libel to align it with the stance of this paper of requiring general speech to incite violence before reaching a criminal threshold. In 1909 a High Court judge directed the jury that the test to be applied on a seditious libel charge:

[The test] is not either the truth of the language, or the innocence of the motive with which he published it, but the test is this: was the language used calculated… to promote public disorder or physical force or violence in a matter of State?[90]

Accordingly, the offence of seditious libel—that had been used repeatedly in earlier times, such as to drive the English radical, Thomas Paine, into exile in France—fell out of use, particularly after the case of Caunt. The facts of Caunt have an echo today. In 1947 a wave of indignation and disorder aimed at Jews swept through Britain in response to the murder of two British sergeants in Palestine by members of the Irgun, fighting to create the state of Israel. Violence broke out in Liverpool, Manchester, and elsewhere, for which a number of people were receiving stiff sentences for what a Lancashire magistrate described as these “un-British and un-patriotic acts”. This prompted the editor of an obscure newspaper in Lancashire to publish an avowedly antisemitic editorial:

On the morn of the announcement of “another catalogue of pains and penalties” there is very little about which to rejoice greatly except the pleasant fact that only a handful of Jews despoil the population of the Borough! The foregoing sentence may be regarded as an outburst of anti-Semitism. It is intended to be and we make no apology, neither do we shirk any responsibility nor repercussions.… If British Jewry is suffering today from the righteous wrath of British citizens, then they have only themselves to blame for their passive inactivity. Violence may be the only way to bring them to the sense of their responsibility to the country in which they live.

The jury acquitted the editor on a seditious libel charge after the judge, Norman Birkett, directed it firstly to bear in mind the importance of free speech: “it is in the highest degree essential, and I cannot overemphasise the importance of it to you, that nothing should be done in this Court to destroy or weaken the liberty of the Press.”[91] Then the judge stated that the question was whether the defendant had “published that libel with the intention of promoting violence by stirring up hostility and ill-will between different classes of His Majesty’s subjects”.[92]

But it was not just the crime of seditious libel that the twentieth century judiciary were holding at bay in the cause of free speech:

• In 1923, in the context of conflict between Britain and Irish nationalists in the Irish Free State, the court declined to give effect to a law, on constitutional grounds, but only after noting that: “You really believe in freedom of speech, if you are willing to allow it to men whose opinions seem to you wrong and even dangerous…”[93]
• In 1971, play at the Wimbledon tennis championships was interrupted for a few minutes whilst about ten protesters stayed on the court with banners, placards and leaflets protesting about South Africa’s policy of apartheid. The spectators made loud and hostile shouts and when the police removed the lead protestor, Dennis Brutus, some attempted to strike him. The magistrates concluded however, that Brutus’ behaviour had not been insulting within the meaning of the Public Order Act 1936, section 5 (box 1 above). On an appeal the highest court in the land, the House of Lords, declined to disturb the acquittal after noting that Parliament had addressed “the difficult question of how far freedom of speech or behaviour must be limited in the general public interest”. With this balancing exercise in mind the magistrates had been entitled to conclude that Brutus’ behaviour did not fall on the criminal side of the divide.[94]
• In 1994 the Court of Appeal relied on free speech to dismiss an injunction application sought by a mother to prevent the identification of the father of her son (who was a convicted paedophile). Hoffmann LJ gave a resolute defence of free speech:

But a freedom which is restricted to what judges think to be responsible or in the public interest is no freedom. Freedom means the right to publish things which government and judges, however well motivated, think should not be published. It means the right to say things which “right-thinking people” regard as dangerous or irresponsible.[95]

• In 1999 the case of three Christian women came before the High Court after they had been convicted for a public order matter. Their preaching on the steps of Wakefield Cathedral had attracted a hostile crowd of over one hundred. The High Court overturned the appellant’s conviction after Lord Justice Sedley noted that a police officer’s right to intervene arose only if the preachers “are threatening violence or behaving in a manner that might provoke violence”. In his now celebrated words he proceeded to note, with references to the nation’s traditions, that:

Free speech includes not only the inoffensive but the irritating, the contentious, the eccentric, the heretical, the unwelcome and the provocative provided it does not tend to provoke violence. Freedom only to speak inoffensively is not worth having. What Speaker’s Comer (where the law applies as fully as anywhere else) demonstrates is the tolerance which is both extended by the law to opinion of every kind and expected by the law in the conduct of those who disagree, even strongly, with what they hear. From the condemnation of Socrates to the persecution of modern writers and journalists, our world has seen too many examples of State control of unofficial ideas. … We in this country continue to owe a debt to the jury which in 1670 refused to convict the Quakers William Penn and William Mead for preaching ideas which offended against State orthodoxy.[96]

The European Convention on Human Rights

Geoffrey Robertson has declared that since Britain is “a nation largely bereft of great cases or constitutional guarantees, we have few explanations, other than from Milton, of the rationale for free speech”. Having overlooked the towering influence of John Stuart Mill and others he then argues: “The best we have done is to borrow. The 1998 Human Rights Act enshrines in UK law Article 10 of the European Convention”. Yet the frequently held view that the ECHR taught Britain a lesson in free speech is misleading for three reasons.[97]

