Why ban hate speech?
About words that hurt, devalue or exclude groups of people. Should such language be banned?
People can say truly awful things. Words can hurt; sometimes it feels like being smacked in the face, minus the bruises. When it happens again and again, the damage cuts deeper: people can feel isolated, humiliated, and fearful that hateful words might turn into hateful deeds. This article explores where to draw the line. What should we think of arguments for translating our abhorrence of offensive language into a ban?
First, let’s explore what we are talking about.
What counts as ‘saying something’?
Philosophers can spend days unpacking what qualifies as “speech”. Let’s not. For simplicity’s sake, let’s call it communication that typically involves language. Spoken or written words, sure. But let’s also throw in cartoons, videos, video games. Even non-verbal stuff counts: a glare, a rude gesture maybe. But where’s the line? Well, a punch in the face does communicate something, but let’s agree that’s more about the punch than the message. Some grey areas, like legal texts or warning signs, are arguably more about function than expression. Let’s set those aside too.
What’s clearly out of bounds?
There are some things we can probably all agree you simply can’t say:
You can’t incite crime. Don’t encourage the looting of Chinese-owned shops. Don’t go shouting that some convicted pedophile deserves a beating. No fatwas on foreign authors. No “burn that mosque down” rhetoric.
You can’t knowingly spread false information that could cause harm. Don’t yell “fire!” in a packed concert hall when there isn’t one. Don’t joke about bombs on planes. Don’t claim you can cure cancer with sugar pills or arsenic.
You can’t slander people with malicious lies that ruin their reputations.
You can’t share things meant to stay secret. Doctors must keep patient info private. Found the secret formula for Coca-Cola during a break-in? Keep it to yourself. State secrets? Still secrets. You may not distribute nude photos of your ex, or child pornography.
You can’t expose children to content they’re too young for — like violent films.
And then there’s the precautionary principle: don’t confront people with shocking content without fair warning. Give people the choice to look away.
So far, so clear. But should we go further? Should hate speech also be banned?
It’s about groups
The big question is: beyond the restrictions already mentioned, should the law limit what people can say when it comes to public remarks that wound, exclude, or marginalise entire groups?
We’re talking here about group identities — things people didn’t choose and can’t change easily, but which define who they are: race, ethnicity, religion, worldview, age, sexuality, physical traits. Several of these traits have, historically, been the fault lines of bloody conflict. Not because of individuals, but because people were treated as mere extensions of a group, stripped of their individuality, boxed in, blamed or hated for something they are, not something they did.
Little wonder, then, that many argue for stricter limits on language that humiliates, excludes, or threatens such groups of people.
Heated speech
Let’s talk categories of speech — and how heated they get.
Category 0: No one bats an eye. Bland, agreeable, inoffensive. Not our concern here.
Category 1: Some people find these comments annoying, upsetting, or disagreeable. Maybe they’d rather not hear them. But they’re not illegal, and shouldn’t be.
Category 2: Now it gets hot. These are statements that provoke strong emotional reactions. People get angry, hurt, offended. Think racist slurs, religious insults, dehumanising language, harmful views that go against scientific consensus, or ideas that may be twisted by extremists. This is where the real debate begins.
Category 3: Incitement to illegal, immoral acts — especially through violence. No freedom of speech protects that.
Most of us can agree that Categories 0 and 1 are allowed, and Category 3 is forbidden. It’s Category 2 — the murky middle — that sparks the fiercest arguments.
New arguments against free speech
So, we’re talking about public statements that fall into what we earlier called Category 2 — the kind that offend, provoke, or marginalise, but don’t directly incite violence or crime. Objections to such speech aren’t new. Historically, plenty of reasons were given to clamp down on what people could say:
Don’t go telling the public lies.
Don’t tempt people into immoral behaviour or corrupt their minds.
Don’t mock the sacred.
Don’t undermine authority or disturb public order.
Don’t insult or slander others.
Some of these objections still hold water today, though they’ve been nudged aside somewhat by the growing emphasis on freedom of speech. Blasphemy, for instance, is no longer a criminal offence in many countries. In the next article, we’ll look at the main arguments for free speech.
