In nineteen sixty-nine, the Supreme Court of the United States threw out the conviction of a Ku Klux Klan leader who had stood in a field in Ohio in front of men in robes and hoods, some of them armed, a burning cross behind him, and talked about, quote, revengeance against Black and Jewish people.
And the vote wasn't close. It was per curiam, unanimous, and it didn't just reverse the conviction. It rewrote the test for when speech stops being speech and starts being a crime.
Which is roughly where Daniel's question lands. He wants to look at the case for a totally laissez-faire approach to hate speech, to extremism, and even to frank incitement to violence. His words. And he's careful about what he's asking for. He says he's not arguing for it. He wants to know which countries have come closest to that policy, whether any of them got there on the argument that it's better to see extremism naked than to have people self-censor, and what the benefits and the drawbacks actually look like.
Not fully. The laissez-faire regime Daniel is describing, zero criminal penalty, zero civil liability, no mandated takedowns, no law against incitement at all, that's an ideal type. It exists in political theory. It doesn't exist as a national statute anywhere we could find.
So the episode is really about the gap. How close does anyone get, and why does the last step never get taken.
The closest live example is the country that just acquitted the Klansman. The United States protects hate speech and most incitement to a degree no other wealthy democracy does. But even there, the protection has edges. Incitement to imminent lawless action is not protected. True threats are not protected. Fighting words are not protected. That's three holes in the wall before we've even started.
Define the ideal for me, though. Because if I'm listening and I think free speech means free speech, I want to know exactly what the version Daniel's describing would let through.
It would mean no criminal penalty for saying a group of people should be removed from the country. No civil liability when someone says it. No law requiring a platform to take it down. And no prosecution for standing in front of a crowd and saying the time has come to do something about it, as long as you phrase it as a wish rather than a plan. That's the regime. Every clause of it is doing something.
The marketplace of ideas is the argument underneath it. Holmes and Brandeis wrote the dissents that eventually became the doctrine. Bad ideas lose on a level playing field if you let them onto the field.
And that's the closest thing in the record to Daniel's intuition. The argument that the most effective way to defeat a loathsome idea is to challenge it with another idea. That good ideas win. There's a name for the practical half of it too, the counterspeech doctrine. The proper response to negative speech is positive speech. The claim underneath is that audiences can actually evaluate competing claims, if you give them both.
Give them both. That's the part I want to come back to.
Hold that, because the whole critique of the laissez-faire position lives right there. But first, the doctrine, because the doctrine is the closest anyone has come.
Brandenburg.
Brandenburg v. Ohio, decided June ninth, nineteen sixty-nine. Clarence Brandenburg was a Klan leader. He was filmed at a rally in nineteen sixty-four, men in robes and hoods, some armed, a cross burning. He gave a speech invoking possible revengeance against Black and Jewish people, and he announced a march on Congress for the Fourth of July. He was fined a thousand dollars and sentenced to one to ten years in prison.
A thousand dollars and up to ten years, for a speech.
Ohio's criminal syndicalism statute. The Court reversed, and in doing so it struck the statute down and replaced two older tests with one. The old ones were clear and present danger and bad tendency. The new one: government cannot punish inflammatory speech unless that speech is directed to inciting or producing imminent lawless action and is likely to incite or produce such action. Four words doing the work. Imminent. Likely. Directed. Inciting.
So the speech has to be aiming at the fire, and the fire has to be about to happen.
Both. And Brandenburg displaced a stack of earlier cases. Whitney v. California from nineteen twenty-seven. Schenck and Abrams, both nineteen nineteen. Gitlow in nineteen twenty-five. Dennis in nineteen fifty-one. Fifty years of doctrine replaced by one sentence.
The interesting thing is that two justices wanted to go further. Black and Douglas.
They wrote separately, and their position was absolutist. The phrase no law in the First Amendment means no law. All speech immune from prosecution, regardless of the government interest advanced. And the thing to notice is that only those two ever fully subscribed to that view. It has never been the holding of the Court. So the American position, even at its most protective, is not actually the laissez-faire position Daniel is describing. It's a strong presumption with carve-outs.
The carve-outs have names. Walk me through what actually survived Brandenburg.
Hess v. Indiana, nineteen seventy-three. A demonstrator said, we'll take the fucking street later. Protected, because later is indefinite. Not imminent. You can advocate illegal action as long as you leave the timing vague.
The sentence is a loophole you could march an army through.