First, the cases cited are invariably ones that are not concerned with either general speech or communications speech. The ECtHR cases relied on by those with a dim view of domestic judicial attitudes towards free speech are ones where there has been a potentially strong countervailing interest, such as national security[98], reputation (libel)[99], confidentiality[100], or the maintenance of judicial authority[101]. Of these cases the one that stands out arose from the Spycatcher litigation in which the British government tried in 1986 to suppress publication of a memoir written by former spy Peter Wright who had worked for MI5. The government persisted with its suppressive efforts even after the memoir had been published in America. Granted, this case did not portray the British judiciary in a good light, particularly in light of the ECtHR’s ruling of 1991, which found that the former spy’s Article 10 rights had been unlawfully abridged. But it would be wrong to draw too many conclusions from this case, especially as the House of Lords ruling went against free speech by 3 votes to 2 and Lord Bridge, in the minority, gave a trenchant defence of it:

Freedom of speech is always the first casualty under a totalitarian regime. Such a regime cannot afford to allow the free circulation of information and ideas among its citizens. Censorship is the indispensable tool to regulate what the public may and what they may not know. The present attempt to insulate the public in this country from information which is freely available elsewhere is a significant step down that very dangerous road. The maintenance of the ban, as more and more copies of the book Spycatcher enter this country and circulate here, will seem more and more ridiculous. If the Government are determined to fight to maintain the ban to the end, they will face inevitable condemnation and humiliation by the European Court of Human Rights in Strasbourg. Long before that they will have been condemned at the bar of public opinion in the free world.[102]

Secondly, in a democracy the courts should not be making new law, save that judges have a role to develop the common law. But most of the ECtHR cases footnoted above are ones that required parliamentary intervention to modernise and update laws that had become otiose. For example, the Sunday Times thalidomide case was concerned with the maintenance of judicial authority arising from the publication of information that was said to be prejudicial to ongoing civil litigation. Distillers—who manufactured thalidomide and were being sued by those claiming to have been damaged by it—obtained an injunction in 1972 to prevent the Sunday Times from publishing an article about the drug’s harm.

The High Court noted that it was not for it to balance the competing interest between the press and others that arose in a contempt of court case. Indeed, before and during the ongoing litigation politicians were aware that the law of contempt needed reform, as was confirmed in three reports:

• 1969: a government inquiry, The Law of Contempt in Relation to Tribunals of Inquiry (Cmnd 4078)
• 1974: a judge-led report (by Phillimore LJ who in 1973 had supported the Court of Appeal’s decision to discharge the injunction against The Sunday Times) proposed various reforms having concluded that the balance of interests had moved too far against freedom of the press (Report of the Committee on Contempt of Court (Cmnd 5794)).
• 1978: a government Green Paper, Contempt of Court: A Discussion Paper (Cmnd 7145) rehearsed the recommendations made in 1974

The ECtHR’s ruling was delivered in 1979 and by a narrow majority of 11 to 9 it concluded that the prohibition on publication by the Sunday Times had been an unlawful interference with press freedom. An often-overlooked point is that that was also the view that three judges in the Court of Appeal had come to in 1973 (Lord Denning, Phillimore, Scarman LJJ), before being overturned by the House of Lords the following year. But the key fact here is that the issue of what could be published and said about ongoing litigation should have been addressed by Parliament much earlier than was ultimately the case when it passed the Contempt of Court Act 1981. The problem was with Parliament, not the judiciary.

Thirdly, in the sort of cases that this paper is primarily concerned with, the ECtHR is not helpful. In particular, the ECtHR has been an enthusiastic supporter of hate speech laws, which in recent decades have become the primary threat to free speech in this country. As Jacob Mchangama notes, the ECtHR “has ruled against the speaker in 62% of hate speech cases brought before it”. This is because the ECtHR has ruled that speech should readily yield to so called “hate speech” restrictions on the basis that:

Tolerance and respect for the equal dignity of all human beings constitute the foundations of a democratic, pluralistic society. That being so, as a matter of principle it may be considered necessary in certain democratic societies to sanction or even prevent all forms of expression which spread, incite, promote or justify hatred based on intolerance…[103]

We saw above how Mark Norwood was convicted in 2003 for aggravated hostility towards Islam for a poster in response to the 9/11 terrorist attacks that read “Islam out of Britain – Protect the British People”. He took his case to the ECtHR arguing that free speech embraces the right to be offensive, unwelcome, and provocative so long as it does not tend to provoke violence. The court dismissed the application at a preliminary stage after concluding that the poster constituted unprotected speech as it was “a general, vehement attack against a religious group, linking the group as a whole with a grave act of terrorism” which “is incompatible with the values proclaimed and guaranteed by the Convention, notably tolerance, social peace and non-discrimination”.[104]

In short, the ECHR cannot address the UK’s free speech recession, indeed it is more likely to worsen it. Any government that is serious about protecting free speech will withdraw the UK from the ECHR.

The American approach

The philosophical approach of Milton and Mill became a legal right when it informed the American Bill of Rights in 1791 with its first amendment endorsement of free speech:

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press.

Yet in America it was not until the early twentieth century that the Milton-Mill approach became a legal right that could defeat hate speech. This happened under the clear-sightedness of Justice Oliver Wendell Holmes (1841-1935) in a series of rulings which culminated in Whitney v California in 1927[105], as noted above.