A new era, new fears
After World War II, new arguments began to gain ground, shaped by disappointing experiences with Nazism, the wave of decolonisation, and the enduring racial inequality faced by Black Americans. The world became more sensitive to racism, and especially antisemitism. Genocide had left deep scars.
Europe responded with new laws designed to ensure that the atrocities of the Nazi era could never happen again. For obvious reasons, Germany’s laws were the strictest. But similar measures were adopted across Europe and by international bodies like the UN. Decolonisation also played a role: former colonies saw an opportunity to reset the boundaries after centuries of ethnic oppression by European powers. Racist language had to be combated, even if freedom of speech would sometimes have to take a back seat.
The term hate speech rose to prominence in the 1980s, driven by thinkers associated with critical race theory. This academic movement argued that America’s Black population continued to suffer from systemic oppression. Only radical intervention could undo centuries of injustice. That meant not just aiming for equal opportunities, but equal outcomes — and rooting out all forms of discrimination, especially the unconscious kind baked into society itself.
Meanwhile, far-right groups were re-emerging — loud, provocative, and unapologetically offensive. They wrapped themselves in the First Amendment, claiming the law gave them every right to spew hate. For critical race theorists, that was intolerable. They argued that hate speech should be treated as a special case: an exception to constitutional free speech protections.
In US academia, critical race theory gained widespread traction. In politics, it has a significant influence in the Democratic Party. But its radical edge and its specific American context meant it didn’t gain the same following in Europe. That said, the concept of hate speech did cross the Atlantic, but in a different cultural package. Here, it fused with older traditions of dignity, civility, and public decency. We’ll return to those elements later.
Foucault: language is power
Enter Michel Foucault (1926–1984), the French philosopher who had a thing or two to say about language and power. Power, he argued, doesn’t just censor, it also creates. It defines the terms, names the categories, tells us who fits where. Take the “homosexual”. Before the 19th century, same-sex behaviour was seen as something some people did, some more than others. But then doctors and lawyers arrived and invented a new category: the homosexual. Conventions emerged about the place of the homosexual in society, and over time homosexuals also came to see themselves according to conventions: as a human category.
For Foucault, language isn’t a neutral tool. It’s a force that shapes society, that assigns status, that locks people into identities. Another example: the “madman”. Once upon a time, eccentric or troubled people had their quirks, but still belonged to the social fabric. Then psychiatry arrived. Suddenly, people weren’t just odd; they were paranoid, or psychotic. Diagnoses turned them into patients, confined them, stripped them of autonomy. Language is not merely a passive description of reality, but also a technology of power that establishes norms, assigns identities and reinforces social order.
This insight inspired the critical race theorists, the crits. But Foucault was mainly describing. He spoke of the “vérité comme champ de bataille”, truth as a battlefield. He distrusted the idea that the state can neutrally determine what is true. He would say: if the state determines what is hate, then power merely shifts. Language as a technology of power continues to exist, wherever you place it.
Critical race theory goes further than Foucault: action is also needed. Power must be shifted. A proponent of critical race theory such as the American lawyer Mari Matsuda (b. 1956) wants to legally combat hate speech, because language categorises, excludes and marginalises people. There must be a fight back, and coercion is legitimate, to fight against the power of language and the language of power, on behalf of the marginalised.
Arguments for banning hate speech
So, what are the main arguments for banning hate speech?
Words can hurt like punches
Physical violence is illegal. Then why not verbal violence too? Some words can wound more deeply than a fist ever could. Verbal abuse should be punishable as much as physical abuse.It harms society
If a group is constantly under verbal attack, they may withdraw, become isolated, integrate less, fall behind socially, or even become radicalised.It can lead to violence
Hate speech isn’t just hot air. It can set the stage for violence, even ethnic cleansing or genocide.It violates human dignity
Targeted individuals are reduced to labels. Their uniqueness is erased. They’re treated as inferiors, as stereotypes.It’s morally wrong
We have a moral duty to treat everyone equally. Hate speech is discrimination, and discrimination deserves punishment.It undermines democratic values
If we believe in democratic equality, then we must challenge speech that denies that equality.
These arguments aren’t ridiculous. They’re sincere, sometimes persuasive. But they’re not immune to criticism either. So next, we’ll take them apart. After we’ve done that, we’ll be able to judge how they hold up when weighed against the principles of free speech. We’ll do that in the next article.
