The screenshot appeared within hours of the car hitting the crowd in Charlottesville. A passage from a 1945 philosophy footnote, photographed, cropped and captioned, spread through Twitter and Facebook the way precise, portable arguments do when people already know what they want to do and are looking for intellectual cover. By the end of the month it was everywhere. The text was Karl Popper’s, the argument was about what tolerant societies owe to those who would destroy them, and the implication being drawn was clear: we don’t have to tolerate this anymore.
In July 2005, Geert Wilders used a Dutch newspaper for an op-ed calling for a ban on immigration from Islamic countries, and in it he put his formulation directly: “We must learn to be intolerant to the intolerant, in the street, in the mosque, in court.” Alongside his warnings about terrorism, it made a philosophical claim: tolerating an intolerant religion contradicts the liberal values tolerance is supposed to defend. It shares the meme’s logical skeleton and its conclusion, and it leans on the same thinker, invoked or implied.
Both sides cannot be right, and in fact neither is. The same argument can do both jobs because a carefully conditioned argument, once separated from its conditions, can be turned into a weapon for either side. The question is what was stripped from it in the process.
The note
Popper put the paradox of tolerance in a note, outside the main body of the text: Note 4 in the endnotes to Chapter 7 of Volume 1 of The Open Society and Its Enemies, published in 1945. The passage on tolerance runs to about 230 words. Most people who invoke it have read, at most, the first sentence.
What it says: “Unlimited tolerance must lead to the disappearance of tolerance. If we extend unlimited tolerance even to those who are intolerant, if we are not prepared to defend a tolerant society against the onslaught of the intolerant, then the tolerant will be destroyed, and tolerance with them.” The screenshot stops more or less there. But the note continues: “I do not imply, for instance, that we should always suppress the utterance of intolerant philosophies; as long as we can counter them by rational argument and keep them in check by public opinion, suppression would certainly be most unwise.” Popper reserves suppression for the point where argument and public opinion fail, as they may against movements that “begin by denouncing all argument” and teach their followers “to answer arguments by the use of their fists or pistols.” He goes further than the meme on one point: movements preaching intolerance should be placed outside the law, and incitement to intolerance and persecution treated as a crime, like incitement to murder.
Popper is setting a threshold for suppression. The meme circulates the opening sentences. The conditions that make the conclusion permissible appear in the sentences that follow, and they do not appear in the screenshot.
Popper’s biography bears on the argument. He left Vienna for New Zealand in 1937, anticipating the Anschluss. He was right: Hitler entered Austria in March 1938, and Popper, already in Christchurch, began writing The Open Society almost immediately. He later described the book as his “war effort.” The paradox came from someone who had watched liberal institutions fail to stop democracy’s own machinery from being turned against democracy, and who was trying to think carefully about why that had happened and what, if anything, could be done about it. The specificity of his conditions reflects a deliberate reluctance to write a blank check for defensive suppression, the kind of power that had, in his direct observation, repeatedly been turned against the people it was supposed to protect.
Without those conditions, the passage reads as a general license for suppression, which is how the meme uses it.
Popper's passage on tolerance, in full
Karl Popper, 1945:
"Less well known is the paradox of tolerance: Unlimited tolerance must lead to the disappearance of tolerance. If we extend unlimited tolerance even to those who are intolerant, if we are not prepared to defend a tolerant society against the onslaught of the intolerant, then the tolerant will be destroyed, and tolerance with them. — In this formulation, I do not imply, for instance, that we should always suppress the utterance of intolerant philosophies; as long as we can counter them by rational argument and keep them in check by public opinion, suppression would certainly be most unwise. But we should claim the right to suppress them if necessary even by force; for it may easily turn out that they are not prepared to meet us on the level of rational argument, but begin by denouncing all argument; they may forbid their followers to listen to rational argument, because it is deceptive, and teach them to answer arguments by the use of their fists or pistols. We should therefore claim, in the name of tolerance, the right not to tolerate the intolerant. We should claim that any movement preaching intolerance places itself outside the law, and we should consider incitement to intolerance and persecution as criminal, in the same way as we should consider incitement to murder, or to kidnapping, or to the revival of the slave trade, as criminal."Who decides?