On free speech issues the British judiciary is often wary of the approach taken by American courts under the First Amendment because “the approach in this jurisdiction and under the [ECHR] is different.”[106] There is much to admire in the approach of American courts to free speech, but the constitutional approach on which it hangs cannot and should not be adopted in Britain. The UK does not have a written constitution whereby something like the first amendment could be entrenched or protected from easy amendment. The essence of our parliamentary democracy is that no Parliament may bind another. This is a strength of our system, which allows Parliament to act on the basis of assumed need at any particular time. And in any event the common law position is that free speech prevails by default in the absence of any restrictions established either by Parliament, or, until the UK withdraws from the ECHR, by Strasbourg. As Hoffmann LJ observed in 1994:

It cannot be too strongly emphasised that outside the established exceptions, or any new ones which Parliament may enact in accordance with its obligations under the [European Convention on Human Rights], there is no question of balancing freedom of speech against other interests. It is a trump card which always wins.[107]

This paper has sought to stress that the UK’s current free speech crisis is an exception to the norm and that for much of the nineteenth and twentieth century Britain probably led the world in advocating and supporting the importance of free speech. The seeds of the current recession were sown in 1965 when the Race Relations Act took the criminal law into new and dangerous territory by outlawing opinions and beliefs on race without any need for violence to have been incited. The Public Order Act 1986 challenged speech more generally by criminalising words that could be “threatening, abusive or insulting” without any need for the prosecution to establish that the words incited violence. Between 1998 and 2008 the Blair-Brown governments passed no fewer than five statutes to criminalise the “wrong” speech on race, religion, sexuality and disability, to which the coalition government of 2012 did the same for gender.

When speech of a certain kind, dubbed “hateful”, is criminalised and the public sees ordinary folk being arrested, prosecuted, convicted, and imprisoned for things they have said, the culture changes. Conversation becomes cancellation and debate yields to denunciation. Society reaps what statute sows.

This is not a problem caused by the judiciary. Judges do not make the law, they apply it. If they are required to enforce “speech crimes” they will do it, and moreover they will express justifications for doing it. That is their function in a democracy. Neither is this a problem that can be solved by appealing to judges in Strasbourg. The ECHR’s approach to hate crimes is far worse than it is in Britain.

The answer to the UK’s free speech recession in a future Free Speech Act is simple.

First, in criminal law, we must:

• Repeal section 5 of the Public Order Act 1986.
• Repeal three existing communications offences under the Malicious Communications Act 1988, the Communications Act 2003, and the Online Safety Act 2023, and replace them with one offence of communication that is intentionally menacing and targeted.
• Repeal all so-called “hate speech” laws outlined in Table 1 of this paper.

Second, in civil law, we must:

• Remove the protections for religion and belief from the Equality Act and replace them with a tort of unlawful interference with free speech.
• Introduce a costs regime to protect those whose speech is interfered with and to deter interference in the first place.

These measures will protect free speech and empower ordinary folk to sue those who have the audacity to challenge one of their most fundamental rights: the right to speak freely. Just as bad laws beget a bad culture, so good laws beget a good one.

With regards to the UK’s free speech recession this paper began by noting that one of our politicians recently asked, “At what point did we become North Korea?” The question should now be: “At what point will we once again become Britain?”

“Allison Pearson sues police chief over social media investigation”. The Telegraph. 24 March 2026. https://www.telegraph.co.uk/news/2026/03/24/allison-pearson-sues-essex-police-chief-social-media/?msockid=2ba0cfc8cee0604b2989d8f9cf8461df

Attorney General v Guardian Newspapers Ltd (Spycatcher No. 1) [1987] UKHL 13, [1987] 1 WLR 1248. https://www.bailii.org/cgi-bin/format.cgi?doc=/uk/cases/UKHL/1987/13.html&query=spycatcher

Attorney General v Observer Ltd (Spycatcher No. 2) [1990] 1 AC 109, Lord Goff 283E. https://www.bailii.org/cgi-bin/format.cgi?doc=/uk/cases/UKHL/1988/6.html&query=spycatcher

BBC. “‘Gay’ police horse case dropped”. 12 January 2006. http://news.bbc.co.uk/1/hi/england/oxfordshire/4606022.stm

BBC. “Harry Miller: Legal victory after alleged transphobic tweets”. 20 December 2021. https://www.bbc.co.uk/news/uk-england-lincolnshire-59727118

BBC. “Man jailed for anti-Semitic tweet to Luciana Berger MP”. 20 October 2014. https://www.bbc.co.uk/news/uk-england-29695846

Bracchi, Paul. “It may have been a victory for free speech, but why did breakfast insult of Muslim’s faith case ever come to court?” Daily Mail. 10 December 2009. https://www.dailymail.com/news/article-1234680/It-victory-free-speech-did-breakfast-insult-Muslims-faith-case-come-court.html

Bromovsky, Lettice. “Arrested for ‘racist’ tweet”. Daily Mail. 26 September 2025. https://www.dailymail.com/news/article-15136539/police-arrest-man-post-Hamas-Islam-social-media.html

Brutus v Cozens [1973] AC 854, HL, Lord Reid, 862D. https://www.bailii.org/uk/cases/UKHL/1972/6.html

Butlin, Sarah Fraser, Catherine Barnard, and Maayan Menashe. Reimagining Employment Dispute Resolution and Enforcement. London: Hart Publishing. 2026. https://openaccess.city.ac.uk/id/eprint/37528/1/9781509990511_web.pdf

Caunt, James. An Editor on Trial: Rex v. Caunt, Alleged Seditious Libel. London: Morecambe Press Ltd. 1947.