Words that wound
You’re not allowed to punch someone in the face — so why are you allowed to hurt someone with words that sting just as badly, if not worse? That’s the argument: the boundary between action and speech is artificial.
To the extent that speech can inflict a wound at once psychic, social, and somatic, it is not only expression, but conduct: it acts on people in injurious ways. When speech causes psychological, social, or even physical harm, it stops being “just expression” and starts functioning as behaviour — damaging behaviour.
— Judith Butler, Excitable Speech (1997/2021)
Does hate speech actually cause harm?
In 2015, social scientists looked into the psychological impact of hate speech in Australia, a country with not only a significant Aboriginal population (over 3%), but also large, visibly identifiable immigrant communities: Arab Muslims, Indians, Chinese, Vietnamese, Filipinos, Sudanese, and many others.
The researchers found measurable psychological effects. Participants reported anxiety, feelings of inferiority, victimhood, and social withdrawal. The effects were often described in physical terms, echoing Judith Butler’s words.
“It was like crushing emotionally and spiritually. And physically.”
These weren’t isolated cases. The interviews painted a bleak picture of life as an ethnic minority in what’s usually seen as a civilised, liberal democracy:
“I’m hurting, I feel that I’m living here and I’m [earning] my livelihood here, my family is here but I’m still not in their [good]will. I’m not having the ownership of – yeah. I feel that.”
“If I’m in a busy train, I wouldn’t read my Arabic newspaper, so people would not recognize me as a middle eastern.”
“When people experience hate speech ‘they do nothing. They just feel angry. When they come back home after that thing, they think about it. Again they feel very desperate, frustrating.”
In Australia, public vilification or intimidation based on race, skin colour, or ethnic origin is against the law. The Racial Discrimination Act covers public speeches, publications, and online posts.
Interestingly, most of the examples in this study — even those with demonstrable psychological damage — didn’t fall under what I’d strictly call hate speech. They were private, interpersonal interactions rather than public statements. But critical race theorists tend to group them together, and many also want personal verbal abuse to be punishable.
And there’s something to that. In fact, I’ll argue later that some forms of interpersonal hate speech should be criminalised. But for this article, we’re mostly focusing on public expression.
Where public speech was involved in the study, it was generally something more ambiguous: politicians arguing for tighter immigration controls; rejection of Islamic extremism; criticism of Israeli policy; TV shows featuring ethnic stereotypes; local protests against a proposed mosque. In other words: Category 1 content. Upsetting to some, yes. But not over the line.
The real takeaway? Ethnic minorities face real hardship, and racism between individuals is very much alive, also within robust liberal democracies. But that struggle is tangled up with broader issues: socio-economic vulnerability, fear of being judged for extremism or crime from within their community, and sensitivity to criticism of controversial aspects of Islam or Israeli policy. That sensitivity can also fuel less credible claims:
“The media hate our community. They want South Sudanese to be frustrated and feel as if they are not Australians.”
Sure, public hate speech would likely make things worse. But the study shows that without it, the psychological well-being of ethnic minorities is already precarious. Public hate speech isn’t the core of the problem.
Should hurtful words be punishable?
The question remains: should we punish hurtful words as much as we punish hurtful fists? Let’s put it to the test.
Imagine this. You and your family are Belgian, born and raised, but of clearly African descent. You’re used to subtle (and not-so-subtle) racism, sometimes directed right at you. You sometimes feel like a stranger in your own country.
You’re walking through the park with your partner and your child. Your child accidentally bumps into a passer-by with her bike. You apologise. But the person erupts in rage:
“F*** off, filthy n****rs! You don’t belong here! Go back to Africa!”
Should that be a crime?
If you’re looking for a neat legal rule to determine when behaviour becomes a criminal offence — sorry, it doesn’t exist. The blunt truth: something is illegal if the law says it is. But that’s hardly satisfying. So let’s turn to the philosopher Joel Feinberg, who offers a more thoughtful approach to when speech should be punishable.
Feinberg suggests four key tests:
Does it cause harm?
Yes. The Australian study shows that racist insults can inflict real psychological damage.Is it a serious act?
Yes. This wasn’t an accidental slip or a joke gone wrong. It was a calculated, malicious outburst.Does it serve any legitimate purpose, like contributing to public debate?