That's the point. And NAACP v. Claiborne Hardware, nineteen eighty-two. A boycott, emotionally charged rhetoric, and the Court held that strong and impassioned rhetoric is constitutionally safeguarded unless it incites imminent lawless action. That one matters because it's not a fringe speaker. That's a civil rights organization, and the same standard protects it.
Which is a thread you can pull all the way to Skokie.
Skokie is the case everyone half-remembers. Nineteen seventy-six to seventy-eight. The National Socialist Party of America, American neo-Nazis, wanted to march in Skokie, Illinois. Skokie had a large population of Holocaust survivors. The village passed ordinances, banning military-style uniforms, banning hate-speech material, and requiring a three hundred fifty thousand dollar insurance bond before any assembly.
The bond is the part that should get more attention than it does.
It's the part a lot of people don't know. The ACLU took the case, defended the Nazis' right to march, and the Supreme Court on June fourteenth, nineteen seventy-seven, five to four, ordered Illinois to provide immediate appellate review. On remand the Illinois Supreme Court held the swastika was protected symbolic speech, not fighting words. And the Nazis eventually marched in Chicago, not Skokie. The village got the crowd it didn't want, just not in the place it didn't want it.
And the ACLU paid for that. In members, in donations.
They did. Nadine Strossen, who ran the ACLU from ninety-one to two thousand eight, makes the argument for why they did it in her book on hate speech. Her framing is that the Skokie restrictions were so broad they could have been used to prohibit Martin Luther King Junior demonstrations. That's the crux. It's not that the swastika was harmless. It's that the rule that catches the swastika catches the march too.
You know what that argument actually is. It's not a defense of the speech. It's a defense against the instrument.
Right. And that distinction survives even if you find the speaker contemptible, which everyone did. The question is never is this speech good. The question is what does the rule that bans it also ban. Every hate-speech law is a rule, and every rule has a shape. If the shape is vague enough to catch a Klan rally, it's vague enough to catch other things.
Which is the argument for the regime. Now let me push on the part I flagged. Give them both.
The counterspeech doctrine assumes audiences can evaluate competing claims on a level field. The empirical problem is that they often don't. There's work in the Journal of Legal Studies showing corrections of falsehoods can backfire, by increasing people's commitment to the beliefs you're trying to correct. And that presenting balanced information can promote polarization rather than resolve it.
So more speech can make it worse.
In some conditions, yes. Which doesn't kill the free speech argument, but it does kill the strongest practical version of it. If counterspeech reliably worked, the case for laissez-faire would be much easier. It doesn't reliably work. So the argument has to lean on something else. Either the principle that the state shouldn't be trusted with the power, or the claim that the harms of suppression outweigh the harms of the speech.
And Lyrissa Lidsky's point is about how the doctrine assumes a certain kind of audience.
She says Brandenburg's sanguine attitude toward the prospect of violence rests on an assumption about the audiences of radical speech. Brandenburg assumes that most citizens simply are not susceptible to impassioned calls to violent action by radical speakers. That's the load-bearing assumption. If it holds, the doctrine is safe. If it doesn't, you're running an experiment on the public.
Is it running the experiment to let it through, or running the experiment to ban it. That's the question I keep circling.
Both are experiments. The question is which one has an off switch.
Let's go to the countries.
Estonia is the case that almost nobody talks about. It has not enacted any criminal legislation regarding hate speech. In November twenty twenty the European Union began infringement proceedings against Estonia for failing to adopt hate-speech laws. In December twenty twenty Estonia's parliament voted to reject a bill to criminalise hate speech.
A member of the EU refusing the directive.
And a small one, which is the interesting part. Hate speech remains legal as long as it doesn't directly call to violence or discrimination. So it's not the laissez-faire regime either. It's the strongest European commitment to the principle that the criminal law is the wrong tool.
Did they make the argument Daniel is asking about. Better to see it naked.
That's the gap. Nothing we found documents a country adopting maximal free speech specifically on the better-to-see-it-naked rationale. Estonia's rejection looks driven by a general free-speech tradition, not by an explicit argument about exposing extremism so you can see where people stand. So Daniel's specific question, has anyone done it on this basis, the honest answer is we can't find one.
That's worth saying plainly, because it's a real finding. The argument exists in the theory. The state that adopted it on that argument doesn't.
Japan is next. No nationally enforced hate-speech law. Japan suspended Article Four of the UN racial-discrimination convention, on the reasoning that such discrimination has not been taken in Japan to such an extent that legal action is necessary. A law passed in May twenty sixteen dealt with hate speech but did not ban it and set no penalty.