Cobban v DPP [2024] EWHC 1908 (Admin). https://www.bailii.org/ew/cases/EWHC/Admin/2024/1908.html

Cunningham, Naomi and Shazia Khan. “Legal costs as damages for discrimination”. ELA Briefing. Vol 31 7. Accessed 23 June 2026. https://www.outertemple.com/wp-content/uploads/2024/08/LEGAL-COSTS-AS-DAMAGES-FOR-DISCRIMINATION-ELA-Briefing-August-2024.pdf

Deborah Hicks v DPP [2023] EWHC 1089 (Admin). https://www.bailii.org/ew/cases/EWHC/Admin/2023/1089.html

DPP v Collins [2005] EWHC 1308 (Admin). https://www.bailii.org/ew/cases/EWHC/Admin/2005/1308.html

DPP v Collins [2006] UKHL 40. https://www.bailii.org/uk/cases/UKHL/2006/40.html

DPP v Hamit Coskun [2026] EWHC 427 (Admin), Warby LJ and Obi J. https://www.judiciary.uk/wp-content/uploads/2026/02/DPP-v-Coskun-2026-EWHC-427-final.pdf

Dr Shahrar Ali v Members of the Green Party, County Court at Mayor’s and City of London, 9 February 2024, HHJ Hellman. https://www.judiciary.uk/wp-content/uploads/2024/02/J00CL858-Ali-v-Reason-and-Nott-judgment-reboot-1-1.pdf

Forbes, Steve. “People are being thrown in UK prisons over what they’ve said online. Can free speech be saved?” Forbes. 9 September 2025. https://www.forbes.com/sites/steveforbes/2025/09/09/people-are-being-thrown-in-uk-prisons-over-what-theyve-said-online-can-free-speech-be-saved/

Grainger v Nicholson [2010] ICR 360. https://www.bailii.org/uk/cases/UKEAT/2009/0219_09_0311.html

Harry Hammond v DPP [2004] EWHC 69 (Admin), May LJ. https://www.bailii.org/ew/cases/EWHC/Admin/2004/69.html

Hate speech. ECtHR Factsheet. November 2023. https://www.echr.coe.int/documents/d/echr/FS_Hate_speech_ENG

HC Deb, 16 November 1936, vol. 317, cc1349-473. https://api.parliament.uk/historic-hansard/commons/1936/nov/16/public-order-bill#column_1349:~:text=with%20a%20view%20to%20checking%20action%20which%20is%20calculated%20to%20destroy%20the%20liberty%20that%20we%20wish%20to%20preserve.4

HC Deb, 3 May 1965, Hansard, vol. 711, cols. 926-43. https://hansard.parliament.uk/Commons/1965-05-03/debates/28e9ee04-70cb-4d4c-b514-c681106cd10f/RaceRelationsBill

Higgs v Farmor’s School [2025] EWCA Civ 109. https://www.bailii.org/ew/cases/EWCA/Civ/2025/109.html

HM Government. “Anti-terrorism, Crime and Security Act 2001”. S39. https://www.legislation.gov.uk/ukpga/2001/24/contents/enacted

HM Government. Crime and Courts Act 2013″. https://www.legislation.gov.uk/ukpga/2013/22/contents

HM Government. “Crime and Disorder Act 1998”. https://www.legislation.gov.uk/ukpga/1998/37/contents/enacted

HM Government. “Crime and Disorder Act 1998”. https://www.legislation.gov.uk/ukpga/2026/20/contents/enacted

HM Government. “Criminal Justice Act 2003”. https://www.legislation.gov.uk/ukpga/2003/44/contents/enacted

HM Government. “Criminal Justice and Immigration Act 2008”. https://www.legislation.gov.uk/ukpga/2008/4/contents/enacted

HM Government. “Hate Crime and Public Order (Scotland) Act 2021”. https://www.legislation.gov.uk/asp/2021/14/contents

HM Government. “Industrial Relations Act 1971”. https://www.legislation.gov.uk/ukpga/1971/72/contents/enacted

HM Government. “Industrial Training Act 1964”. https://www.legislation.gov.uk/ukpga/1964/16/contents/enacted

HM Government. “Justices of the Peace Act 1361”. https://www.legislation.gov.uk/aep/Edw3/34/1/data.html

HM Government. “Post Office (Amendment) Act 1935”. https://www.legislation.gov.uk/ukpga/Geo5/25-26/15/pdfs/ukpga_19350015_en.pdf

HM Government. “Public Order Act 1986”. https://www.legislation.gov.uk/ukpga/1986/64/contents/enacted

HM Government. “Race Relations Act 1965”. https://www.legislation.gov.uk/ukpga/1965/73/contents/enacted

HM Government. “Race Relations Act 1976”. https://www.legislation.gov.uk/ukpga/1976/74/enacted

HM Government. “Racial and Religious Hatred Act 2006”. https://www.legislation.gov.uk/ukpga/2006/1/contents/enacted

HM Government. “Sentencing Act 2020”. https://www.legislation.gov.uk/ukpga/2020/17/contents

HM Government. “Sentencing Act 2020”. Schedule 21. https://www.legislation.gov.uk/ukpga/2020/17/schedule/21/data.pdf

Johnston, Philip. Feel Free to Say It: Threats to Freedom of Speech in Britain Today. London: Civitas. 2013. https://www.civitas.org.uk/publications/feel-free-to-say-it/

Karen Sunderland v The Hut.Com Limited. Employment Tribunal 2300911/2022. 24 May 2023. https://assets.publishing.service.gov.uk/media/6481d85a5f7bb7000c7fa72b/Ms_K_Sunderland_v__1__The_Hut.com_Limited__2__Ms_P_Cohen_2300911-2022_Judgment___Reasons.pdf

Lester, Anthony and Geoffrey Bindman. Race and Law. London: Penguin Books. 1972

Locke, John. An Essay Concerning Human Understanding. Project Gutenberg. 1689. https://www.gutenberg.org/files/10615/10615-h/10615-h.htm

Maya Forstater v CGD Europe [2022] ICR 1. https://www.bailii.org/uk/cases/UKEAT/2021/0105_20_1006.html

Mchangama, Jacob. Free Speech: A Global History from Socrates to Social Media. London: Basic Books. 2022.