No. It was pure verbal aggression. No argument, no context.Was the victim a voluntary participant?
Definitely not. You didn’t choose to hear it, couldn’t anticipate it, and couldn’t escape it.
By Feinberg’s criteria, there’s a strong case for making this kind of personal verbal attack a criminal offence. The fact that it happened in private doesn’t change that. A punch is illegal even in private. So are private grave insults. Why not this?
Public vs private speech
This example was deliberately private — a direct insult to your face. But what if instead, the same man had climbed onto a soapbox and shouted to the whole park:
“All filthy n****rs should go back to Africa!”
Should that be illegal too?
It’s not so simple. If he were addressing you specifically, it is you who doesn’t belong here and has to go to Africa, the police should rightly get involved. But shouting about a whole group — as nasty and offensive as that is — edges into political speech. It’s no longer personal abuse; it’s a general (if vile) statement meant for public consumption.
Why tolerate public expressions of hatred in political debate? Because democracies rest on free speech, even ugly speech. Because speech, even offensive speech, can be challenged, debated, disarmed. But we’ll dig into those arguments in the next article.
In her book Words that wound, Mari Matsuda gives harrowing examples of racist abuse: an FBI agent finding a monkey photo stuck over a picture of his child; burning crosses on someone’s lawn. These are not grey areas. They deserve disciplinary action or criminal charges.
But in this article, we’re sticking to public expressions not directed at individuals.
Hate speech causes social harm
We’ve already seen evidence of real psychological harm. And when individuals suffer that kind of pain, it affects the whole group. That’s when private suffering becomes a public issue.
Group members find it difficult not only to participate in the collective life, but also to lead autonomous and fulfilling personal lives. They lead ghettoized and isolated lives with a knock-on effect on their children’s education and career choices.
Targets of hate speech understandably feel nervous in public spaces lest they should be humiliated. They are afraid to speak their minds and behave normally, and they worry constantly about how the negative stereotypes that others hold of them will lead them to interpret their words and actions. As a result, they are likely to feel alienated from the wider society, to lead shadowy lives, and to feel trapped in a cramped mode of being.
— Bhikhu Parekh, Is there a case for banning hate speech? (2012)
Minorities withdraw into their own community to feel more protected, which makes integration more difficult, with consequences for future generations. They withdraw from the social dialogue, do not feel like a full member of society, with a risk of radicalisation.
It’s a real problem. But again, public hate speech generally isn’t the root cause. Just as with personal insults, it’s part of a wider issue: racism, socio-economic precarity, and slow integration.
Does hate speech lead to social harm? That is difficult to demonstrate when other causes of harm are so obvious. Moreover: what is the general interest, Friedrich Hayek wrote. In that respect too, social harm is difficult to determine. Moreover: suppression of free speech to combat hate speech can also lead to harm. Less freedom is also social harm, stated H.L.A. Hart.
Everyone deserves respect and dignity
Hate speech also violates the dignity of the members of the target group by stigmatizing them, denying their capacity to live as responsible members of society, and ignoring their individuality and differences by reducing them to uniform specimens of the relevant racial, ethnic, or religious group.
— Bhikhu Parekh, Is there a case for banning hate speech? (2012)
Let’s start with that last point. Truth is, we reduce people to group stereotypes all the time, consciously or not. We make assumptions about car salesmen, Bulgarians, people with lisps, Jehovah’s Witnesses, drug dealers, baby boomers, people with face tattoos, golfers… the list goes on. It’s part of being human. You can’t stop it.
It would be absurd to ban negative public comments about Bulgarians, Buddhists, or boomers. So why treat negative public comments about Muslims, Sudanese, or Jews any differently?
I have previously discussed the argument that one should not violate people's dignity. The idea that dignity must never be violated sounds noble. But it can’t be a magic word used to silence any uncomfortable thought. Dignity-based arguments often struggle with a core problem: why should the dignity of the person being offended trump the dignity of the person being silenced?
The civility argument
At this point, many people fall back on civility. Yes, it’s deeply rude to make monkey noises at a black footballer. If I managed the stadium, I’d throw the culprit out. But should it be illegal?
