***

Tolerance does not come naturally to mankind. For most of human history, what people have believed about morality, about government and about the natural world was laid down by authority, usually by people claiming to speak in the name of God. It is salutary to be reminded that freedom of expression, as an idea, has a short history. It is essentially the legacy of the European scientific revolution of the 17th century and the European enlightenment of the 18th century. These movements rejected the way that men had thought for centuries. They rejected mere authority as a source of truth, in favor of observation, reasoning and rational discourse. It took much longer for the same notions to penetrate people’s ideas about politics, government and law. The legal protection of free speech is barely two centuries old in Europe and North America. And it has only ever been accepted with reservations.

Recent criticisms of Europe by American politicians like JD Vance have reminded us that there is a profound gulf between American and European conceptions of free speech. The United States comes closer than any other country in the world to absolute free speech. The First Amendment to the US Constitution provides that “Congress shall make no law… abridging the freedom of speech”. No ifs, no buts. Since the early 20th century, this has been held to apply to state legislatures as well. The American courts were eventually persuaded to recognize exceptions for incitement to imminent violence, obscene and defamatory material, child pornography and fraud, but that is all. Even these exceptions have been very narrowly interpreted.

The European model has been very different. Freedom of expression is an important but not an absolute public interest in Europe. In principle, it may be overridden by a wide variety of other public interests if there is a sufficiently compelling case for doing so. Article 10 of the European Convention on Human Rights protects freedom of expression, but subject to a long list of potential exceptions: national security, territorial integrity or public safety, the prevention of disorder or crime, the protection of health or morals, the protection of the reputation or rights of others, the protection of confidences, or maintaining the authority and impartiality of the judiciary.

In Britain, the list of exceptions recognized by the law grows longer all the time: speech that discloses secret intelligence, speech that prejudices current legal proceedings, speech that supports proscribed terrorist groups, hate speech directed at racial, religious or sexual minorities, “grossly offensive” messages sent through the post or over electronic networks, and so on. Most of these exceptions are concerned in one way or another with public order and social harmony. In Britain we have not usually specified opinions that it is illegal to hold in public, but other countries have. In Germany, Austria and Belgium it is a criminal offense to deny, trivialize or approve of the Holocaust. Spain, France and Switzerland go even further. It is an offense in those countries to question the existence or gravity of anything which is regarded as a crime against humanity. In many countries it is illegal to display the swastika. In Latvia and Lithuania, that extends to the hammer and sickle and other symbols of Soviet communism.

There is an inevitable tendency for people to want to exclude from freedom of expression whatever ideas they personally cannot abide. Milton’s Areopagitica, published in 1644, is one of the most eloquent pleas for freedom of expression in the English language, but he excluded Catholics and atheists, whose views he regarded as beyond the pale. So did John Locke, although he at least offered reasons for their exclusion. He thought that Catholics were intolerable because they were not interested in truth, and atheists because having no fear of God, they could not be relied upon to honor any social obligations. The authors of the famous early 18th-century tracts known as Cato’s Letters argued the case for free political speech but made an exception for what they called “factious scribblers” and “turbulent babblers”.

The problem, as George Orwell pointed out, is that free speech is worthless unless it extends to things that people do not want to hear. We all have different lists of things that we do not want to hear. Suggested candidates for legally enforced censorship include a variety of allegedly extreme or unacceptable positions such as gender-criticism, climate denial, antisemitism, blasphemy against the Prophet Mohammed or the Koran, totalitarianism, eugenics, satanism and many others. Many of these demands reflect the sensitivities of particular ethnic, social or religious groups who say that their position is special. Censorship is often the child of collective grievances. People who feel that the world has been unjust to them are likely to claim that their position is special too. If some opinions are to be banned as beyond the pale, then why should my list of unacceptable beliefs be privileged over yours? The truth is that if we believe in freedom of expression, we have to tolerate things that we regard as intolerable. Otherwise the whole concept is meaningless.

Since we have to draw a line, where do we draw it? The starting point, I suggest, is to ask why we instinctively believe in freedom of expression in the first place. I think that there are three main reasons.

The first is that constraints on our ability to communicate facts or opinions invade our autonomy as moral agents and thinking creatures. When the constraints are imposed by the state, they imply a relationship between the individual and the state which is morally repellent. They reduce us to the status of mere instruments of other people’s power. Whether the suppression of free speech comes from governments or pressure groups or public opinion, it ushers in a narrow-minded, intolerant and authoritarian world in which the fear of giving offense or challenging existing shibboleths eliminates some of the most creative and original products of the human spirit.

The second reason for believing in freedom of speech is that we live in a democracy whose entire political order depends on the free exchange of ideas. It depends on our ability to freely receive information, on an acceptance of diversity of opinion and on a large measure of tolerance of intellectual and emotional differences. Our collective life depends on the resolution of issues between citizens by marshaling objectively verifiable facts. It depends on ordered debate about their implications under common rules that exclude coercion, and on a culture in which the outcome of our processes of collective decision-making is accepted even by those who disagree with it.

Suggested reading

Inside Britain’s two-tier justice system

By M.L. Friend

A third reason why we believe in freedom of expression is perhaps the most fundamental. It is the one which received its most eloquent expression from Karl Popper’s classic work, The Open Society and its Enemies. No one person or group, however respectable or expert, has exclusive access to wisdom or truth. This is not just because of the limits on the capacity of the human mind. It is also because truth and wisdom are dynamic concepts. They are informed by a stock of experience which is constantly increasing. All statements of fact or opinion are provisional. They reflect the current state of knowledge and experience. But knowledge and experience are not closed or immutable categories. They are inherently liable to change in response to fresh discoveries or insights. Knowledge advances by testing conflicting arguments, not by suppressing them. Understanding increases by exposure to uncomfortable truths. In a world of free expression, some of what people say will certainly be wrong, hurtful or even objectively harmful. But the principle which we would have to accept in order to justify censoring these statements is more damaging than the statements themselves. We cannot have truth without accommodating the possibility of error, or wisdom without accommodating the possibility of folly. In the end we have to accept the implications of human inquisitiveness, creativity and imagination. The alternative is to entrust significant parts of our intellectual world to external authorities whose capacity for objectivity, truthfulness and wisdom is no greater than our own.

Ever since the 17th century, our civilization has been based on the notion that truth is independent of human will. It may be only partly knowable, and more or less difficult to identify, but it exists somewhere out there whether we like it or not. We have built our intellectual world on the footing that we get closest to the truth by objective study, logical reasoning and open debate, by what Mill called the “collision of adverse opinions”. I am not suggesting that this is guaranteed to produce truth or the wisest possible outcomes, but experience suggests that in the long term it is more likely to do so than any system of intellectual controls.

These reasons for believing in freedom of expression suggest a number of conclusions that we should bear in mind when we come to think about exceptions.

In the first place, freedom of expression is concerned with the transmission of information or opinion. Not everything that is spoken or written deserves protection. If I write you a letter or an email consisting of nothing but foul-mouthed abuse, I may be committing an offense under the Malicious Communications Act 1988 or the Communications Act 2003. Low level communications of this kind are devoid of intellectual, cultural or political value. In a sense I may be regarded as having communicated my opinion of you, but that hardly counts as an infringement of my right of freedom of expression.

The second conclusion to be drawn from our reasons for believing in free speech is that we should not suppress the expression of facts or opinions simply because they give offense, however understandable that offense may be. If we are not free to say things that other people regard as objectionable, we have no freedom of speech at all. In a famous statement of principle, Lord Justice Sedley observed in the High Court in a case about a Christian fundamentalist street preacher 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 promote violence. Freedom only to speak inoffensively is not worth having.”

Thirdly, it cannot be a good reason for suppressing some assertion or opinion that we think it is wrong or even that everybody thinks it is wrong. Few propositions are incontrovertibly false, in the sense that they are inherently incapable of being affirmed by hitherto unknown facts or fresh analysis. As a general rule, the best answer to misleading speech is counter-speech, i.e. reasoned refutation. Of course, counter-speech does not always work. The barrage of rubbish generated by the internet, generally mediated by algorithms which serve people up with material that confirms their existing prejudices, is a serious challenge. But counter-speech is more likely to work than suppression. Once upon a time, the authorized consensus was that the sun moved round the earth, because that view was supported by Biblical authority, then regarded as a source of incontrovertible truth. Galileo was condemned for denying it. The same point can be made about statements of opinion. Once upon a time the religious and social consensus justified slavery and the imprisonment of homosexuals, because this had been regarded as obvious for centuries. Once upon a time it was taken for granted in Britain that black people were inherently inferior. These views have almost entirely died out in advanced societies in the face of rational discourse.

In 2021 the Royal Society, Britain’s premier scientific society, called for legislation to ban the dissemination by anti-vaxxers of “false information” about vaccines, by which they meant information rejected by the scientific consensus. The case for vaccination is not inherently irrefutable. But in the current state of our knowledge, the debate has been won hands down by supporters of vaccination. That has only been possible because there has been a debate. Suppressing public opposition to vaccination will simply drive it underground where it will be more difficult to refute. Holocaust denial is another example. It is an absurd position which has been discredited not by legislation but exposure to the evidence. The great American jurist Oliver Wendell Holmes, who wrote some of the greatest judgments of the US Supreme Court in support of free speech, once observed that the guiding principle was that almost all certainty is an illusion. We might always be wrong, he said. All of us can all point to public statements which are untrue, mischievous or absurd, sometimes manifestly so. But manifestness is not a very useful criterion for censoring other peoples’ opinions. Nearly every received opinion which rational examination has discredited was once regarded as manifest by those who held it.

Finally, the concept of free speech assumes that adults are capable of distinguishing what is good from what is bad in the torrent of information to which we are all exposed. In a democratic society, it is necessary to proceed on that assumption, even though we know that it is not always true. In a society which treats every person as entitled to equal consideration and respect, we cannot distinguish between adults according to their degree of intelligence or discernment or their skill in ordering their lives. In a democracy, we have to assume that adults are capable of making discerning choices in their own interest, because the alternative is an authoritarian world in which their choices are taken away from them by the state.

All of this suggests that the threshold for banning the transmission of information or opinion should be exacting. Any exception must, I suggest, be based on some consistent principle, not just on people’s subjective likes and dislikes. The exception should also be consistent with the basic rationale for wanting freedom of expression in the first place. And it must be clearly defined, because if the boundaries of the exception are unclear, we will end up with a freedom that is at least partly discretionary. A freedom which depends on the discretion of the enforcers is no freedom at all.

I suggest that the only speech which we are justified in suppressing by law is speech which is coercive. By that I mean speech which deprives us of our choices because it is threatening, or directly incites violence or manipulates people in highly vulnerable categories, especially children.