Metropolitan Police Act 1839. Clause LIV (13). https://www.legislation.gov.uk/ukpga/Vict/2-3/47/enacted?view=plain

MGN Limited v. The United Kingdom, Application no. 39401/04. European Court of Human Rights. 18 January 2011. https://globalfreedomofexpression.columbia.edu/wp-content/uploads/2021/12/CASE-OF-MGN-LIMITED-v.-THE-UNITED-KINGDOM.pdf

Mill, John Stuart. On Liberty. Project Gutenberg. 1859. https://gutenberg.org/files/34901/34901-h/34901-h.htm

Milton, John. Areopagitica. Project Gutenberg. 1644. https://www.gutenberg.org/files/608/608-h/608-h.htm

Ministry of Justice. Human Rights Act Reform: A Modern Bill of Rights. Consultation response. June 2022. https://consult.justice.gov.uk/human-rights/human-rights-act-reform/results/modern-bill-rights-consultation-response.pdf

Norfolk, Andrew and Greg Hurst. “Race hate laws to be changed after BNP case fails”. The Times. 11 November 2006. https://www.thetimes.com/best-law-firms/profile-legal/article/race-hate-laws-to-be-changed-after-bnp-case-fails-h6kzqwzmzp8

Norfolk, Andrew. “BNP chief claims acquittal is his victory for freedom”. The Times. 11 November 2006. https://www.thetimes.com/uk/politics/article/bnp-chief-claims-acquittal-is-his-victory-for-freedom-bq2sz883glk

Norwood v DPP [2003] EWHC 1565, Auld LJ. https://www.bailii.org/ew/cases/EWHC/Admin/2003/1565.html

Norwood v United Kingdom. Application no. 23131/03. Lodged 16 July 2003. https://hudoc.echr.coe.int/eng#{%22itemid%22:[%22001-67632%22]}

Pannick, David. “Why freedom of expression is an essential safety valve for society”. The Times. 21 November 2006. https://www.thetimes.com/uk/law/article/why-freedom-of-expression-is-an-essential-safety-valve-for-society-n58q5q35x73

Paul Chambers v DPP [2012] EWHC 2157 (Admin). https://www.bailii.org/ew/cases/EWHC/Admin/2012/2157.html

Poser, Norman. Lord Mansfield: Justice in the Age of Reason. Montreal: McGill-Queens University Press. 2013.

R v Aldred (1909) 22 Cox C.C.1., Coleridge J.

R v Britton [1967] 2 QB 51.

R v Central Independent Television Plc [1994] Fam 192, 203A. https://www.independent.co.uk/arts-entertainment/law-report-broadcast-cannot-be-restricted-regina-v-central-independent-television-plc-court-of-appeal-lord-justice-neill-lord-justice-hoffmann-and-lord-justice-waite-9-february-1994-1394627.html

R v Hamit Coskun. Southwark Crown Court. 10 October 2025. Bennathan J. https://www.judiciary.uk/wp-content/uploads/2025/10/Rex-v-Hamit-Coskun.pdf

R v Home Secretary ex p Simms [2000] 2 AC 115, Ld Steyn. https://www.bailii.org/uk/cases/UKHL/1999/33.html

R v Howell [1982] QB 416.

R v Robert Stewart MP Southwark Crown Court, 23 February 2024, Bennathan J.

R v Rogers [2007] UKHL 8, Lady Hale. https://www.bailii.org/uk/cases/UKHL/2007/8.html

R v Secretary of Home Affairs, ex parte O’Brien [1923] 2 K. B. 361, Scrutton LJ, 382.

Redmond-Bate v DPP [1999] EWHC Admin. https://www.bailii.org/ew/cases/EWHC/Admin/1999/733.html

Robertson, Geoffrey. Lawfare: How Russians, the Rich and the Government Try to Prevent Free Speech and How to Stop Them. London: TLS Books, 2023.