It’s rude to insult whole communities, to cause hurt, to be a jerk. But rudeness isn’t a crime, nor should it be. If we started punishing all bad manners, where would it end?
And here’s a tougher question: what’s worse: a rude truth or a polite lie?
When Martin Luther called the Catholic clergy “soul-murdering thieves”, should he have been gagged for lack of decorum? His words were certainly offensive to Catholics, but not baseless. He often spoke harshly, yes, but from a deep conviction.
Now, compare that to low-value speech: mindless abuse, crass outbursts, pure verbal venom with no ideas behind it. No one will be any wiser if you publicly call your opponent a dick. On the other hand: should we really be concerned about this?
Moreover, it is difficult to determine whether there is an opinion behind it. Think of someone projecting a swastika onto a synagogue wall, calling a politician a traitor, or depicting migrant workers as locusts. That kind of bile might seem an easy target for bans.
But here’s the rub: what if the politician is a traitor? Who decides whether crude language hides deeper meaning? And what’s really wrong with blunt speech? Calling a minister a retard may lack elegance, but it’s not necessarily meaningless.
Censorship’s cat-and-mouse game
Censorship also leads to creative evasion. In Germany, saying “Heil Hitler” is banned. So neo-Nazis began using “88” (H being the 8th letter of the alphabet). Then that got banned. So they moved on to “2x44”. What next: banning all maths problems that result in 88? It’s a game of cat and mouse.
Hurt feelings are tricky
Another challenge: hurt feelings are subjective. When exactly is someone allowed to feel offended?
Imagine a public figure constantly mocked for her hair. It’s wild, patchy, and unruly. All hairdressers despair. She’s sick of it, and it genuinely hurts her. Then another cheeky columnist publishes yet another mocking piece. She’s had enough. She sues for hate speech, saying her hair has become the target of constant ridicule. She wants to live with dignity — no wigs, no taunts — and she wants respect. Should the court convict the columnist?
Once you ban speech based on hurt feelings, you also incentivise people to exaggerate their pain. Imagine a devout believer who refuses to hear criticism of her religion. She could exaggerate her emotional distress as a weapon to ban even mild satire or reasoned critique.
Inferiority and the law
The German Constitution starts with: “Human dignity shall be inviolable. All state authority shall respect and protect it.” Dignity is violated, according to legal doctrine, when a person is portrayed as inferior, stripped of equality, or denied a rightful place in society.
Fair enough. That’s a noble aim. But what happens when legitimate opinion is swept up in the same net?
Take Germany’s Alternative für Deutschland (AfD) party. They oppose asylum immigration and are critical of cultural clashes, especially with Muslim newcomers — some of whom have lived in Germany for generations, hold German passports, and still struggle with integration.
These are real issues. The AfD taps into concerns shared by many citizens. Problems don’t fix themselves, they argue; they must be addressed. Whether you like the party or not, that’s a legitimate view.
Yet Germany’s domestic intelligence agency (BfV) recently declared that parts of the AfD are unconstitutional. Why? Because the party allegedly reduces people to second-class citizens based on their ethnicity. The AfD supposedly draws a line between “real Germans” and “passport Germans” — the latter being seen as lesser because of their migration background.
Are AfD members always polite? Far from it. Nuance tends to vanish in polarised societies. And with millions of supporters, you can’t reasonably expect everyone to behave.
But while Germany’s constitution wants to protect dignity, it can also become a tool to silence those who raise uncomfortable but real questions about immigration and integration. It shows how stretchy the concept of dignity can become, and how easily it bends to politics.
Everyone counts
British philosopher Bhikhu Parekh (b. 1935) argues that hate speech seeks to deny some people full membership of society. That feels reasonable, especially when you recall the fate of Jews under the Nazi regime.
But we shouldn’t forget that what’s considered “full membership” shifts over time. There was once a time when slavery was normal. Women couldn’t vote. Married women needed their husband’s permission to sign legal documents. And what about people nowadays without citizenship, migrant workers, the undocumented, or those under legal guardianship? They don’t have full membership either.
Given how much our view has changed over the years, do we really want to shut down further moves in the future? Even so drastically that no one may even publicly speak about it?
Imagine someone arguing that women should no longer be legally independent. Or that atheists should be denied a law degree. These are ridiculous ideas, deserving of mockery. But should they also be illegal to express?