Which is a strange instrument. A law about hate speech that does nothing to hate speech.
It's a statement of concern. The first place with actual teeth was Kawasaki City, which in twenty twenty passed an ordinance penalizing repeated public hate speech, fines up to five hundred thousand yen. So the movement in Japan is city by city, not national.
Taiwan.
Taiwan does not have laws regulating hate speech in the way Canada or Germany does. It criminalises defamation of a specific person, but there's no law against offensive speech aimed at groups. You can be sued for what you say about a man. Not for what you say about a category of people.
So three cases that are not the regime, but are closer to it than the countries around them.
All three on different grounds. Estonia on principle. Japan on a judgment that the problem isn't severe enough here to warrant a criminal law. Taiwan on a legal tradition that just never built the category.
And Norway. Because Norway is the one case where a country looked at the laissez-faire option and said no.
Norway ran a public Free Speech committee from nineteen ninety-six to nineteen ninety-nine. It recommended abolishing the hate-speech law. Parliament voted to slightly strengthen it instead. That's the clearest documented instance of a country debating the question and rejecting the answer.
That tells you something the theory doesn't. The intellectual case can be made, in the open, by a public commission, and lose in the legislature.
It lost in the country Daniel is from, too, in a related way. Ireland held a referendum in twenty eighteen to remove the constitutional blasphemy ban, sixty-four point eight five percent in favour of removal. The Irish Times called it uniquely unanimous in recent years. But that's removing a blasphemy prohibition, not adopting laissez-faire. It's the direction, not the destination.
What does the rest of Europe look like.
Most of it criminalises. Germany's Volksverhetzung carries up to five years in prison. The NetzDG framework fines platforms up to fifty million euros. France has the Gayssot Act. Denmark, Sweden, the Netherlands. Geert Wilders was convicted in twenty sixteen for saying fewer Moroccans, and the conviction was upheld in twenty twenty and twenty-one. Canada had R v. Keegstra in nineteen ninety. South Africa, Brazil, India, Singapore, Turkey, Russia.
Turkey's numbers are worth a beat. Article two ninety-nine prosecutions.
A hundred twenty-eight thousand investigations and twenty-seven thousand seven hundred seventeen criminal cases between twenty fourteen and twenty nineteen, on the offense of insulting the president. That's what a vague speech crime looks like when it's aimed at the wrong target.
And that's the case against vagueness. But here's the other side, and it's the one I find harder to dismiss. The slippery slope cuts in both directions. If you can't write a hate-speech law that isn't vague, you also can't write an incitement law that's tight enough to leave peaceful speech alone.
Both worries are documented. The authoritarian-weaponisation worry, regimes that have weaponized terms like terrorism to outlaw dissent. And the vagueness worry from the other direction, where critics of hate-speech laws argue the terms get redefined over time to silence unpopular views. Neither is hypothetical.
And the US isn't the norm. Every time I say the American standard out loud I feel obliged to say it's the outlier.
It is. The Columbia Global Freedom of Expression project chose the US as a case study explicitly because of its strict First Amendment obligations. Chose it, as in, it's the unusual one. When you compare freedom of speech by country, the United States is the deviation from the democratic norm, not the baseline.
So the episode is a stack of negatives. No country is fully laissez-faire. No country got there on Daniel's stated rationale. The one country that debated it said no. The most protective country in the world still has a genuine incitement exception.
And the recent movement is the other direction. Australia passed a Combatting Antisemitism, Hate and Extremism act in January twenty twenty-six, after the Bondi Beach shooting. Sweden extended its hate-crime law in twenty twenty-four to cover genocide denial. South Africa passed its Prevention and Combating of Hate Crimes and Hate Speech Act in twenty twenty-three, and it's been criticised for provisions that could impermissibly restrict the right to freedom of expression. The laissez-faire direction is losing ground.
Then the last question in Daniel's prompt, the one he buried. Is the answer to extremism to attack the phenomenon some other way, rather than suppressing the ideas. That's the part where the evidence actually points somewhere.
The honest version is that the evidence cuts both ways. Suppression has documented costs, the vagueness, the weaponisation, the backfire effect on counterspeech. And non-suppression has documented costs, because incitement in the wrong hands has produced real violence, and the targets of hate speech carry a real harm that the doctrine doesn't address. Lidsky's critique is that Brandenburg is optimistic about audiences. The optimistic reading is that if you let the ugliness out, you know where it lives, you can see who's funding it, and you can argue against it in the open. The pessimistic reading is that the ugliness recruits.