The suppression or punishment of speech is a coercive act. The traditional principle of English law has been that it is only justified if the alternative is coercion of a different kind, the coercion which follows from a breakdown of public order. In his Commentaries on the Laws of England, the earliest systematic survey of the common law, published in the 1760s, Sir William Blackstone described a free press as “essential to the nature of a free state”. This was why the common law did not allow pre-publication censorship. Blackstone added that after publication, an author could be punished or mulcted in damages if his words were contrary to public order. This was because the state’s ability to maintain public order was also essential to the nature of a free state and to its role as the guardian of liberty. In Blackstone’s day, the courts were much more willing to take action against provocative speech, because the state had fewer means of maintaining public order. There was no general police force in Britain until the 1830s. In the first half of the 19th century, as Britain became a more orderly, better policed society, speech became less dangerous and the courts became more active in protecting it. However, direct incitement to violence is still the main exception sanctioned by English law to the principle of freedom of expression.

The most interesting analyses of this question come from the American case law. The Americans have a related doctrine known as the “fighting words” exception. Their case law denies that this amounts to censorship, because they distinguish between the content of the message and its consequences. Restrictions on free speech are unacceptable if they are directed to the content of the message. They have to be addressed exclusively to the circumstances in which the words were uttered and to their likely consequences. A rather similar line has been taken in some decisions of the European Court of Human Rights in Strasbourg. A good example is the case of Faber v. Hungary, decided in 1992. Faber was a Right-wing activist who had been convicted in Hungary of standing at a site in Budapest where Jews had been massacred in the Second World War and holding up a banner with a fascist symbol. His conviction was held by the Strasbourg court to violate the Human Rights Convention because the mere act of holding up the banner was incapable of disturbing public order or inciting violence. The fact that it was offensive and liable to outrage those who saw it was not enough. That was an objection to the content of Faber’s statement, not to its likely consequences. I think that that is also the approach of English law, even though English judges have been less articulate in saying so. Calling for violence in a university debating society is objectionable but not illegal, whereas calling for violence in front of a howling mob in a city center is likely to be both.

In September of last year, Lucy Connolly was sentenced to 31 months imprisonment for an inflammatory tweet. The case aroused much controversy. Her unsuccessful appeal against the sentence was supported by the Free Speech Union, and the sentence was roundly denounced in The Daily Telegraph and on GB News. This seemed to me to be strange. Her tweet said: “Mass deportation now. Set fire to all the fucking hotels full of the bastards for all I care.” If she had expressed this opinion at a meeting of the Women’s Institute, there would have been no problem. The actual circumstances were that the murder of three children in Southport had been attributed (in fact wrongly) to an asylum seeker, at a time when there was rising tension against immigrants in much of the Midlands. If a rabble-rouser stood on a soap-box in front of an angry mob and urged them to head for the nearest immigration hostel and burn it down, the vice of words like these would have been obvious. Doing it on social media is worse because the reach of the internet is so much greater. Its algorithms thrust words like Connolly’s under the noses of people who are already likely to agree. The internet is not a private conversation. It can whip up a violent mob in minutes. Connolly soon regretted her post and deleted it three and a half hours later, but in the short time that it was up, it was viewed 310,000 times and reposted 940 times. It was one of a number of wild statements on the internet following the Southport murders which provoked serious riots across England later that day and violent attacks on mosques and immigration hostels. Connolly was not a free speech martyr. Her conviction was in line with a perfectly rational principle reflecting a centuries-old limitation on free speech.

Incitement to violence is not the only restriction that our law places on free speech. Some of the other restrictions are principled exceptions which are perfectly consistent with the underlying principle of free speech. Others are unprincipled exceptions which should have no place in a rational system of law.

Let’s start with the legal restrictions which I suggest we should accept. The Online Safety Act, which was passed in 2023, is a complex and in some ways clumsily drafted piece of legislation, which hits many targets that were not intended. However, its principal purpose is to require providers of online content to have age verification checks so as to control access by children to manipulative or corrupting material such as pornography or material advocating suicide or self-harm. I have pointed out that the concept of freedom of expression involves treating adults as capable of discernment and of intelligently ordering their lives. The same assumptions cannot apply to children, who are by definition in a state of developing but incomplete maturity, and who do not have the full range of liberties available to adults. It is fair to regard the manipulation of children’s sensibilities as closely analogous to the coercion from which we seek to protect adults.

Next, there are laws which regulate the way in which our right of free speech is exercised, rather than the content of our speech. These too are, in principle, acceptable. For example, you cannot picket abortion clinics within 150 meters of their doors or buy advertising time for political causes on television. The law restricts the picketing of abortion clinics because it harasses women at an emotionally difficult time of their lives, but you can still protest against abortion in ways that do not amount to harassment: you can stand with your placards 160 meters away, demonstrate in Trafalgar Square, write articles in the press, publish podcasts on YouTube, distribute leaflets, and so on. The law forbids political advertising on television in order to provide some limit on the power of big money to buy influence over political debate in a democracy. However, you can still do party political broadcasts, put up posters on billboards, take advertising space in newspapers, or hold political rallies. These restrictions are not directed against the content of any spoken or written statement. There is no infringement of the right to freedom of expression simply because there are some places where you cannot speak. A notice saying “Silence” in a courtroom, church or library is not an infringement of free speech.

My final example is more controversial. It is about the laws relating to defamation and privacy. The law of libel and slander allows people to claim damages from those who have told lies about them. The law of privacy allows those whose privacy has been invaded without any justification in the public interest to claim damages from the intruder. There are legitimate concerns about the granting of injunctions, especially in privacy cases, but I see no reason why the law should not, after proper investigation, correct the financial consequences of lies and intrusions into our private lives by awarding compensation for the damage which these things can cause. I do not regard that as censorship. It simply adjusts as between private individuals the financial consequences of free speech.

If these are legitimate restrictions on freedom of speech, there are others which have been imposed over the last half-century and which ought to cause us real concern. They are designed to protect people not from coercion or manipulation but simply from being offended.

A good example, which is currently topical, is the provision of the Terrorism Act 2000, which forbids public expressions of support for proscribed terrorist organizations. The Act forbids any conduct which involves assisting or participating in the work of a proscribed terrorist organization, which is fair enough. However, it then goes on to forbid expressions of support even where they do not assist the work of the organization but merely amount to the peaceful expression of a favorable opinion about it. Unlike the restrictions which I have just been discussing, this is a direct attack on the content of certain statements. There are people who think that terrorism in the service of a good cause is fine. I think that this is foolish, but should not be criminal. Arresting hundreds of people in Parliament Square for holding up placards saying that they support Palestine Action is an unprincipled interference with freedom of expression.

The main restrictions of this kind come under the broad heading of hate crimes, which is probably the most intrusive and least justifiable interference with freedom of expression in Britain. Hate crimes are based on the Public Order Act of 1986. Their scope has been expanded by successive amendments since the original Act was passed. The position can be summarized as follows: it is an offense to use language which stirs up racial hatred, religious hatred or hatred based on sexual orientation. In the case of religious hatred and hatred based on sexual orientation, the language used has to be threatening. This probably means words which threaten imminent violence, which broadly accords with the law’s traditional approach to breaches of the peace. However, in the case of stirring up racial hatred, the offense is much wider. The language used need not be threatening. It is enough that it should be abusive or offensive. In the case of religious hatred, there is a broadly drawn exception which allows people to discuss or criticize, or express antipathy, dislike, ridicule, insult or abuse of any religious belief or practice. This may be described as a proviso for the benefit of Richard Dawkins, the late Christopher Hitchens, and the late Bertrand Russell, all of whom have had harsh things to say about religious beliefs. Yet there is no such exception in the case of race or sexual orientation. In 2024, 140,561 hate crimes were recorded by the 43 police forces in England and Wales. The breadth of the offense can be seen from the facts that the police have arrested Christian preachers for their street sermons, parents engaged in a dispute with their daughter’s school, and the Irish comedian Graham Linehan for tweeting that women confronted by trans men in women-only spaces should “make a scene, call the cops and if all else fails, punch him in the balls”.

In addition to hate crimes properly so-called, there is an anomalous category known as “non-crime hate incidents”. The police have a common law power to record incidents. Non-crime hate incidents are not criminal offenses, but are simply incidents which the police are administratively directed to record. They have real consequences and will usually entail intimidating visits and warnings from the police. They will show up on enhanced criminal record checks and may stop people getting jobs. The police defend the practice on the ground that such incidents may lead to real crimes if repeated. This seems rather far-fetched. The job of the police is to enforce the criminal law, not to regulate behavior which is perfectly lawful but may be loutish, offensive, politically incorrect or contrary to received opinion. Different police forces have recorded as non-crime hate incidents gender-critical tweets, tweets critical of the police, accidental damage done by schoolchildren to a copy of the Koran, and even speeches by ministers proposing restrictions on immigration. Harry Miller, who was visited by Humberside police at his workplace and questioned about his gender-critical tweets, took them to court. The conduct of the police, said the judge, offended against a “cardinal democratic principle”. “In this country,” he said, “we have never had a Cheka, a Gestapo or a Stasi. We have never lived in an Orwellian society.” Yet last year, more than 13,000 non-crime hate incidents were recorded by police forces in England and Wales.

Suggested reading

Corporate America's dirty trick

By Lee Fang

“Hatred”, “threatening” and “abusive” are vague terms which cry out for precise definition, but there is no definition of these words in the Public Order Act. “Offensive” is an even vaguer term, which is likewise undefined. On the face of it, the word extends to anything that causes offense. As for non-crime hate incidents, they have no statutory basis at all and therefore no definition either. The looseness of these concepts means that in reality action depends on the discretion of individual police officers. This is not a proper function for a statutory public service or a satisfactory basis for penalizing speech. A police state can fairly be defined as a state in which the police have discretion as to what constitutes an offense. The prosecuting authorities have tried to resolve this problem in a highly unsatisfactory way. The Crown Prosecution Service and the police have agreed to define a hate crime as anything which is perceived by the victim or anyone else to be motivated by hostility or prejudice. In other words, the definition which they use is subjective. If the complainant thinks it is a hate crime, then it is a hate crime. One would think that hatred implied a very high degree of animosity, but the Crown Prosecution Service has suggested that it can include ill-will, ill-feeling, spite, contempt, prejudice, unfriendliness, antagonism, resentment, and dislike. Yet all of these, humanity being what it is, are hazards which most of us must expect to live with. Remarkably, until 2021 the College of Policing, which sets standards for the police, after explaining that a hate crime depended on the perception of the victim, added the following observation:

“The victim does not have to justify or provide evidence of their belief, and police officers and staff should not directly challenge this perception. Evidence of hostility is not required for an incident or crime to be recorded as a hate crime or hate incident.”

Carried to its logical conclusion, this approach gives racial minorities a right of veto; an entitlement to silence opinions to which they object.