Rufo, Yasmin. “Conservative MP Bob Stewart guilty of racially aggravated public order offence”. BBC. 4 November 2023. https://www.bbc.co.uk/news/uk-england-london-67310954

Simpson, Thomas and Eric Kaufmann. Academic freedom in the UK. London: Policy Exchange. November 2019. https://policyexchange.org.uk/publication/academic-freedom-in-the-uk/

Sky News. “Britain ‘out of whack’ on free speech, warns Nick Clegg”. 10 September 2025. https://www.msn.com/en-gb/news/uknews/britain-out-of-whack-on-free-speech-warns-nick-clegg/ar-AA1MeSz7

Somerset v Stewart (1772) 98 ER 499, Lord Mansfield. https://www.nonhumanrights.org/wp-content/uploads/Somerset-v.-Stewart.pdf

Steel and Morris v United Kingdom (2005) 41 EHRR 22. https://hudoc.echr.coe.int/eng#{%22itemid%22:[%22001-68224%22]}

Steven Thomas v Surrey and Borders Partnership NHS Foundation Trust [2024] EAT 141. https://assets.publishing.service.gov.uk/media/66d97c5cfb86ba5a1f214e47/Mr_S_Thomas_v_1__Surrey_and_Borders_Partnership_NHS_Foundation_Trust_2__Ms_A_Brett__2024__EAT_141.pdf

The Observer and The Guardian v United Kingdom (1991) 14 EHRR 153. https://hudoc.echr.coe.int/eng#{%22itemid%22:[%22002-10248%22]}

The Sunday Times v United Kingdom (the Thalidomide case) (1979) 2 EHRR 245. https://hudoc.echr.coe.int/eng#{%22itemid%22:[%22001-57584%22]}

Times Newspapers Ltd v UK (Nos 1 & 2) [2009] EMLR 14. https://hudoc.echr.coe.int/eng#{%22itemid%22:[%22001-91706%22]}

Tomkins, Adam. On the Law of Speaking Freely. London: Hart Publishing. 2025.

UK Ministry of Justice. Civil Procedure Rules 46, Section IX Costs Limits in Aarhus Convention Claims. Accessed 23 June 2026. https://www.justice.gov.uk/courts/procedure-rules/civil/rules/part-46-costs-special-cases#x

UK Parliament. House of Commons Public Bill Committee, Protection of Freedoms Bill: Memorandum submitted to the Public Bill Committee (PF 7). 23 March 2011. https://publications.parliament.uk/pa/cm201011/cmpublic/protection/memo/pf7.htm

United Nations Economic Commission for Europe. Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters. Aarhus Convention. 25 June 1998. https://unece.org/DAM/env/pp/documents/cep43e.pdf?__cf_chl_f_tk=c7u1X5bM2y4jUks2WiCmlQ5quELsedvLnDXOzPz0HGw-1782918007-1.0.1.1-YURdMVV60lAG6Ijx2ANfkWtQVvT9_VFGgj7I8qrUuLU

Whitney v California [1927] 274 US. https://tile.loc.gov/storage-services/service/ll/usrep/usrep274/usrep274357/usrep274357.pdf

[1] In this paper the term “free speech” also embraces other forms of free expression.

[2] Relying on the precedent set in Maya Forstater v CGD Europe [2022] ICR 1. (link)

[3] Most notably illustrated in Higgs v Farmor’s School [2025] EWCA Civ 109. (link)

[4] Such as when Mr Nicholson succeeded in establishing that his belief in the catastrophic effect of climate change was protected: Grainger v Nicholson [2010] ICR 360. (link)

[5] Steven Thomas v Surrey and Borders Partnership NHS Foundation Trust [2024] EAT 141. (link)

[6] Attorney General v Observer Ltd (Spycatcher No. 2) [1990] 1 AC 109, Lord Goff 283E. (link)

[7] Sky News, “Britain ‘out of whack’ on free speech, warns Nick Clegg”, 10 September 2025. (link)

[8] Steve Forbes, “People are being thrown in UK prisons over what they’ve said online. Can free speech be saved?”, Forbes, 9 September 2025. (link)

[9] Thomas Simpson and Eric Kaufmann, Academic freedom in the UK (London: Policy Exchange, November 2019). (link)

[10] R v Home Secretary ex p Simms [2000] 2 AC 115, Ld Steyn 126F. (link)

[11] R v Home Secretary ex p Simms [2000] 2 AC 115, Ld Steyn, 126E.

[12] BBC, “Harry Miller: Legal victory after alleged transphobic tweets”, 20 December 2021. (link)

[13] “Allison Pearson sues police chief over social media investigation”, The Telegraph, 24 March 2026. (link)

[14] Lettice Bromovsky, “Arrested for ‘racist’ tweet”, Daily Mail, 26 September 2025. (link)

[15] John Stuart Mill, On Liberty, 1859, Project Gutenberg, accessed 22 June 2026. (link)

[16] R v Hamit Coskun, Southwark Crown Court, 10 October 2025, Bennathan J, §19. (link)

[17] Adam Tomkins, On the Law of Speaking Freely, 2025 (London: Hart Publishing, 2025), 78.

[18] Whitney v California [1927] 274 US 357, 375. (link)

[19] Whitney v California, 375.

[20] Whitney v California, 376.

[21] Whitney v California, 376.

[22] See “Justices of the Peace Act 1361”. (link)

[23] R v Howell [1982] QB 416.

[24] Metropolitan Police Act 1839, clause LIV (13). (link)

[25] HC Deb, 16 November 1936, vol. 317, cc1349-473 (link); cited by Philip Johnston, Feel Free to Say It: Threats to Freedom of Speech in Britain Today (London: Civitas, 2013), 24 (link)

[26] Anthony Lester and Geoffrey Bindman, Race and Law (London: Penguin Books, 1972), 351.

[27] One judge has described s5 as “a bumpy collection of elements”. See Norwood v DPP [2003] EWCH 1564, Auld LJ, §17 (link).