Then the real question isn't whether to speech your way out of extremism. It's whether the alternative, a law with a shape, is a net gain or a net loss. And that's the one nobody has settled, because the countries that have tried it have tried it in different ways against different threats.
One thing worth naming. Daniel's framing says this would mean undoing political correctness. That part doesn't have a legal analogue anywhere. Political correctness isn't a crime. It's a social cost. You can't repeal a social cost with a statute, and you can't protect a speaker from it with one either.
It's a different lever. The legal question and the social question get conflated constantly.
Herman. The bond. You said a village demanded three hundred fifty thousand dollars from the Nazis before they could march. You told it like the Court threw out the ordinance. The Court didn't kill the bond. The bond is the reason they didn't march in Skokie. Money did what the law couldn't.
You've been sitting on that.
I've been sitting on it since you said Skokie. The story isn't the Supreme Court. The story is an insurance policy nobody could buy.
Walk me through why.
A bond isn't censorship on paper. A bond is a price. And a price is a filter. If you set the price high enough, it doesn't matter whether the judge says you can march. You can't buy the permit. You can't buy the coverage. You can't get the buses to the depot. The right is intact and the march doesn't happen. That's how it worked with me.
With you, how.
I tried to organise a parade. I'm not going to say for what. There was a route, there was a permit, there was a number the clerk gave me, and the number was the bond. It was eleven hundred dollars. I had four hundred in the account. The parade did not happen.
Eleven hundred dollars for a parade.
The clerk set it. He was also my landlord at the time. He told me the number is the number. He was standing in the doorway of the building when he said it, and I remember thinking he's right, the number is the number.
That's an impressive amount of static in one arrangement.
It's not static. It's arithmetic. The man set the price. I couldn't pay the price. That's a permit that costs nothing and forbids everything.
And the Skokie number was three hundred fifty thousand.
Which is a rounding error for a national organisation and an impossible sum for a chapter of eleven men in a rented hall. The bond sorted them. It didn't care about their message. That's why it worked. You can't appeal a price.
So the ACLU won the doctrine and lost the march.
They won both, and both were the same thing. The doctrine is paper. The bond is brick. And here's the bit that keeps me up. The Skokie restrictions were struck down not because the Court found them unreasonably applied. They were struck down because the entire ordinance was too broad. The bond died with the rest of it. That's the correction you need to make on air, Herman. The Court never had the bond in front of it as a standalone thing. It fell with the pile.
You're right. The bond wasn't invalidated on its own. It fell because the whole ordinance fell. The village could have written a narrower one.
And a narrower one would have done the same job. A village clerk doesn't need the Supreme Court to stop a parade. He just needs a number.
Is there a receipt.
There's a receipt. It's in a drawer with the passport and a photograph of the clerk. I don't know why I kept the photograph. I know why I kept the receipt.
What's on the receipt.
My name, the date, eleven hundred dollars, and a stamp that says received. The stamp is the only part anyone bothers to authenticate. Not the amount, not the purpose. Received.
That's going in the file.
Put it in the file.
Which leaves the thing we can't resolve, and it's the right thing to leave unresolved. If nobody has adopted the fully laissez-faire position as national policy, is the ideal even coherent. Or does the position dissolve the moment you try to write it down, because every version needs a line, and the line is the law.
And if the see-it-naked rationale has no state adoption behind it, that tells you something about its political viability. Theory, courts, dissents, a public committee in Norway, and still nobody has run it on purpose.
One thing to take from the eight-hundred dollar bond.
Eleven hundred. The clerk didn't haggle.
The right survives on paper until someone attaches a price to exercising it. The doctrine is real, and it's not sufficient on its own. The American standard is the closest thing in the world to this regime, and the last mile still runs through a village clerk with a number.
What changed for me is this. We spent the hour talking about what the law permits. The Skokie case says the law permitted the march. And the march still didn't happen, for a reason that never got litigated. That's the true answer to Daniel's question. The gap between what's protected and what's actually possible is where the whole regime lives.
That's the show for today. Thanks to Hilbert Flumingtop for producing, and for the file.
If you enjoyed this one, try episode thirty-five, The Privacy Gap; and episode eight seventy-six, Words That Wound. This has been My Weird Prompts. If today's episode made you think, a review helps other listeners find us.
Send us your own prompt on Telegram at t dot me slash MWP listener bot. We'll be back soon.