In 2021, the guidance was changed in response to criticism by the Court of Appeal in Miller’s case. The Court of Appeal did not object to the subjective character of the definition, but it held that the guidance was unlawful because it was far too wide. It interfered with Miller’s right of freedom of expression and was likely to have a chilling effect on the discussion of controversial issues like gender. However, the changes to the guidance were minimal. The guidance now says that at the point of receiving the complaint, no justification or evidence is required, leaving open the possibility of seeking such evidence later. The effect is that the investigation of a hate-crime is still dependent on the personal sensitivities of the complainant and the discretion of the police.

All of this originated with the Macpherson Report on the murder of Stephen Lawrence, which was published in 1999. Macpherson made two recommendations which are relevant for present purposes. First, he recommended that the police should systematically record “racist incidents”, a category which he defined as “any incident which is perceived to be racist by the victim or any other person”. That is the origin of the subjective definitions currently used by the police and the Crown Prosecution Service. Secondly, he recommended that racist incidents should include both crimes and non-crimes. That is the origin of the category of non-crime hate incidents. Experience has shown the Macpherson recommendations have proved to be extremely damaging. Speaking in the House of Lords last November, Lord Hogan-Howe, a former Metropolitan Police Commissioner, observed that the recording of hate crime was a well-intended change which had gone wrong. “The definition of a crime is objective and the definition of hate crime is subjective,” he said, “during the time this has been in place, online crime and online hate crime have grown massively, and the application of the hate crime definition has been inconsistent.” Recently, Britain’s largest police force, the Metropolitan Police, announced that it would stop dealing with non-crime hate incidents. The current commissioner, it was announced, “has been clear he doesn’t believe officers should be policing toxic culture war debates, with current laws and rules on inciting violence online leaving them in an impossible position”. So far the Met’s example has not been followed by other police forces.

The most famous test for determining what can and what cannot properly be suppressed is the “harm principle” proposed by John Stuart Mill, the great Victorian apostle of free speech, in his 1859 treatise On Liberty. Mill taught that the only purpose for which power might properly be exercised against individuals against their will was to prevent harm to others. Hate crime and non-crime hate incidents are symptoms of a modern tendency to redefine harm so that it extends to the discomfort caused by having to endure rudeness, contradiction or dissent. When interest groups object to someone’s opinion of them, they commonly call for a subjective approach to its impact. Harm is whatever the relevant target group perceives as harm. It depends on their “lived experience”, as the phrase goes. This way of looking at the harm done by free speech is particularly common when the offended group is a racial, religious or sexual minority. The argument is that words may undermine the dignity of the groups against whom they are directed. Words wound, it is said, especially when they relate to another person’s identity or status. Words are therefore viewed as a form of violence comparable to physical assault. But there are obvious problems with this approach. One is that concepts such as undermining human dignity are too uncertain and subjective to serve as part of the definition of a criminal offense. Another is that there is an important difference between violence and words. Violence is coercive. Words, even if offensive, are not coercive except in those cases where they are calculated to provoke violence.

Today, racial, religious or sexual minorities are said to need protection from hurtful words. The desire to accommodate groups who feel themselves oppressed is understandable. It assists social inclusion. Racial, religious and sexual differences are highly sensitive matters, but they are also major issues in our society, as well as in other countries of Western Europe and North America. They need to be openly discussed without fear of criminal liability. Discussion certainly cannot be suppressed without inviting serious suppressed anger which will ultimately be destabilizing and very bad for community relations. Historically, minorities have been the main victims of censorship and other forms of intellectual and social intolerance and the chief beneficiaries of freedom of speech. As I have pointed out, people once believed in the persecution of sexual minorities, the inferiority racial minorities and the exclusion of religious minorities. These notions were supported by the social consensus of the age, but died out in the face of rational discourse. In general, rational counter-argument is a better way of addressing hate speech than the attempt to suppress it through criminal law.

The suppression or punishment of speech tends to start from the best of motives. Its progress is gradual and frequently unperceived. But the ultimate result is objectionable. Recent legislation and police practice has, I fear, moved away from the distinction between coercion and liberty, which represented the only rational basis for curtailing freedom of expression. The result has been to reduce large parts of the law in this area to incoherence.

Once you start proscribing speech on the sole ground that it offends or distresses people, you open the door to the censorship of any words that people do not like. Without a coherent principle governing the permissible limits of free speech, there is effectively no free speech on some issues, except for those who confine themselves to the inoffensive and the anodyne. I think that that would be a retreat into an unpleasant world.

***

 This article was adapted from the Shrieval Lecture, given at Hereford Town Hall on 12 November 2025.

Recent criticisms of Europe by American politicians like JD Vance have reminded us that there is a profound gulf between American and European conceptions of free speech. The United States comes closer than any other country in the world to absolute free speech. The First Amendment to the US Constitution provides that “Congress shall make no law… abridging the freedom of speech”. No ifs, no buts. Since the early 20th century, this has been held to apply to state legislatures as well. The American courts were eventually persuaded to recognize exceptions for incitement to imminent violence, obscene and defamatory material, child pornography and fraud, but that is all. Even these exceptions have been very narrowly interpreted.

The European model has been very different. Freedom of expression is an important but not an absolute public interest in Europe. In principle, it may be overridden by a wide variety of other public interests if there is a sufficiently compelling case for doing so. Article 10 of the European Convention on Human Rights protects freedom of expression, but subject to a long list of potential exceptions: national security, territorial integrity or public safety, the prevention of disorder or crime, the protection of health or morals, the protection of the reputation or rights of others, the protection of confidences, or maintaining the authority and impartiality of the judiciary.

In Britain, the list of exceptions recognized by the law grows longer all the time: speech that discloses secret intelligence, speech that prejudices current legal proceedings, speech that supports proscribed terrorist groups, hate speech directed at racial, religious or sexual minorities, “grossly offensive” messages sent through the post or over electronic networks, and so on. Most of these exceptions are concerned in one way or another with public order and social harmony. In Britain we have not usually specified opinions that it is illegal to hold in public, but other countries have. In Germany, Austria and Belgium it is a criminal offense to deny, trivialize or approve of the Holocaust. Spain, France and Switzerland go even further. It is an offense in those countries to question the existence or gravity of anything which is regarded as a crime against humanity. In many countries it is illegal to display the swastika. In Latvia and Lithuania, that extends to the hammer and sickle and other symbols of Soviet communism.

There is an inevitable tendency for people to want to exclude from freedom of expression whatever ideas they personally cannot abide. Milton’s Areopagitica, published in 1644, is one of the most eloquent pleas for freedom of expression in the English language, but he excluded Catholics and atheists, whose views he regarded as beyond the pale. So did John Locke, although he at least offered reasons for their exclusion. He thought that Catholics were intolerable because they were not interested in truth, and atheists because having no fear of God, they could not be relied upon to honor any social obligations. The authors of the famous early 18th-century tracts known as Cato’s Letters argued the case for free political speech but made an exception for what they called “factious scribblers” and “turbulent babblers”.

The problem, as George Orwell pointed out, is that free speech is worthless unless it extends to things that people do not want to hear. We all have different lists of things that we do not want to hear. Suggested candidates for legally enforced censorship include a variety of allegedly extreme or unacceptable positions such as gender-criticism, climate denial, antisemitism, blasphemy against the Prophet Mohammed or the Koran, totalitarianism, eugenics, satanism and many others. Many of these demands reflect the sensitivities of particular ethnic, social or religious groups who say that their position is special. Censorship is often the child of collective grievances. People who feel that the world has been unjust to them are likely to claim that their position is special too. If some opinions are to be banned as beyond the pale, then why should my list of unacceptable beliefs be privileged over yours? The truth is that if we believe in freedom of expression, we have to tolerate things that we regard as intolerable. Otherwise the whole concept is meaningless.

Since we have to draw a line, where do we draw it? The starting point, I suggest, is to ask why we instinctively believe in freedom of expression in the first place. I think that there are three main reasons.

The first is that constraints on our ability to communicate facts or opinions invade our autonomy as moral agents and thinking creatures. When the constraints are imposed by the state, they imply a relationship between the individual and the state which is morally repellent. They reduce us to the status of mere instruments of other people’s power. Whether the suppression of free speech comes from governments or pressure groups or public opinion, it ushers in a narrow-minded, intolerant and authoritarian world in which the fear of giving offense or challenging existing shibboleths eliminates some of the most creative and original products of the human spirit.

The second reason for believing in freedom of speech is that we live in a democracy whose entire political order depends on the free exchange of ideas. It depends on our ability to freely receive information, on an acceptance of diversity of opinion and on a large measure of tolerance of intellectual and emotional differences. Our collective life depends on the resolution of issues between citizens by marshaling objectively verifiable facts. It depends on ordered debate about their implications under common rules that exclude coercion, and on a culture in which the outcome of our processes of collective decision-making is accepted even by those who disagree with it.

Suggested reading

Inside Britain’s two-tier justice system

By M.L. Friend

A third reason why we believe in freedom of expression is perhaps the most fundamental. It is the one which received its most eloquent expression from Karl Popper’s classic work, The Open Society and its Enemies. No one person or group, however respectable or expert, has exclusive access to wisdom or truth. This is not just because of the limits on the capacity of the human mind. It is also because truth and wisdom are dynamic concepts. They are informed by a stock of experience which is constantly increasing. All statements of fact or opinion are provisional. They reflect the current state of knowledge and experience. But knowledge and experience are not closed or immutable categories. They are inherently liable to change in response to fresh discoveries or insights. Knowledge advances by testing conflicting arguments, not by suppressing them. Understanding increases by exposure to uncomfortable truths. In a world of free expression, some of what people say will certainly be wrong, hurtful or even objectively harmful. But the principle which we would have to accept in order to justify censoring these statements is more damaging than the statements themselves. We cannot have truth without accommodating the possibility of error, or wisdom without accommodating the possibility of folly. In the end we have to accept the implications of human inquisitiveness, creativity and imagination. The alternative is to entrust significant parts of our intellectual world to external authorities whose capacity for objectivity, truthfulness and wisdom is no greater than our own.

Ever since the 17th century, our civilization has been based on the notion that truth is independent of human will. It may be only partly knowable, and more or less difficult to identify, but it exists somewhere out there whether we like it or not. We have built our intellectual world on the footing that we get closest to the truth by objective study, logical reasoning and open debate, by what Mill called the “collision of adverse opinions”. I am not suggesting that this is guaranteed to produce truth or the wisest possible outcomes, but experience suggests that in the long term it is more likely to do so than any system of intellectual controls.

These reasons for believing in freedom of expression suggest a number of conclusions that we should bear in mind when we come to think about exceptions.

In the first place, freedom of expression is concerned with the transmission of information or opinion. Not everything that is spoken or written deserves protection. If I write you a letter or an email consisting of nothing but foul-mouthed abuse, I may be committing an offense under the Malicious Communications Act 1988 or the Communications Act 2003. Low level communications of this kind are devoid of intellectual, cultural or political value. In a sense I may be regarded as having communicated my opinion of you, but that hardly counts as an infringement of my right of freedom of expression.