Hate speech undermines democratic values
Hate speech (...) weakens democracy by arousing passions and irrational fears, making sweeping and indiscriminate generalizations about groups, creating a sense of insecurity among target groups, and discouraging their political participation. Like free speech, mutual respect, spirit of tolerance, equal dignity, and unconstrained and enthusiastic participation in the conduct of public life are also the lifeblood of democracy.
— Bhikhu Parekh, Is there a case for banning hate speech? (2012)
Parekh is making two main claims here:
Hate speech discourages political participation.
Hate speech undermines core democratic values.
Both claims aren’t very persuasive. You could just as easily argue the opposite: that hate speech actually mobilises the very communities it targets. People don’t just sit back and take it. They organise, resist, and fight for recognition. Minorities come together, demand change, and begin to assert their political influence.
As for the idea that democracy suffers when respect and equality break down, well, democracies often thrive on friction. Friction often leads to higher voter turn-out, and intensifies political engagement. Groups who wouldn’t usually get along are also forced to find ways to work together, whether by shouting at each other in parliament or — more commonly — by reaching compromise. Ethnic and religious minorities often hold the balance of power in coalition governments, trading support for policy concessions.
Legitimacy and the state
Parekh also offers a different angle:
When hate speech is allowed uninhibited expression, its targets rightly conclude that the state either shares the implied sentiments or does not consider their dignity, self-respect, and well-being important enough to warrant action. In either case, the state forfeits its legitimacy in their eyes and weakens its claim to represent them and demand their loyalty.
— Bhikhu Parekh, Is there a case for banning hate speech? (2012)
In other words, people who feel offended need the state to have their back. If the state stays silent, they’ll lose faith in it, and possibly withdraw from society altogether.
The angle is not without flaws. If your cat runs away, the police won’t help you look for it, even if it breaks your heart. Disappointment is often a matter of expectations. If it’s clearly understood that the state won’t intervene when, let’s say, a prophet is being ridiculed, then people won’t expect it to, and won’t be as disillusioned.
Sometimes, principled neutrality is better than a state constantly trying to please everyone or becoming a pawn in every culture war.
Is discrimination a crime?
Discrimination generally isn’t a criminal offence, but certain discriminatory statements are. That strange legal contradiction comes from the 1965 UN International Convention on the Elimination of All Forms of Racial Discrimination. It focuses on race alone — not gender, religion, or sexual orientation — and it requires states to ban the spread of ideas based on racial superiority or hatred, along with incitement to racial discrimination or violence.
At the same time, we know everyone discriminates. Our brains are wired to make snap judgments. Governments are expected to treat everyone equally — and rightly so. But individuals? We can’t. We’re not built that way. In fact, countless studies show we’re all unconsciously biased in some form.
So, discrimination is universal, but publicly promoting it is banned. That’s… a bit hypocritical.
Can racism be “unlearned”?
Legal scholar Mari Matsuda doesn’t see it that way. Racism, she argues, is learned, and what can be learned can also be unlearned, with help from the law.
Racism as an acquired set of behaviors can be disacquired, and law is the means by which the state typically provides incentives for changes in behavior.
— Mari Matsuda, Words that wound (1993)
But there are problems with Matsuda’s view. First, criminal law is a blunt instrument. It’s not good at changing deep-seated behaviours. Symbolic? Sure. Transformational? Rarely.
Second, racism isn’t just learned. It’s deeply rooted in our cognitive hardware. Education and culture shape how it’s expressed, but we instinctively distrust outsiders and judge people based on familiar features. Outlawing racist language may change what people say, but not necessarily what they think. Then a ban on racist language becomes similar to the criminalisation of homosexuality: outwardly you may not notice it anymore, but you don't have to cut deeply to see that it secretly occurs everywhere.
Third, limiting hate speech only in public makes little sense. Private speech between likeminded individuals remains untouched — unless we want a surveillance state. It’s like banning drug dealing only if it happens in a supermarket.
And finally, why criminalise only racist speech? Hate can target many features people can’t change: religion, background, sexuality, age, disability. If hate speech is about discrimination, why draw the line only at race?