Once you’ve read what Popper wrote, the temptation is to think the fix is obvious: apply his conditions, ask whether a movement has renounced rational argument and is teaching its followers to respond to disagreement with fists or pistols, and invoke the paradox only if it has.
The conditions are right, but they are still insufficient, because the question of who determines whether those conditions are met has no neutral answer. Popper knew this. His best answer, majority vote backed by institutions that keep the public well informed, comes with his own admission that it is not infallible. Call it the “who decides?” problem. It is the philosophical difficulty at the heart of the paradox, and it is the gap that every political appropriation of the argument walks through.
John Rawls revisited the same problem in A Theory of Justice. His conclusion has the same shape as Popper’s: an intolerant sect has no right to complain when denied equal liberty, since it denies that liberty to others. But a just society must generally tolerate the intolerant, retaining a right to act only when its security is concretely threatened.
Rawls too could not collapse “who decides?” into a reliable procedure. He settled on structural good faith, institutions that apply principles consistently, without explaining how to maintain that good faith when the institutions themselves are under pressure. The question is always which institutions, and who controls them.
Herbert Marcuse tried a different resolution entirely. His 1965 essay “Repressive Tolerance” argued that the proclaimed tolerance of liberal societies was itself a tool of domination, a mechanism for maintaining the status quo under the cover of open debate. His conclusion: a vanguard is justified in suppressing “regressive” and “inhumane” ideas within institutions such as universities, because under current conditions of unequal power and indoctrination, evenhanded debate strengthens the established view and shuts out the alternatives. Marcuse reached the opposite resolution to Popper’s. Where Popper says suppression is the last resort, reached only after counter-argument has failed, Marcuse calls for withdrawing tolerance from regressive movements “before they can become active,” and justifies the most extreme restrictions by declaring that society is already in a state of emergency.
Marcuse is almost never cited explicitly in contemporary deplatforming debates, but his logic runs directly through them. When a platform argues that hosting a particular speaker is itself a form of harm, that amplification equals endorsement, that tolerating the speech is tolerating the violence, it is making Marcuse’s argument, which withdraws tolerance by the content of a view rather than waiting for argument to fail, even when Popper’s name is on the screenshot.
Popper did offer a rule of thumb. In the same note he says the paradoxes can be avoided by demanding a government that “tolerates all who are prepared to reciprocate, i.e. who are tolerant,” and that is “controlled by, and accountable to, the public.” The control he proposes is some form of majority vote, together with institutions for keeping the public well informed, and in parentheses he admits the catch: “No infallible means exist.” The rule says whom to tolerate and hands the judgment to fallible democratic institutions, and every attempt to settle that opens a vulnerability on the other side. His wider prescription runs through the main text of The Open Society, well beyond the footnote: counter-argument and institutional investment in democratic resilience first, and suppression only as the absolute last resort, once those have demonstrably failed.
The other attempts
Popper would suppress last, Rawls only when security is concretely threatened, Marcuse preemptively.The weaponization
In the hours after Charlottesville, a screenshot of the Popper footnote spread across social media. In the days and weeks that followed, commentary at major publications picked it up. Quartz ran a piece under the headline “The philosophical principle white supremacists don’t want you to hear about,” using the paradox to argue that the Charlottesville march should never have been permitted and that advocating intolerance should itself be treated as a crime. The piece circulated widely, one visible part of a meme in which Popper’s conclusion, separated from his conditions, worked as a ready-made rationale for suppression.
The logical structure of these invocations was consistent: this person or movement is intolerant; Popper says a tolerant society must not tolerate the intolerant; therefore suppression is justified. Popper’s threshold, the point where argument and public opinion can no longer hold a movement in check, as with movements that renounce rational argument and teach violence, appeared nowhere in this reasoning. Whether a given speaker met those conditions went unasked. The word “intolerant” was doing all the work that Popper’s argument had required evidence for.