The second conclusion to be drawn from our reasons for believing in free speech is that we should not suppress the expression of facts or opinions simply because they give offense, however understandable that offense may be. If we are not free to say things that other people regard as objectionable, we have no freedom of speech at all. In a famous statement of principle, Lord Justice Sedley observed in the High Court in a case about a Christian fundamentalist street preacher 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 promote violence. Freedom only to speak inoffensively is not worth having.”

Thirdly, it cannot be a good reason for suppressing some assertion or opinion that we think it is wrong or even that everybody thinks it is wrong. Few propositions are incontrovertibly false, in the sense that they are inherently incapable of being affirmed by hitherto unknown facts or fresh analysis. As a general rule, the best answer to misleading speech is counter-speech, i.e. reasoned refutation. Of course, counter-speech does not always work. The barrage of rubbish generated by the internet, generally mediated by algorithms which serve people up with material that confirms their existing prejudices, is a serious challenge. But counter-speech is more likely to work than suppression. Once upon a time, the authorized consensus was that the sun moved round the earth, because that view was supported by Biblical authority, then regarded as a source of incontrovertible truth. Galileo was condemned for denying it. The same point can be made about statements of opinion. Once upon a time the religious and social consensus justified slavery and the imprisonment of homosexuals, because this had been regarded as obvious for centuries. Once upon a time it was taken for granted in Britain that black people were inherently inferior. These views have almost entirely died out in advanced societies in the face of rational discourse.

In 2021 the Royal Society, Britain’s premier scientific society, called for legislation to ban the dissemination by anti-vaxxers of “false information” about vaccines, by which they meant information rejected by the scientific consensus. The case for vaccination is not inherently irrefutable. But in the current state of our knowledge, the debate has been won hands down by supporters of vaccination. That has only been possible because there has been a debate. Suppressing public opposition to vaccination will simply drive it underground where it will be more difficult to refute. Holocaust denial is another example. It is an absurd position which has been discredited not by legislation but exposure to the evidence. The great American jurist Oliver Wendell Holmes, who wrote some of the greatest judgments of the US Supreme Court in support of free speech, once observed that the guiding principle was that almost all certainty is an illusion. We might always be wrong, he said. All of us can all point to public statements which are untrue, mischievous or absurd, sometimes manifestly so. But manifestness is not a very useful criterion for censoring other peoples’ opinions. Nearly every received opinion which rational examination has discredited was once regarded as manifest by those who held it.

Finally, the concept of free speech assumes that adults are capable of distinguishing what is good from what is bad in the torrent of information to which we are all exposed. In a democratic society, it is necessary to proceed on that assumption, even though we know that it is not always true. In a society which treats every person as entitled to equal consideration and respect, we cannot distinguish between adults according to their degree of intelligence or discernment or their skill in ordering their lives. In a democracy, we have to assume that adults are capable of making discerning choices in their own interest, because the alternative is an authoritarian world in which their choices are taken away from them by the state.

All of this suggests that the threshold for banning the transmission of information or opinion should be exacting. Any exception must, I suggest, be based on some consistent principle, not just on people’s subjective likes and dislikes. The exception should also be consistent with the basic rationale for wanting freedom of expression in the first place. And it must be clearly defined, because if the boundaries of the exception are unclear, we will end up with a freedom that is at least partly discretionary. A freedom which depends on the discretion of the enforcers is no freedom at all.

I suggest that the only speech which we are justified in suppressing by law is speech which is coercive. By that I mean speech which deprives us of our choices because it is threatening, or directly incites violence or manipulates people in highly vulnerable categories, especially children.

The suppression or punishment of speech is a coercive act. The traditional principle of English law has been that it is only justified if the alternative is coercion of a different kind, the coercion which follows from a breakdown of public order. In his Commentaries on the Laws of England, the earliest systematic survey of the common law, published in the 1760s, Sir William Blackstone described a free press as “essential to the nature of a free state”. This was why the common law did not allow pre-publication censorship. Blackstone added that after publication, an author could be punished or mulcted in damages if his words were contrary to public order. This was because the state’s ability to maintain public order was also essential to the nature of a free state and to its role as the guardian of liberty. In Blackstone’s day, the courts were much more willing to take action against provocative speech, because the state had fewer means of maintaining public order. There was no general police force in Britain until the 1830s. In the first half of the 19th century, as Britain became a more orderly, better policed society, speech became less dangerous and the courts became more active in protecting it. However, direct incitement to violence is still the main exception sanctioned by English law to the principle of freedom of expression.

The most interesting analyses of this question come from the American case law. The Americans have a related doctrine known as the “fighting words” exception. Their case law denies that this amounts to censorship, because they distinguish between the content of the message and its consequences. Restrictions on free speech are unacceptable if they are directed to the content of the message. They have to be addressed exclusively to the circumstances in which the words were uttered and to their likely consequences. A rather similar line has been taken in some decisions of the European Court of Human Rights in Strasbourg. A good example is the case of Faber v. Hungary, decided in 1992. Faber was a Right-wing activist who had been convicted in Hungary of standing at a site in Budapest where Jews had been massacred in the Second World War and holding up a banner with a fascist symbol. His conviction was held by the Strasbourg court to violate the Human Rights Convention because the mere act of holding up the banner was incapable of disturbing public order or inciting violence. The fact that it was offensive and liable to outrage those who saw it was not enough. That was an objection to the content of Faber’s statement, not to its likely consequences. I think that that is also the approach of English law, even though English judges have been less articulate in saying so. Calling for violence in a university debating society is objectionable but not illegal, whereas calling for violence in front of a howling mob in a city center is likely to be both.

In September of last year, Lucy Connolly was sentenced to 31 months imprisonment for an inflammatory tweet. The case aroused much controversy. Her unsuccessful appeal against the sentence was supported by the Free Speech Union, and the sentence was roundly denounced in The Daily Telegraph and on GB News. This seemed to me to be strange. Her tweet said: “Mass deportation now. Set fire to all the fucking hotels full of the bastards for all I care.” If she had expressed this opinion at a meeting of the Women’s Institute, there would have been no problem. The actual circumstances were that the murder of three children in Southport had been attributed (in fact wrongly) to an asylum seeker, at a time when there was rising tension against immigrants in much of the Midlands. If a rabble-rouser stood on a soap-box in front of an angry mob and urged them to head for the nearest immigration hostel and burn it down, the vice of words like these would have been obvious. Doing it on social media is worse because the reach of the internet is so much greater. Its algorithms thrust words like Connolly’s under the noses of people who are already likely to agree. The internet is not a private conversation. It can whip up a violent mob in minutes. Connolly soon regretted her post and deleted it three and a half hours later, but in the short time that it was up, it was viewed 310,000 times and reposted 940 times. It was one of a number of wild statements on the internet following the Southport murders which provoked serious riots across England later that day and violent attacks on mosques and immigration hostels. Connolly was not a free speech martyr. Her conviction was in line with a perfectly rational principle reflecting a centuries-old limitation on free speech.

Incitement to violence is not the only restriction that our law places on free speech. Some of the other restrictions are principled exceptions which are perfectly consistent with the underlying principle of free speech. Others are unprincipled exceptions which should have no place in a rational system of law.

Let’s start with the legal restrictions which I suggest we should accept. The Online Safety Act, which was passed in 2023, is a complex and in some ways clumsily drafted piece of legislation, which hits many targets that were not intended. However, its principal purpose is to require providers of online content to have age verification checks so as to control access by children to manipulative or corrupting material such as pornography or material advocating suicide or self-harm. I have pointed out that the concept of freedom of expression involves treating adults as capable of discernment and of intelligently ordering their lives. The same assumptions cannot apply to children, who are by definition in a state of developing but incomplete maturity, and who do not have the full range of liberties available to adults. It is fair to regard the manipulation of children’s sensibilities as closely analogous to the coercion from which we seek to protect adults.

Next, there are laws which regulate the way in which our right of free speech is exercised, rather than the content of our speech. These too are, in principle, acceptable. For example, you cannot picket abortion clinics within 150 meters of their doors or buy advertising time for political causes on television. The law restricts the picketing of abortion clinics because it harasses women at an emotionally difficult time of their lives, but you can still protest against abortion in ways that do not amount to harassment: you can stand with your placards 160 meters away, demonstrate in Trafalgar Square, write articles in the press, publish podcasts on YouTube, distribute leaflets, and so on. The law forbids political advertising on television in order to provide some limit on the power of big money to buy influence over political debate in a democracy. However, you can still do party political broadcasts, put up posters on billboards, take advertising space in newspapers, or hold political rallies. These restrictions are not directed against the content of any spoken or written statement. There is no infringement of the right to freedom of expression simply because there are some places where you cannot speak. A notice saying “Silence” in a courtroom, church or library is not an infringement of free speech.

My final example is more controversial. It is about the laws relating to defamation and privacy. The law of libel and slander allows people to claim damages from those who have told lies about them. The law of privacy allows those whose privacy has been invaded without any justification in the public interest to claim damages from the intruder. There are legitimate concerns about the granting of injunctions, especially in privacy cases, but I see no reason why the law should not, after proper investigation, correct the financial consequences of lies and intrusions into our private lives by awarding compensation for the damage which these things can cause. I do not regard that as censorship. It simply adjusts as between private individuals the financial consequences of free speech.

If these are legitimate restrictions on freedom of speech, there are others which have been imposed over the last half-century and which ought to cause us real concern. They are designed to protect people not from coercion or manipulation but simply from being offended.

A good example, which is currently topical, is the provision of the Terrorism Act 2000, which forbids public expressions of support for proscribed terrorist organizations. The Act forbids any conduct which involves assisting or participating in the work of a proscribed terrorist organization, which is fair enough. However, it then goes on to forbid expressions of support even where they do not assist the work of the organization but merely amount to the peaceful expression of a favorable opinion about it. Unlike the restrictions which I have just been discussing, this is a direct attack on the content of certain statements. There are people who think that terrorism in the service of a good cause is fine. I think that this is foolish, but should not be criminal. Arresting hundreds of people in Parliament Square for holding up placards saying that they support Palestine Action is an unprincipled interference with freedom of expression.

The main restrictions of this kind come under the broad heading of hate crimes, which is probably the most intrusive and least justifiable interference with freedom of expression in Britain. Hate crimes are based on the Public Order Act of 1986. Their scope has been expanded by successive amendments since the original Act was passed. The position can be summarized as follows: it is an offense to use language which stirs up racial hatred, religious hatred or hatred based on sexual orientation. In the case of religious hatred and hatred based on sexual orientation, the language used has to be threatening. This probably means words which threaten imminent violence, which broadly accords with the law’s traditional approach to breaches of the peace. However, in the case of stirring up racial hatred, the offense is much wider. The language used need not be threatening. It is enough that it should be abusive or offensive. In the case of religious hatred, there is a broadly drawn exception which allows people to discuss or criticize, or express antipathy, dislike, ridicule, insult or abuse of any religious belief or practice. This may be described as a proviso for the benefit of Richard Dawkins, the late Christopher Hitchens, and the late Bertrand Russell, all of whom have had harsh things to say about religious beliefs. Yet there is no such exception in the case of race or sexual orientation. In 2024, 140,561 hate crimes were recorded by the 43 police forces in England and Wales. The breadth of the offense can be seen from the facts that the police have arrested Christian preachers for their street sermons, parents engaged in a dispute with their daughter’s school, and the Irish comedian Graham Linehan for tweeting that women confronted by trans men in women-only spaces should “make a scene, call the cops and if all else fails, punch him in the balls”.