From hate to harm
Violence is a necessary and inevitable part of the structure of racism. It is the final solution, as fascists know, barely held at bay while the tactical weapons of segregation, disparagement, and hate propaganda do their work. The historical connection of all the tools of racism is a record against which to consider a legal response to racist speech.
— Mari Matsuda, Words that wound (1993)
Speech affects beliefs and behavior over time, gradually moving people toward condoning or committing violence against members of another group.
— Catherine Buerger, Susan Benesch, Dangerous speech as an atrocity early warning indicator (2024)
It’s often said that hate speech leads to violence. And history is cited as proof: the fall of the democratic Weimar Republic that preceded Hitler in Germany, the Rwandan genocide (1990–1994), the wars in Yugoslavia (1991–1995). Since I’m writing this from the former Yugoslav republic of North Macedonia, I chose to look more closely at that last example.
The Yugoslav wars: a case study
In 1991, Yugoslavia began to unravel. Slovenia and Croatia declared independence. The war in Slovenia lasted ten days. Croatia was bloodier, mainly due to a large Serbian minority supported by the Yugoslav army. Roughly 200,000 ethnic Serbs eventually fled Croatia.
But the worst came in Bosnia and Herzegovina. That republic declared independence in 1992. Its population was evenly split between Croats, Serbs, and Bosniaks (Muslim Bosnians). Although they had lived peacefully side by side in many places for centuries, there were also regions in Bosnia where one ethnic group was dominant.
Bosnian Serbs, backed by Belgrade, rejected Bosnian independence. They declared autonomous regions and they besieged, among other places, the capital Sarajevo, while many Bosniaks lived there who saw no point in Serbian rule. The war that followed involved ethnic cleansing, concentration camps, mass killings, and a full-blown propaganda war.
Hate speech in wartime
All sides used media and political rhetoric to demonise the enemy.
In the Serbian press, Croats were portrayed as fascists, citing Croatia’s WWII fascism. But the worst abuse was aimed at Bosniaks: they were called “filth,” “dogs,” “gypsies,” “bitches” and “beasts.”
Wild rumours spread: Bosniaks feeding Serb babies to zoo animals. Or the other way around: Serbs feeding pregnant Bosniak women to the tigers and lions. A local police chief claimed that Bosniaks planned to circumcise all Serb boys, murder all men, and put the women in harems.
Of course, these kinds of messages fueled the fighting spirit on both sides. A sign of radicalisation with bloody consequences. More than 100,000 people died in those wars. Millions were displaced.
Hate speech and fake news, we must be extremely alert to that. As soon as you notice it, action must be taken before the conflict gets out of hand. That is at least the often-heard premise among those who combat hate speech. But if you take a closer look at the escalation in Yugoslavia, you see that this is not true.
Here’s the twist: the hate came after the guns came out.
Hate didn’t start the war
Before June 1991, things were relatively civil. Nationalist rhetoric existed, and ethnic stereotypes and historical grievances were no exception, but it was still within Category 1. Patriotic? Yes. Controversial? Sure. But no slurs, grave insults, or dehumanising language.
Croatian leaders spoke of a “centuries-old dream” of independence. Slovenians wrote of “Serbian domination”. In Serbia, the 1986 “Memorandum” from the Serbian Academy of Sciences argued for minority rights. Although the memorandum was nationalistically coloured and one-sided, there was not a single unpleasant word in it. The memorandum was a harbinger of the rising nationalism in Serbia, from which the later president Milošević profited. Even from Milošević, a dubious populist, not a single unpalatable word was recorded until the war.
The real link: radicalisation
The Yugoslav case shows a common misconception: that hate speech causes violence. Despite numerous attempts, the scientific literature has been able to demonstrate correlation but not a causal link between public hate speech and violence. Presumably, the causal chain is not:
discontent → hate speech → violencebut:
discontent → [hate speech + violence]Discontent does not get out of hand through hate speech, but through radicalisation. Understanding radicalisation is key. In future articles, we’ll dig deeper into that. But here’s a sneak preview.
Iranian psychologist Fathali Moghaddam (b. 1951) describes radicalisation as a five-storey pyramid. The broad base is not part of the radical organisation, but the radicals do stand up for the base. Who forms that broad base is easy to determine. The community of believers, the proletariat, the Western white race, the Basque people.