[28] “Crime and Courts Act 2013”, s57(2), which came into force on 1 February 2014. (link)

[29] Harry Hammond v DPP [2004] EWHC 69 (Admin), May LJ, §§19, 33. (link)

[30] BBC, “‘Gay’ police horse case dropped”, BBC News, 12 January 2006. (link)

[31] UK Parliament, House of Commons Public Bill Committee, Protection of Freedoms Bill: Memorandum submitted to the Public Bill Committee (PF 7), 23 March 2011. (link)

[32] Paul Bracchi, “It may have been a victory for free speech, but why did breakfast insult of Muslim’s faith case ever come to court?”, Daily Mail, 10 December 2009. (link)

[33] Deborah Hicks v DPP [2023] EWHC 1089 (Admin). (link)

[34] R v Hamit Coskun, Southwark Crown Court, 10 October 2025, Bennathan J. (link)

[35] DPP v Hamit Coskun [2026] EWHC 427 (Admin). (link)

[36] HM Government, “Post Office (Amendment) Act 1935”, s10(2)(a). (link)

[37] BBC, “Man jailed for anti-Semitic tweet to Luciana Berger MP”, 20 October 2014. (link)

[38] DPP v Collins [2006] UKHL 40, Lord Bingham, §7. (link)

[39] DPP v Collins [2005] EWHC 1308 (Admin). (link)

[40] DPP v Collins [2006] UKHL 40.

[41] Paul Chambers v DPP [2012] EWHC 2157 (Admin). (link)

[42] Cobban v DPP [2024] EWHC 1908 (Admin). (link)

[43] It should be noted that in Cobban all the police officers were dismissed from service. Those dismissals may well have been appropriate because as is noted in the section below on civil law, different considerations apply to speech in the workplace and beyond.

[44] DPP v Collins [2006] UKHL 40, Lord Bingham, §11.

[45] Cobban v DPP [2024] EWNC 1908 (Admin), §§63-67. (link)

[46] A message can only be “menacing” if it creates fear or apprehension in those to whom it is communicated, or who may reasonably be expected to see it. See Paul Chambers v DPP [2012] EWHC 2157, LCJ Lord Judge, §30.

[47] Norwood v DPP [2003] EWHC 1565, Auld LJ, §§4-13, 33, 40.

[48] Andrew Norfolk, “BNP chief claims acquittal is his victory for freedom”, The Times, 11 November 2006. (link)

[49] David Pannick, “Why freedom of expression is an essential safety valve for society”, The Times, 21 November 2006. (link)

[50] Andrew Norfolk and Greg Hurst, “Race hate laws to be changed after BNP case fails”, The Times, 11 November 2006. (link)

[51] R v Rogers [2007] UKHL 8, Lady Hale, §12. (link)

[52] Yasmin Rufo, “Conservative MP Bob Stewart guilty of racially aggravated public order offence”, BBC, 4 November 2023. (link)

[53] R v Robert Stewart MP, Southwark Crown Court, 23 February 2024, Bennathan J. (link)

[54] Rt Hon. Sir Frank Soskice QC MP, Second Reading debated, 3 May 1965, Hansard, vol. 711, cols. 926-43. (link)

[55] HM Government, “Race Relations Act 1965”, s6. (link)

[56] The Race Relations Act 1976, s70, inserted s5A into the Public Order Act 1936 (incitement to racial hatred):

A person commits an offence if –

(a) he publishes or distributes written matter which is threatening, abusive or insulting; or

(b) he uses in any public place or at any public meeting words which are threatening, abusive or insulting,

in a case where, having regard to all the circumstances, hatred is likely to be stirred up against any racial group in Great Britain by the matter or words in question. (link)

[57] HM Government, “Public Order Act 1986”, Part III (ss17-29). (link)

[58] HM Government, “Crime and Disorder Act 1998”, ss28-32. (link)

[59] HM Government, “Anti-terrorism, Crime and Security Act 2001”, s39. (link)

[60] HM Government, “Criminal Justice Act 2003”, ss145-146. (link)

[61] HM Government, “Criminal Justice Act 2003”, schedule 21. (link)

[62] HM Government, “Racial and Religious Hatred Act 2006”. The schedule inserted ss29A-29N, as Part 3A, into the Public Order Act 1986. (link)

[63] HM Government, “Criminal Justice and Immigration Act 2008”. Schedule 16 amended Part 3A of the Public Order At 1986. (link)

[64] These provisions are now in the Sentencing Act 2020, s66. (link)

[65] When s145 of the Act comes into force, on a date to be appointed by the Home Secretary, it will amend the Crime and Disorder Act 1998, s28, to extend existing racially and religiously aggravated offences to cover hostility related to disability, sex, sexual orientation and transgender identity. (link)

[66] R v Britton [1967] 2 QB 51. The Court of Appeal quashed the conviction of the “wretched little youth” on the ground that his actions did not amount to publication or distribution to “any section of the public” as required by the section.

[67] Lester and Bindman, Race and Law, 371-72.

[68] HM Government, “Race Relations Act 1976”, s70.

[69] HM Government, “Public Order Act 1986”, Part III, s18.

[70] Similar provisions exist in Scotland under the Hate Crime and Public Order (Scotland) Act 2021. (link)

[71] Although, as shown in table 1, the offence of stirring up hatred was only extended from race to embrace religion and sexuality.

[72] HM Government, “Sentencing Act 2020”, schedule 21, para 3(2)(g) & (h). (link)

[73] Dr Shahrar Ali v Members of the Green Party County Court at Mayor’s and City of London, 9 February 2024, HHJ Hellman. (link)

[74] It might, for example, be a legitimate stance for the Labour Party to take with regards to its members and employees.