In addition to hate crimes properly so-called, there is an anomalous category known as “non-crime hate incidents”. The police have a common law power to record incidents. Non-crime hate incidents are not criminal offenses, but are simply incidents which the police are administratively directed to record. They have real consequences and will usually entail intimidating visits and warnings from the police. They will show up on enhanced criminal record checks and may stop people getting jobs. The police defend the practice on the ground that such incidents may lead to real crimes if repeated. This seems rather far-fetched. The job of the police is to enforce the criminal law, not to regulate behavior which is perfectly lawful but may be loutish, offensive, politically incorrect or contrary to received opinion. Different police forces have recorded as non-crime hate incidents gender-critical tweets, tweets critical of the police, accidental damage done by schoolchildren to a copy of the Koran, and even speeches by ministers proposing restrictions on immigration. Harry Miller, who was visited by Humberside police at his workplace and questioned about his gender-critical tweets, took them to court. The conduct of the police, said the judge, offended against a “cardinal democratic principle”. “In this country,” he said, “we have never had a Cheka, a Gestapo or a Stasi. We have never lived in an Orwellian society.” Yet last year, more than 13,000 non-crime hate incidents were recorded by police forces in England and Wales.

Suggested reading

Corporate America's dirty trick

By Lee Fang

“Hatred”, “threatening” and “abusive” are vague terms which cry out for precise definition, but there is no definition of these words in the Public Order Act. “Offensive” is an even vaguer term, which is likewise undefined. On the face of it, the word extends to anything that causes offense. As for non-crime hate incidents, they have no statutory basis at all and therefore no definition either. The looseness of these concepts means that in reality action depends on the discretion of individual police officers. This is not a proper function for a statutory public service or a satisfactory basis for penalizing speech. A police state can fairly be defined as a state in which the police have discretion as to what constitutes an offense. The prosecuting authorities have tried to resolve this problem in a highly unsatisfactory way. The Crown Prosecution Service and the police have agreed to define a hate crime as anything which is perceived by the victim or anyone else to be motivated by hostility or prejudice. In other words, the definition which they use is subjective. If the complainant thinks it is a hate crime, then it is a hate crime. One would think that hatred implied a very high degree of animosity, but the Crown Prosecution Service has suggested that it can include ill-will, ill-feeling, spite, contempt, prejudice, unfriendliness, antagonism, resentment, and dislike. Yet all of these, humanity being what it is, are hazards which most of us must expect to live with. Remarkably, until 2021 the College of Policing, which sets standards for the police, after explaining that a hate crime depended on the perception of the victim, added the following observation:

“The victim does not have to justify or provide evidence of their belief, and police officers and staff should not directly challenge this perception. Evidence of hostility is not required for an incident or crime to be recorded as a hate crime or hate incident.”

Carried to its logical conclusion, this approach gives racial minorities a right of veto; an entitlement to silence opinions to which they object.

In 2021, the guidance was changed in response to criticism by the Court of Appeal in Miller’s case. The Court of Appeal did not object to the subjective character of the definition, but it held that the guidance was unlawful because it was far too wide. It interfered with Miller’s right of freedom of expression and was likely to have a chilling effect on the discussion of controversial issues like gender. However, the changes to the guidance were minimal. The guidance now says that at the point of receiving the complaint, no justification or evidence is required, leaving open the possibility of seeking such evidence later. The effect is that the investigation of a hate-crime is still dependent on the personal sensitivities of the complainant and the discretion of the police.

All of this originated with the Macpherson Report on the murder of Stephen Lawrence, which was published in 1999. Macpherson made two recommendations which are relevant for present purposes. First, he recommended that the police should systematically record “racist incidents”, a category which he defined as “any incident which is perceived to be racist by the victim or any other person”. That is the origin of the subjective definitions currently used by the police and the Crown Prosecution Service. Secondly, he recommended that racist incidents should include both crimes and non-crimes. That is the origin of the category of non-crime hate incidents. Experience has shown the Macpherson recommendations have proved to be extremely damaging. Speaking in the House of Lords last November, Lord Hogan-Howe, a former Metropolitan Police Commissioner, observed that the recording of hate crime was a well-intended change which had gone wrong. “The definition of a crime is objective and the definition of hate crime is subjective,” he said, “during the time this has been in place, online crime and online hate crime have grown massively, and the application of the hate crime definition has been inconsistent.” Recently, Britain’s largest police force, the Metropolitan Police, announced that it would stop dealing with non-crime hate incidents. The current commissioner, it was announced, “has been clear he doesn’t believe officers should be policing toxic culture war debates, with current laws and rules on inciting violence online leaving them in an impossible position”. So far the Met’s example has not been followed by other police forces.

The most famous test for determining what can and what cannot properly be suppressed is the “harm principle” proposed by John Stuart Mill, the great Victorian apostle of free speech, in his 1859 treatise On Liberty. Mill taught that the only purpose for which power might properly be exercised against individuals against their will was to prevent harm to others. Hate crime and non-crime hate incidents are symptoms of a modern tendency to redefine harm so that it extends to the discomfort caused by having to endure rudeness, contradiction or dissent. When interest groups object to someone’s opinion of them, they commonly call for a subjective approach to its impact. Harm is whatever the relevant target group perceives as harm. It depends on their “lived experience”, as the phrase goes. This way of looking at the harm done by free speech is particularly common when the offended group is a racial, religious or sexual minority. The argument is that words may undermine the dignity of the groups against whom they are directed. Words wound, it is said, especially when they relate to another person’s identity or status. Words are therefore viewed as a form of violence comparable to physical assault. But there are obvious problems with this approach. One is that concepts such as undermining human dignity are too uncertain and subjective to serve as part of the definition of a criminal offense. Another is that there is an important difference between violence and words. Violence is coercive. Words, even if offensive, are not coercive except in those cases where they are calculated to provoke violence.

Today, racial, religious or sexual minorities are said to need protection from hurtful words. The desire to accommodate groups who feel themselves oppressed is understandable. It assists social inclusion. Racial, religious and sexual differences are highly sensitive matters, but they are also major issues in our society, as well as in other countries of Western Europe and North America. They need to be openly discussed without fear of criminal liability. Discussion certainly cannot be suppressed without inviting serious suppressed anger which will ultimately be destabilizing and very bad for community relations. Historically, minorities have been the main victims of censorship and other forms of intellectual and social intolerance and the chief beneficiaries of freedom of speech. As I have pointed out, people once believed in the persecution of sexual minorities, the inferiority racial minorities and the exclusion of religious minorities. These notions were supported by the social consensus of the age, but died out in the face of rational discourse. In general, rational counter-argument is a better way of addressing hate speech than the attempt to suppress it through criminal law.

The suppression or punishment of speech tends to start from the best of motives. Its progress is gradual and frequently unperceived. But the ultimate result is objectionable. Recent legislation and police practice has, I fear, moved away from the distinction between coercion and liberty, which represented the only rational basis for curtailing freedom of expression. The result has been to reduce large parts of the law in this area to incoherence.

Once you start proscribing speech on the sole ground that it offends or distresses people, you open the door to the censorship of any words that people do not like. Without a coherent principle governing the permissible limits of free speech, there is effectively no free speech on some issues, except for those who confine themselves to the inoffensive and the anodyne. I think that that would be a retreat into an unpleasant world.

***

This essay is adapted from the Shrieval Lecture, given at Hereford Town Hall on 12 November 2025

We yell things like “We have rights!” when we feel that somebody might be trying to take one of them away from us. But where does it say we have them in the first place? It’s not in the Bible, or any other religious source. Yes, yes, I know there is a Bill of Rights in America, and any country worth its salt can point towards a number of things that their own countrymen (and even their women, on occasion) should enjoy as members of that particular society. But is that an absolute truth — or is it just hot air?

I’m starting to think it’s as much the latter as it is the former: just ask the Taliban… or the Mennonites… or…

For without the right to die — to decide how, when and where — what else do we have? In the United States, we have the right to make a will, to be of “sound mind and body”, and to say how we want to divide up our earthly possessions. But in very few states is that sovereignty afforded us upon the hour of our death. Even if we’re awake and aware and able to coherently make that decision on our own, the law and the state and even modern medicine itself steps in and tells us, “Not so fast…”. And what we want, as the patient, is often swept aside in the drive to keep us alive for as long as possible, even if that means in pain and despair and even if it’s the absolute opposite of our own wishes and desires.

And that sucks.

Suggested reading

There's no dignity in assisted dying

By Kathleen Stock

I wrote a play, How to Fight Loneliness, which is on at the Park Theatre in London, in which one of the characters, a roughneck, blue-collar American named Tate, is asked by a young couple for his help in ending the life of the wife, a cancer patient named Jodie. Tate has been tapped for this duty because it is whispered that he helped an ailing step-brother pass away when he was younger. And, in the present, Tate is open to doing the same for Jodie. Is this because he believes so fervently in “assisted suicide”?

Hardly.

Is it because he’s a man of great religious conviction and he’s sure this is what God would want if only He would make His will evident to his followers?

Absolutely not.

Tate is willing to support Jodie because they went to school together and she was nice to him and, most of all, because she’s asking him to do it since no one else in her life is willing or able to, including her husband and herself.

So Tate does it because he believes he’s helping her, not hurting her, and because he believes it is Jodie’s right to make that decision and no one else’s. Simple as that.

When my mother passed away not long ago, she begged me for my help in her final days, but I was only able to read to her and whisper kindnesses and hope for the best. In the end I signed the papers for her to go to a hospice and that was that. She passed away in due time and in relatively little pain by that point, but long after the time when she wanted all the agony and suffering to go away but no one listened to her: not the doctors nor the specialists nor the many nurses who bustled in and out of her room, asking her to “be quiet and stop shouting”.

Suggested reading

What the death bill tells us about life

By Graeme Archer

And there I stood, hoping and wishing and praying but not really doing anything at all. Did I make a fuss? Yes, some… but not enough.

So this woman who never wished ill on anybody in her lifetime was denied the right to die, to be relieved of her misery and to leave this Earth when she decided it was time — Socratic oaths be damned — which is what leads me to ask how many rights we really do have and what those rights might be.