As you move up the pyramid, the isolation increases. The fifth floor is focused on action. This is where you live to commit or organise violence. The more radical and the more prepared to take action, the more isolated the activist. Those who are prepared to use violence do not advertise it.
The most dangerous radicals don’t shout in public. They operate quietly. Take the 2019 murder of German politician Walter Lübcke. A video of him supporting asylum seekers went viral. It was shared by far-right groups. Public outrage followed. But the killer didn’t comment online. He left no digital footprint. He acted in silence.
Barking dogs don’t bite. The quiet ones do.
Conclusion: arguments against free speech
We’ve now worked through the main arguments in favour of banning hate speech. Some hold water, others less so. In the next article, we’ll turn to the counterarguments — those for free speech and against banning hate speech. Only then can we properly weigh both sides.
But first, here’s a recap of the strongest points we’ve seen so far for restricting hate speech.
Research shows that racism and discrimination do real harm. Victims often compare the experience to physical pain. It fosters feelings of inferiority, self-imposed isolation, reduced civic engagement, and poor integration into society.
In the most extreme cases — especially when remarks are directed at an individual — a ban seems justified. Think of phrases like:
“They should have gassed you too.”
“Go back to your African monkey tree.”
It doesn’t matter if these are said privately or publicly. They cross a clear line.
But things become more complicated when such language takes the form of a political statement, a policy proposal, or a general contribution to public debate — not targeting one person, but referring to a group. As offensive as such remarks may be, they must still be weighed against the value of free and open public discourse. That’s where we’ll pick things up next time.
It’s certainly a problem when groups are portrayed as inferior, or when their equal right to be part of society is challenged. But we need to be careful not to shut down every policy discussion that relates to specific social or ethnic groups.
As for the other arguments in favour of banning hate speech — they’re on shakier ground:
The link between public hate speech and measurable social harm remains unproven.
The idea of “dignity” is vague and subjective, and must be balanced against the dignity of those who wish to speak freely.
You can’t suppress truths just because they’re expressed rudely. A civility police force would be a hallmark of totalitarianism.
Hurt feelings are too subjective and hard to regulate through law.
Claims that hate speech undermines democracy are poorly substantiated and feel speculative.
Criminalising discrimination by individuals is like banning people from breathing. It ignores basic human psychology.
No clear causal link has been shown between hate speech and acts of violence.
So no, we’re not done with hate speech yet. The two strongest arguments for banning it — social and psychological harm to individuals and social exclusion of groups — still need to be weighed against the principles of free speech.
That’s exactly what we’ll do in the next article.
For further reading
John Stuart Mill, On liberty (1859)
Michel Foucault, L’ordre du discours (1971), also available in English: The discourse on language (1972)
Joel Feinberg, The moral limits of the criminal law, volume 2: Offense to others (1985)
Mari Matsuda et al., Words that wound. Critical race theory, assaultive speech, and the first amendment (1993)
, Excitable speech (1997/2021)
Mark Thomson, Forging war: The media in Serbia, Croatia, Bosnia and Hercegovina (1999)
Bart den Uyl, Het drama Joegoslavië en het Servische Memorandum (2011)
Fatih Moghadam, The staircase to terrorism: a psychological exploration, American Psychologist (2005)
, In dubio. Vrijheid van meningsuiting als het recht om te twijfelen (2008)
Michael Herz, Péter Molnár (ed.), The content and context of hate speech: rethinking regulation and responses (2012):
— Péter Molnár, Interview with Robert Post
— Bhikhu Parekh, Is there a case for banning hate speech?Katharine Gelber, Luke McNamara, Evidencing the harms of hate speech, Social Identities (2015)
, Hate speech and democratic citizenship (2016)
Karsten Müller, Carlo Schwarz, Fanning the flames of hate: social media and hate crime, Journal of the European Economic Association (2021)
Luvell Anderson, Michael Barnes, Hate speech, The Stanford Encyclopedia of Philosophy (2022)
Catherine Buerger, Susan Benesch, Dangerous speech as an atrocity early warning indicator: Measuring changing conflict dynamics, Genocide Studies and Prevention: An International Journal (2024)
This was the fourth article in a series on censorship and freedom of speech. The next article will be published in a few weeks, and will discuss the blessings of free speech.