[75] Karen Sunderland v The Hut.Com Limited, Employment Tribunal 2300911/2022, 24 May 2023, §190. (link)

[76] Higgs v Farmor’s School [2025] EWCA Civ 109.

[77] Industrial and Training Act 1964, s12. (link)  Although the Tribunals only began to hear unfair dismissal cases with the passage of the Industrial Relations Act 1971, ss 22(1) and 106. (link)

[78] Sarah Fraser Butlin, Catherine Barnard and Maayan Menashe, Reimagining Employment Dispute Resolution and Enforcement (London: Hart Publishing, 2026). (link)

[79] The attempt made by Rachel Meade to secure an award of costs as damages arising from her unlawful discrimination failed recently. For an explanation from her lawyers see Naomi Cunningham and Shazia Khan, “Legal costs as damages for discrimination”, ELA Briefing, vol 31 7, accessed 23 June 2026. (link)

[80] United Nations Economic Commission for Europe, Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters (Aarhus Convention), 25 June 1998. (link)

[81] UK Ministry of Justice, Civil Procedure Rules 46, Section IX Costs Limits in Aarhus Convention Claims, accessed 23 June 2026. (link)

[82] Geoffrey Robertson, Lawfare: How Russians, the Rich and the Government Try to Prevent Free Speech and How to Stop Them, 2023 (London: TLS Books, 2023).

[83] Ministry of Justice, Human Rights Act Reform: A Modern Bill of Rights, consultation response, June 2022, foreword and §38. (link)

[84] Prepublication censorship returned in 1643 and prompted Milton to pen Areopagitica, unlicensed the following year. State licensing of the press did not formally lapse until 1695. See Free Speech: A Global History from Socrates to Social Media, Jacob Mchangama (London: Basic Books, 2022), 105-16.

[85] John Milton, Areopagitica (Project Gutenberg, 1644). (link)

[86] John Locke, An Essay Concerning Human Understanding (Project Gutenberg, 1689). (link)

[87] John Stuart Mill, On Liberty (Project Gutenberg, 1859). (link)

[88] “The state of slavery is of such a nature, that it is incapable of being introduced on any reasons, moral or political; but only positive law… it’s so odious, that nothing can be suffered to support it, but positive law. Whatever inconveniences, therefore, may follow from a decision, I cannot say this case is allowed or approved by the law of England; and therefore the black must be discharged.” Somerset v Stewart (1772) 98 ER 499, Lord Mansfield, 509. (link)

[89] Norman Poser, Lord Mansfield: Justice in the Age of Reason (Montreal: McGill-Queens University Press, 2013), 244.

[90] R v Aldred (1909) 22 Cox C.C.1., Coleridge J.

[91] James Caunt, An Editor on Trial: Rex v. Caunt, Alleged Seditious Libel (London: Morecambe Press Ltd, 1947).

[92] Lester and Bindman, Race and Law, 347.

[93] R v Secretary of Home Affairs, ex parte O’Brien [1923] 2 K. B. 361, Scrutton LJ, 382.

[94] Brutus v Cozens [1973] AC 854, HL, Lord Reid, 862D. (link)

[95] R v Central Independent Television Plc [1994] Fam 192, 203A. (link)

[96] Redmond-Bate v DPP [1999] EWHC Admin, §20. (link)

[97] “The truth is that the law in the United Kingdom has come only slowly and reluctantly towards regarding freedom of speech as a right worthy of protection. UK law has been pulled there in large measure under the influence of the ECHR [which has] done more than any other source to help the common law see that freedom of speech merits at least a degree of legal protection.” Tomkins, On the Law of Speaking Freely, 102.

[98] In this Spycatcher case the ECtHR held that injunctions preventing newspapers from publishing material already widely available abroad violated Article 10. See The Observer and The Guardian v United Kingdom (1991) 14 EHRR 153. (link)

[99] In this McLibel case where McDonalds sued Steel and Morris for libel the ECtHR held that the disproportionate damages award violated Article 10. See Steel and Morris v United Kingdom (2005) 41 EHRR 22 (link).

In another case, the ECtHR criticised the UK’s “multiple publication “rule for online archives as a disproportionate interference with Article 10. See Times Newspapers Ltd v UK (Nos 1 & 2) [2009] EMLR 14 (link).

[100] In MGN v UK (2011) the supermodel Naomi Campbell won damages against the Daily Mirror for violating her privacy. The ECtHR ruled that the UK’s “success fees” in privacy and defamation litigation chilled media freedom and violated Article 10. See MGN Limited v. The United Kingdom, Application no. 39401/04, European Court of Human Rights, 18 January 2011 (link).

[101] The Sunday Times v United Kingdom (the Thalidomide case) (1979) 2 EHRR 245. (link)

[102] Attorney General v Guardian Newspapers Ltd (Spycatcher No. 1) [1987] UKHL 13, [1987] 1 WLR 1248, 1286. (link)

[103] Erbakan v Turkey, 6 July 2006, §5, cited in Hate speech an ECtHR Factsheet, November 2023. (link)

[104] Norwood v United Kingdom, application no. 23131/03, lodged 16 July 2003. (link)

[105] Whitney v California [1927] 274 US 357.

[106] DPP v Hamit Coskun [2026] EWHC 427 (Admin), Warby LJ and Obi J, §27.

[107] R v Central Independent Television Plc [1994] Fam 192, 203D.