Don’t hurt anyone. Don’t hate anybody. Talk if you want to and vote for whomever you care to. At least until somebody tells you otherwise or your government decides you really don’t or shouldn’t have those rights and then they’re changed. Is that Orwellian of me? Not really — and I had to name-drop somebody because this is an essay, after all, and it’s supposed to sound important and researched and all the shit that people do when they write these things.

But not this one, not this time. This is simply a question, straight from the gut: if we Earth-dwellers honestly have rights, why wouldn’t “the right to die” stand chief among them?

According to my country, I do have “the right to peaceful assembly” but that doesn’t seem to include putting up a tent or overstaying my welcome. And depending on how a university or the police or even the White House is feeling, that right can be revoked in a dizzyingly short period of time, along with my ass being kicked off campus and maybe even suspended.

Depending on where you live, you can believe that you have a right to love whom you want and to vote for whom you want — but one quick governmental overthrow and some junta will laugh in your face when you screech “free love!” and “free speech!” and all the rest of the 30 “rights” you have on that UDHR list (as they’re applying battery clamps to your nipples or nether regions). Thanks for the effort, United Nations, but a lot of the time that list of yours doesn’t actually mean a damn thing. And it’s great when human rights organizations like Amnesty International jump in when they can to help out — but sometimes there’s absolutely nothing they can do, and you know it and I know it and they know it. Sometimes the baddies win and there’s no justice for it and that’s just the way it is.

I know, I know, I’m preaching to the choir and these all seem like such obvious points, but obvious is what we sometimes overlook or forget about… until it’s too late. I don’t think we have 30 rights. I don’t think we have half that many and, on a bad day (or a bad presidency), we probably have far fewer. They slip away, they get brushed aside and, sometimes, they even get violently taken away. We have rights until somebody bigger or stronger or meaner says we don’t and then we hopefully rise up, we march or talk or scream or fight and sometimes we even win…

And sometimes we don’t.

And another “right” is gone… and another… and another… and pretty soon we’re right back where we were, 50 years ago, or maybe even 100 years ago. Don’t believe me? Just ask any woman you know.

The truth is you’re born and you die (and sometimes you don’t even live long enough to pay taxes), and everything in-between is simply happenstance and luck, good or bad, and to put the rest of it in the hands of “beliefs” and “rights” and what you are “owed” for simply being here with the rest of us is just plain foolishness.

Appreciate what you’ve got, hope you have it again tomorrow, and wish all of it on everybody else. You can’t do much more than, but if you do at least those few little things, then you are so much further ahead of most people in the crowd that it’s not even funny.

And when you really think about this, any of it, for even a second, it’s not really very funny at all.

Of course, you have the right to your own opinion about any of this…

Or do you?

“ZeroCovid” is, after all, a term that elicits confusion and, sometimes, outright hostility. Perhaps that’s why, when leading members of the global ZeroCovid movement met for a three-day international conference last Wednesday, it had a far more innocuous title: the “Covid Community Action Summit”.

But even though this increasingly popular school of thought — which holds that we must not return to normal until the virus is completely eliminated within a country — wasn’t explicitly on the billing, its presence was made clear from the outset. In her introductory remarks, the moderator confirmed to the more than 600 registrants and speakers from across the world that “we are here to end Covid through ZeroCovid and CovidZero policies”. More often at the event, held over Zoom and organised by American scientist Yaneer Bar-Yam, speakers preferred to refer to ZeroCovid as an “elimination strategy”.

Yet the purpose of the event was clear: to share evidence and political advice to help campaigners lobby Western governments to abandon any notion of living alongside the virus, and instead to follow the lead of Asia-Pacific nations in aiming to eliminate the disease entirely within their borders. This group is crucially distinct from people who support ongoing lockdown measures to suppress the virus to a level where it is safe to reopen — for ZeroCovid believers, we cannot rest until that level is zero.

On paper, this approach may sound rather sensible. After all, surely we’d all rather live in a world without Covid? Yet having attended last week’s conference, I keep returning to a question that didn’t seem to particularly trouble the speakers: at what cost?

More from this author

We need Scepticism more than ever

By Freddie Sayers

In the coming weeks, I suspect this is a question we will all be forced to answer. For make no mistake: this is no fringe movement. Their advocates are among the most regular faces in broadcast media; Professor Devi Sridhar, one of its most outspoken advocates, has appeared on Channel 4 News 21 times during the pandemic — more than any other expert.

There’s a UK ZeroCovid chapter, which last month hosted its own well-attended online conference; the Scottish government is committed to their campaign, alongside Independent SAGE, British trade unions and Labour MPs such as Jeremy Corbyn and Diane Abbott. Meanwhile, influential Tory MPs like Jeremy Hunt advocate a strategy of “zero infections and elimination of the disease” and routinely refer to the Asian model. Google search results in the UK and US for “ZeroCovid” are at an all-time high. The campaign has momentum.

Judging by last week’s conference, it is easy to see why. The campaigners are, for the most part, an impressive, sincere and eloquent group. Many of them are young, telegenic and skilled communicators. But there was a mood — a unanimity of world view — that was unsettling; a fusion of overt progressive-Left politics with an ironclad certainty about their interpretation of the science. They referred to people who disagreed with them as needing to be “educated”: “deniers”, “right-wingers”, “conspiracy theorists”, or, perhaps lowest of all, “herd immunity apologists”.

More striking, however, was the time given over to messaging, organising and communications. Alongside talks on subjects ranging from how to reduce transmission in universities to which technologies could purify the air in gyms, some sessions felt more like the War Room of a political campaign.

Suggested reading

Can the vaccine keep up with the Covid variants?

By Tom Chivers

“You have to be playing with the theory of mind of your audience,” said Tomás Ryan, a neuroscientist at Trinity College Dublin and co-founder of Ireland’s “ISAG” group that hopes to persuade the Irish government to adopt a ZeroCovid policy, in a session on strategy. “We’ve made a lot of tactical errors. We didn’t give enough time to selling the reward, which is so we can be like New Zealand.” His ISAG colleague Aoife McLysaght, a professor in molecular evolution, agreed: “We were the merchants of doom for a while… we’ve had to change our messaging. So we are now saying: you can have hugs again. You can go to music festivals. That kind of thing.”

The speakers all agreed on one central tactic: to be successful, ZeroCoviders must present themselves, counter-intuitively, as being anti-lockdown; it is only in order to quit today’s cycle of restrictions, they say, that we need to eliminate the virus completely.

But how to convince Western governments? One place to start could be two articles written by Tomas Pueyo, a tech and marketing executive who explained during a session on  “Communications Strategy & Policy” that his blogposts had garnered over 60 million views. They had even, he claimed, contributed to a global policy change on Covid. For him, it didn’t matter that he wasn’t an epidemiologist. His skill was in a different kind of viral transmission: online content. “I have experience in virals, in communication, in products that people like and share,” he said. “So I designed them for that purpose. I put a narrative structure, and some tips to make them spread virally, and it succeeded.”  The outcome of the pandemic, he explained, hinges on winning the comms game: “The epidemics of the 21st century are going to be fought more in communication than they are in the lab.”

Such a pithy remark is, of course, typical of communications professionals. But why are they now such a central part of the ZeroCovid campaign? Part of the reason might be the sustained vilification aimed at their opposing fringe group — the so-called “Covid deniers”. The ZeroCoviders are determined to avoid the same fate. As Tomás Ryan says: “unfortunately we have been presented as the extreme end of the Overton window — the other end being herd immunity and denialist people.”

More importantly, I suspect it also stems from the realisation that the implications of a country committing to ZeroCovid are highly political. At the moment, the fundamental offer in Western democracies is: accept these awful restrictions now, and the vaccines will soon offer a way out. This means that once deaths and the pressure on the health service are reduced to acceptable levels, people will expect to be able to resume their previous way of life.

But for the ZeroCovider, at that point we would need to do the opposite of relax. Instead, we would need to use ongoing “interventions” and test and trace protocols to drive virus levels even lower — even at a time when there is hardly any Covid about. It would need very talented comms professionals to successfully sell this to the British public.

Suggested reading

Lockdowns don't protect the elderly

By Jennie Bristow

At that point, even if — and that’s a very big if — Britain was able to reach ZeroCovid within our borders, what then? Most people accept the need for sensible border controls to defend against new variants while the vaccine is still being rolled out. But if our goal is an entirely Covid-free land, why would we ever relax border restrictions? You’d have to wait until the disease were eradicated on a global scale, which even the most committed activists don’t think is possible for years. As Irish minister and former Taoiseach Leo Varadkar said last week, ZeroCovid is “a promise you could never fulfil”; if you cut off the country, “when do you ever unseal, because then inevitably, you let the virus back in again”?

Still, perhaps the thorniest question is what life would be like inside our theoretical ZeroCovid fortress. At last week’s conference, speakers explained that they prefer to use Australia and New Zealand as good examples because they garner a more positive response than when they mention Asian countries.

But the country that invented the approach, one which may be a better guide of a large nation pursing the strategy after a severe outbreak, is China. Despite the footage of people celebrating in Wuhan on New Year’s Eve  — provided by the CCP and obediently carried on global networks such as CBS and the BBC — the reality of life in ZeroCovid China is anything but normal. Constantly fearful at the prospect of another outbreak, the country’s already expansive surveillance state has ramped up a gear, ready to withdraw liberty at the slightest sign of Covid. Last month, all 11 million inhabitants of Shijiazhuang were thrown into lockdown after a local outbreak; one week later, a further 5 million people outside Beijing were put into lockdown on the basis of a single case. Only this week, Hong Kong launched a policy of “ambush lockdowns” where residential blocks are sealed off at a moment’s notice.

David Rennie, Beijing bureau chief of The Economist, recently gave an astonishingly candid account of current ZeroCovid life in the Chinese capital:

“China’s strategy, from the start, was to have no infections at all… Still in Beijing, where we have hardly any cases, every time you step outside your door you have to use a smartphone to scan a QR code — every shop, every taxi, every bus, every metro station. You have no privacy at all — it’s all built around this electronic system of contact tracing. To leave Beijing you have to have a Covid test, to come back in you have to have a Covid test…. We basically don’t have the virus here, but the flip side is that they are keeping this place locked down as tight as a drum… It’s very hard to know where Covid containment starts and a Communist police state with an obsession with control kicks in.”

Surely that is the most powerful objection to the approach: that in reality it would require a long-term illiberal regime to achieve and maintain it. ZeroCovid is a totalitarian aim, best delivered by a totalitarian state. Even in Australia, last weekend there was panic buying in Perth as the city re-entered lockdown in response to a single positive test result. So far at least, British voters have not chosen to reject liberal democracy, no matter what the epidemiological allure of a ZeroCovid regime.

For now, the British Government has resisted the campaign’s logic, and the Prime Minister continues to make encouraging signals about easing restrictions and even summer holidays. But as the impact of the vaccine is felt and the number of cases continues to fall, the politically difficult question of what constitutes an acceptable level of infection will have to be addressed.

Whatever that level is, expect well-spoken ZeroCovid campaigners to say it is too high. At each hesitant step towards opening up society, expect it to be called irresponsible and short-termist. No doubt ZeroCoviders sincerely believe their campaign for a Covid-free world is a noble one. But how successful they are at influencing policy will affect the shape of our society for years to come.

The Lib Dems have since disappeared into irrelevance. The lib Cons are still around — and include in their number our current Prime Minister. But could the man who buried David Cameron’s premiership be the one to resurrect ID cards?

There’s no point rehashing the arguments made in the early 2000s. This is an issue in which the case for and against is prone to exaggeration. The fact is that a liberal democracy can have compulsory ID cards without sliding into dictatorship — or not have them and avoid anarchy. There are plenty of examples of both scenarios.

So, instead of having the old argument, let’s have a new one. I’ve always been instinctively opposed to ID cards, but looking at the underlying dynamics of the issue, 2020 looks very different from 2010. The most obvious — and recent — of game-changing developments is the Covid crisis. The future is pathogenic now. Even if there’s no second wave, we cannot assume a new normal in which pandemic disease risk isn’t a major policy consideration.

Suggested reading

Not so rosy: the legacy of the Coalition government

By Peter Franklin

Lockdown has restricted our freedoms in ways that make the concerns of the ID card debate look trivial. We’re now presented with plausible — indeed, likely — scenarios in which our freedom of movement and association depends on us being able to produce documented facts about ourselves: quarantine certificates; vaccination records; immunity status; test, trace and tracking data.

In terms of international travel some of this is already in force — with passports serving as the primary means of proving ID. How long before ID cards serve a parallel function as ‘internal passports’? It’s a disturbing concept, but as an alternative to general lockdown, it may be the least worst option.

Suggested reading

Should we let the aviation industry die?

By Peter Franklin

Another game-changing development is new technology. Back in 2010 we hadn’t yet woken up to the full implications of social media. We didn’t realise just how much the tech companies knew about us. A lot of people weren’t even using smartphones yet. In 2020, we still don’t know the whole story, but we’ve come to accept a significant degree of electronic surveillance in our everyday lives. Thus the idea of being easily identified doesn’t seem so foreign anymore, not when we’re identified every time we go online to shop, socialise and express our opinions.

Of course, the information flows that would be enabled through ID cards would be controlled by the state — and the state, unlike Facebook or Google, has the power to arrest and imprison you. So, the stakes are higher. But, then again, if the authorities want to track your movements and behaviour, they already can.

The last ten years have seen rapid advances across a variety of surveillance technologies, including facial recognition, gait recognition, and even heartbeat recognition. The CCTV network is ever expanding in scope and sophistication and is supplemented by drones and other mobile forms of image capture. And that’s just one form of big data collected by the various arms of the state.

In short, if you’re leading any sort of normal life, the authorities don’t need you to carry around an ID card in order to identify you.

Suggested reading

Big Bezos is watching you

By Peter Franklin

Indeed, the bigger risk may lie in not being identified. This was demonstrated by the Windrush Scandal — in which people who had immigrated to this country from the Caribbean were required to prove to the authorities that they’d been legally resident in the UK. For the Windrush generation who arrived before legal immigration was documented in the way it is now, this was difficult to do. Shamefully, longstanding UK residents were deported because they were unable to assemble the documentary evidence that wasn’t there — or, at least, wasn’t accessible. The scandal was in fact two scandals: first and foremost the inhumanity of plunging innocent people into a Kafkaesque nightmare; but also that it should be so hard to prove the basic facts of your own life.

A future full of big data will be a hyper-bureaucratic future. In our dealings with both the state and private corporations it is, therefore, vital that we have some consistent hold over the way our identities are recorded and recognised. ID cards could provide that control — a key that grants us access to what we need to know and what we want others to know about us. There’s no guarantee that the introduction of such a system would guarantee our rights, but it would be an opportunity to fight for them.

*

We also need to think about the people who slip through the cracks of the system — living undocumented, unofficial lives unsupported and unprotected by the authorities. Only a small minority of these will be citizens who freely choose an ‘off-grid’ existence — a choice that should be respected. The great majority, though, will be vulnerable adults and children, many of them illegal immigrants, some of them trafficked against their will — and all of them at risk of exploitation. A recent report from the Centre for Social Justice estimates that there may be 100,000 or more victims of modern slavery in Britain alone.

We make things easy for this vile trade by tolerating, even celebrating, a bargain basement labour market — the legal and illegal components of which shade into one another. This is further enabled by a context in which it’s still possible for people to live and work without needing to prove who they are — and without their employers and landlords having to confirm those identities. A robust ID card system that made this simple and straightforward — but also inescapable — would be a severe blow to the modern slavers, people smugglers, slum landlords and other exploiters who depend on the anonymity of those they exploit.

Of course, no such system can be foolproof, nor would it be completely free of inconvenience for law abiding citizens. However, the same can be said of our passports system. Ultimately, it is a question of comparing these potential harms against the consequences of the status quo — in which we’re neither able to fully control our borders nor to prevent the everyday violation of basic human rights.

*

What about the practical arrangements? One of the reasons why there was such resistance to New Labour’s ID cards policy was the lack of confidence in the government’s ability to manage IT projects. Indeed, the decade was marked by a series of fiascos — including one occasion when two CDs containing confidential child benefit data on 25 million people got lost in the post. Yes, really. The ensuing outrage was probably what doomed the ID policy before it was even implemented.

Is there any reason why we should have any more confidence in government today? The faltering progress on the promised test-and-trace app isn’t exactly encouraging. However, there’s no reason why governments can’t deliver ambitious IT projects. Singapore and Taiwan have impressed the world with their use of tech to control the spread of Covid. As it happens, both countries also have compulsory ID card systems. Another interesting model is Estonia, a world leader in providing online government services. Estonians can even vote in national elections online. And that’s despite the country being under constant cyber-attack. 

Suggested reading

Covid is a once-in-a-lifetime opportunity

By Peter Franklin

So if they can do it, why can’t we? Well, the fact is that we can. Any ID card system could be based on systems that have been successfully and reliably computerised. The passports system is the obvious example and an appropriate one if ID cards are re-introduced as means of exempting individuals from future lockdown restrictions. 

If HMG ever persuades the British people to accept ID cards it must be in the service of our overall convenience, not theirs. Every ID card should come as part of a package that also includes a free passport and, for those suitably qualified, a free driving license too. As in Estonia, it should provide strong authentication for access to online government services and participating private sector websites. And even if we don’t immediately follow the Estonian example of internet-based voting, ID cards could be used to provide a robust system of confirming identity for voting by post and at polling stations. 

*

Then there’s the question of electronic cash. Thanks to innovations like contactless payments and online retail we’re using less physical cash than ever. Covid has only accelerated the decline of notes and coins, whose circulation through so many hands makes them a potential vector of disease. But to do without them altogether would require some form of e-cash — that is, not a credit or debit card, but a card that carries a digital currency for making instant, frictionless payments. In other words, something like an Oyster Card, but which could be used anywhere.

Suggested reading

Why is Zuckerberg inventing a currency?

By Peter Franklin

The big tech companies smell an opportunity. Some of them, like Facebook with its Libra project, are already on the case. But do we really want private companies running the digital currencies of the future? Or do we want one of the very oldest functions of government to stay in public hands? Writing in The Guardian, the economist Kenneth Rogoff suggests one possible way forward:

“The most radical approach would be a dominant retail central-bank currency which allows consumers to hold accounts directly at the central bank. This could have some great advantages, such as guaranteeing financial inclusion and snuffing out bank runs.”

It doesn’t have to go quite that far, but if currencies do go fully digital then making sure that everyone has access to the means of exchange becomes absolutely essential. At the very least, government needs to provide some form of guaranteed basic account. But that, of course, raises the question of identification. It probably wouldn’t be a good idea to have our electronic cash on the same piece of plastic as our ID, but the two would have to be linked in some way.

*

William Gibson once said that “the future is already here — it’s just not evenly distributed”. Well, a future in which everyone is subject to constant, electronic, centralised identification is certainly on its way.

The things that were decentralised — the little bits of information we give away through our relationships with friends, family, strangers, suppliers, customers and the agencies of the state — are inexorably coming together. But that process of centralisation is not evenly distributed. It privileges the most powerful organisations and is taking place away from public scrutiny.

ID cards might just be our best chance to drag the centralisation of our data — our very identities — into the light. It’s not a question of whether we’ll be ID’d in every moment of our lives, but how.

My ‘under-reported story of the week’ was from the world’s most under-reported country – China. It concerned the arrest of a 31-year old man at a pop concert in the city of Nanchang. The remarkable thing was that the police had managed to spot him in a crowd of 60,000 people. How did they manage that?

Fortunately, the story wasn’t entirely unreported: here’s Amy B Wang for the Washington Post:

“The man, identified only by his surname Ao, was reportedly wanted for ‘economic crimes,’ according to Kan Kan News. Details about Ao had been in a national database, and when he had arrived at the stadium, cameras at the entrances with facial-recognition technology had identified him — and flagged authorities, the news site reported.

‘He was completely shocked when we took him away,’ police officer Li Jin told Xinhua news agency. ‘He couldn’t fathom that police could so quickly capture him in a crowd of 60,000.’”

Recommended listening

Test article

By Charlie Pickles

While the ability of an automated system to pick out one face from so many (and then alert the authorities) is impressive, not to say scary, there were special circumstances – i.e. the layout of a stadium channels people through a restricted number of access points, where cameras can be strategically positioned.

Could facial recognition work nearly so well in a less regimented and much bigger space? The Chinese authorities are determined that it should:

“As The Washington Post‘s Simon Denyer reported, law enforcement and security officials in China hope to use such technology to track suspects and even predict crimes. Ultimately, officials there want to create a comprehensive, nationwide surveillance system known as ‘Xue Liang,’ or ‘Sharp Eyes’ to monitor the movements of its citizens…”

In a previous UnPacked, I wrote that no country is a stronger position to develop a comprehensive facial recognition system. China has the vast centralised bureaucracies, the disregard for civil liberties and the major investments in artificial intelligence required to make it happen. I wouldn’t be surprised if Xue Liang-type systems become a major export opportunity for China’s tech sector.

Recommended reading

Who wants to buy the products of 'Beijing Analytica'? Many of us (probably).

By Peter Franklin

Of course, ‘Sharp Eyes’ technology can only be as good as its actual eyes. Cameras capable of feeding image data for processing by recognition algorithms are getting cheaper and smaller – but they’re not yet cheap or small enough to be ubiquitous.

Give it time, though. For instance, consider the following story from Samuel K Moore for IEEE Spectrum:

“Solar cells convert light to electricity. Image sensors also convert light to electricity. If you could do them both at the same time in the same chip, you’d have the makings of a self-powered camera. Engineers at University of Michigan have recently come up with just that, an image sensor that does both things well enough to capture 15 images per second powered only by the daylight falling on it.

“With such an energy harvesting imager integrated with and powering a tiny processor and wireless transceiver you could ‘put a small camera, almost invisible, anywhere,’ says Euisik Yoon, the professor of electrical engineering and computer science at University of Michigan who led its development.”

Though only at the proof of concept stage, the direction of travel is clear – the integration of self-powered, always-on image sensing throughout every public space (and quite a few private ones too).

The internet, like a week-old kitten, is beginning to open its eyes. ln these initial stages, what it sees of the world is blurry and incomplete – but don’t be fooled, it won’t be long before little is missed.

Will this mean a big brother style society in which the state sees everything we do?

Not quite. Once the technology – its sensors and algorithms – becomes cheap, it’s hard to see the state, or any other major player, retaining exclusive control. As well as being all-pervasive, the technology will be also be open-source. Everyone will be able to watch everyone else – and, as never before, our control over private space will defined by our ability to stop others from looking into it.

Recommended reading

Kiss your privacy goodbye – CCTV is just the beginning

By Peter Franklin

If this stuff was getting the coverage it deserves, we’d be debating implications non-stop. The practice of law enforcement – indeed the entire way in which we interact with the world and with one another is set to be transformed. Concerns over the CCTV systems we have now will seem quaint by comparison.

We don’t even have a name for new technology (automatic facial recognition is only one aspect of it). I’m not sure ‘Xue Liang’ will catch-on beyond China – so let me propose an alternative and more generic label. Pervasive Open Source Surveillance Electronics – or POSSE for short.

Any association with the law enforcement practices of the Wild West is purely coincidental.

But what if the number is more of a score than an identifier – a means of quantifying the value of each person’s life? Arguably, we’re already defined by such numbers: bank balance, body mass index, IQ, etc. But what we don’t have to cope with is a single, state-sanctioned, numeric indicator of our social desirability.

The concept is deeply dystopian, but according to a jaw-dropping report by Rachel Botsman for Wired, it’s about to become fact for the people of China:

“…the government is developing the Social Credit System (SCS) to rate the trustworthiness of its 1.3 billion citizens. The Chinese government is pitching the system as a desirable way to measure and enhance ‘trust’ nationwide and to build a culture of ‘sincerity’. As the policy states, ‘It will forge a public opinion environment where keeping trust is glorious. It will strengthen sincerity in government affairs, commercial sincerity, social sincerity and the construction of judicial credibility.’”

The system is still taking shape, with the government giving “a license to eight private companies to come up with systems and algorithms for social credit scores.”

The sort of data going into these prototype systems includes fairly obvious stuff like financial credit scores and the verifiability of contact details. However, other possible components include an assessment of an individual’s reliability based on the kind of products they purchase. Then there’s social media, a potentially rich source of information – especially about relationships.

Botsman can see the implications and doesn’t like them one bit:

“Posting dissenting political opinions or links mentioning Tiananmen Square has never been wise in China, but now it could directly hurt a citizen’s rating. But here’s the real kicker: a person’s own score will also be affected by what their online friends say and do, beyond their own contact with them. If someone they are connected to online posts a negative comment, their own score will also be dragged down.”

But why would these scores matter? Based on a reading of Chinese government documents, Botsman sets out the consequences of scoring poorly:

“…people with low ratings will have slower internet speeds; restricted access to restaurants, nightclubs or golf courses; and the removal of the right to travel freely abroad with, I quote, ‘restrictive control on consumption within holiday areas or travel businesses’. Scores will influence a person’s rental applications, their ability to get insurance or a loan and even social-security benefits. Citizens with low scores will not be hired by certain employers and will be forbidden from obtaining some jobs…”

But, hang on, aren’t we westerners subject to quite a lot of the same thing? For instance, good luck getting a mortgage if you’ve got a poor credit rating. Also, aren’t we constantly rating one another on various websites? And what about the personal information we give away to social media companies – what are they doing with that?

It’s a point that Botsman concludes on:

“Today China, tomorrow a place near you… If we are not vigilant, distributed trust could become networked shame. Life will become an endless popularity contest, with us all vying for the highest rating that only a few can attain.”

This is a fair and timely warning. Yet the Chinese system represents a difference in kind not just degree. The centralised and soon-to-be compulsory nature of the SCS has no parallel in the West. And nor does the system’s most crucial feature: the fact that information about all sorts of qualitatively distinct things is rolled into a single quantity.

A criminal record, unpaid debts, buying too much alcohol, having the ‘wrong’ friends, sharing unpopular opinions: they all become one – interchangeable points towards a score that will be the first thing that anyone finds out about you and the last thing they forget.

But remember the first rule of information processing: garbage in, garbage out. For instance, feed a computer with the wrong information about your credit history, it’s unlikely to give you the right credit rating.

As Laura Hudson points out in a salutary briefing for FiveThirtyEight, loan approvals are just one of the things that computers are ‘deciding’ these days:

“Rather than relying on human judgment alone, organizations are increasingly asking algorithms to weigh in on questions that have profound social ramifications, like whether to recruit someone for a job, give them a loan, identify them as a suspect in a crime, send them to prison or grant them parole.”

If the alternative is a parole board composed of potentially racist human beings, a prisoner may prefer to take his (or her) chances with a fully computerised points-based system. But if the points are being totted up on the basis of data produced by a structurally-biased criminal justice system, then the prisoner could be worse off:

“Black people are arrested more often than whites, even when they commit crimes at the same rates. Black people are also sentenced more harshly and are more likely to searched or arrested during a traffic stop. That’s context that could be lost on an algorithm (or an engineer) taking those numbers at face value…”

A flesh-and-blood parole board might be using information derived from the same source, but there’s at least a chance that its members will exercise some wisdom in their interpretation of the data. A computer, though, once programmed, is set in its ways:

“Biased data can create feedback loops that function like a sort of algorithmic confirmation bias, where the system finds what it expects to find rather than what is objectively there.”

Compounding the issue is that people tend to believe what their silicon pals tell them. Indeed, a variation on ‘garbage in, garbage out’ is ‘garbage in, gospel out’.

For instance, if you were presented with ‘evidence’ supplied by a police force with a known history of racism, you’d  be on your guard. But what if that same information were to be computerised, combined with other data sources, processed through a complex algorithm and presented in a different place and at a later time? The original context would be lost even if its ultimate effect remained the same. One might call this ‘data laundering’.

Is there anything we can do to counter this effect? Hudson outlines one approach:

“Advocates say the first step is to start demanding that the institutions using these tools make deliberate choices about the moral decisions embedded in their systems, rather than shifting responsibility to the faux neutrality of data and technology.”

But what do “deliberate decisions” mean in this context? Does it mean rewriting algorithms to counteract real (and perceived) biases? If so, the adjustment could itself be subject to bias.

For instance, imagine if testing in schools were to be handled through a national computerised examination system. Such a facility would generate huge amounts of useful data about the way that different groups of pupils learned new facts and skills. But what if the same data were to reveal persistent differences in the cognitive processes of boys and girls (for which a growing body of evidence already exists). To some activists this might be regarded as ‘neurosexism’ – evidence only of bias in the way that the sexes are treated in the classroom and elsewhere. They might therefore demand a tweak to the analytical software in order to ‘correct’ the offending data.

And therein lies a further danger: Rather than trying to establish the existence or non-existence of original biases, we will — in a world where human judgment is increasing mediated through algorithms – seek only to change the algorithm. This would amount the laundering not just of data, but of reality itself.

Then the prosaic reality of Windows and FaceBook kicked-in, and we realised that the internet was basically just an electronic version of our filing cabinets and photo albums – useful or even addictive, but hardly a world of its own.

But that is changing fast. Piece-by-piece we are building a parallel world in truly remarkable detail. And it’s no game. For instance, you wouldn’t expect much mucking about in a German factory – but as an Economist report from a Siemens facility makes clear, virtual reality is being put to productive use:

“…there is another factory, a virtual version of the physical facility that resides within a computer system. This digital twin is identical in every respect and is used to design the control units, test them, simulate how to make them and program production machines. Once everything is humming along nicely, the digital twin hands over to the physical factory to begin making things for real.”

Using this technology, one can modify the digital twin to see how changes to its real world counterpart would work out. Assuming enough computing power, one could also fast-forward the simulation to predict what might happen to the original at a given point in the future – for instance, which machine parts might malfunction first.

Furthermore, components within each simulation could be copied and incorporated into other simulations:

“The spread of digital twins could shake up supply chains. For example, suppliers could be asked to submit a digital twin of their product so that it can be tested in a manufacturer’s virtual factory before an order is placed.”

Presumably, there must come a point at which the exchange of content merges the individual simulations into one giant simulation.

For the moment these digital twins are of objects – from simple products to complex machinery. But what about digital twins of real people?

In a feature for the Proceedings of the National Academy of Sciences, Mitchell Waldrop writes about the use of ‘agent-based models’ – computer simulations of human societies:

“These artificial societies contain thousands or even millions of subroutines, each representing individual people moving through virtual landscapes while behaving in more or less human ways. Modelers can simulate real-world geography and transportation networks; realistic populations that have the right distribution of ages, genders, and jobs; and details about who comes into contact with whom: ‘the entire social structure of households, hospitals, and workplaces’ says Ira Longini, a statistician and modeler at the University of Florida in Gainesville.”

Waldrop’s main focus is on the use of agent-based models to predict and combat the spread of highly infectious diseases like ebola. However, agent-based models are also being used closer to home – for instance, by the Bank of England:

“In November 2016, Andy Haldane, chief economist at the Bank of England, took his discipline to task for failing to anticipate major events, such as the financial collapse of 2007-2008: the economic analog of a global pandemic. A big part of the problem, he said, was that his colleagues typically modeled swings in the economy using equations that focused only on global factors such as interest rates, while ignoring… individual human differences and interactions. Instead, they should be building on decades of academic research that has produced agent-based models of issues such as innovation, inequality, and the stability of the financial system.”

The Bank has already developed two agent-based models – one for the housing market and one for the bond market.

As things stand, the sub-routines representing the individual humans within each model are fairly simple. After all, the purpose of such models is to predict broad outcomes at a population level. But what happens as programming techniques become more sophisticated – allowing the sub-routines to develop into digital twins of real people, informed by data pulled in from social media sites and public sector records?

In an agent-based model owned by a prospective employer, your digital twin would be like a walking, talking CV (or résumé) with a mind of its own (or, rather, the semblance of one). How would you like to have your virtual self tried-out in a simulated workplace? Or what about an agent-based model operated by a crime-fighting agency? What if your ‘twin’ starts misbehaving in one of those?

One would hope that that the predictive limitations of such models would be understood. But the real issue isn’t whether they should be believed, but whether they are believed.