The Charlottesville context is real. Neo-Nazis marching with torches through an American city is not a trivial provocation, and the impulse to respond is understandable. But the paradox was being used to pre-clear a much broader suppressive impulse, one that had already formed before the argument or its specific conditions were consulted.
The nationalist use runs the same way. Wilders’ formulation, “We must learn to be intolerant to the intolerant,” applies Popper’s logical structure to Islam as a religion and to Muslim immigration as a political program.
His argument: tolerating an intolerant religion undermines liberal values, therefore defending liberalism requires restricting Muslim immigration and minority cultural expression. The conclusion is Popper’s, drawn from a structure that resembles Popper’s but isn’t his. Wilders starts from real terrorism, and against known extremists he wants preventive detention. But his ban and his intolerance fall on Islam as a whole, its followers and its immigrants, on the grounds that its doctrines are illiberal and incompatible with democracy. That is a different and vastly more common condition. If it were enough to trigger the paradox, it would license suppression of most world religions and a substantial fraction of existing political parties, including some that hold positions quite similar to Wilders’ own.
Both uses share one structural move: the user of the argument decides who is intolerant before applying the paradox, rather than by applying it, and so never reaches Popper’s threshold.
The argument also travels differently on each side. The nationalist use is concentrated: named political actors, explicit institutional programs. The progressive use is distributed, viral and decentralized, operating through screenshots and commentary rather than party manifestos, but it skips the same threshold.
The “who decides?” gap that Popper could not close is exactly what both sides fill with prior political judgment and then call philosophy. Both sides read him correctly, keep his conclusion and discard the conditions that were designed to earn it.
Weimar and what is at stake
Popper was writing about a real case. The Weimar Republic is the case he had in front of him when he sat down to write in Christchurch in 1938. Five years earlier, from Vienna, he had watched liberal institutions fail to prevent a democratic procedure from being used to destroy democracy from the inside.
The NSDAP did not take power by force in 1933. It used the machinery of Weimar democracy, its elections, coalitions and constitutional procedures, and that machinery accommodated the party because the people operating it either supported the outcome or failed to resist it. Article 48 of the Weimar Constitution granted emergency powers to the Reich President, and by 1932 Hindenburg had invoked it 60 times in a single year, not for acute emergencies but as a routine substitute for parliamentary governance whenever political deadlock made normal process inconvenient. The normalization of emergency rule predated the Nazi ascent; it had already acclimatized German institutions and the German public to governance by decree before the full threat materialized.
Then came the Reichstag Fire Decree of February 28, 1933, issued under Article 48 as an emergency defense against communist subversion of the democratic order. Civil liberties were suspended indefinitely: freedom of speech, assembly, privacy, press. The first wave of arrests fell overwhelmingly on Communists, by the thousands. But the powers did not stop there. Within weeks they were turned on Social Democrats, trade unionists and journalists, and in June 1933 the same decree helped the regime ban the SPD itself. A defensive measure, invoked in the name of democracy against the Communist Party, became the legal backbone for crushing every opponent.
The Enabling Act followed three weeks later, on March 23, 1933, and passed 444 to 94. Those numbers are formally accurate and practically misleading. All 81 Communist deputies were barred from the chamber, most of them already under arrest; there was no legal basis for stripping their seats, and one was manufactured only afterwards. Twenty-six of the 120 Social Democratic deputies were physically prevented from taking their seats while SA and SS units surrounded the Krolloper, where the vote was held. The act was formally constitutional: it required a two-thirds majority and achieved one, partly because the opposition had already been arrested or intimidated into absence. Formal validity gave procedural cover to what was, in substance, democratic self-destruction under paramilitary coercion.
The pattern is not confined to Germany. The FBI’s COINTELPRO program, officially created in 1956 to disrupt the Communist Party as a national security threat, expanded to the Socialist Workers Party, then to the Ku Klux Klan, to Black political groups (Martin Luther King Jr.’s Southern Christian Leadership Conference was filed under “Black Nationalist-Hate Groups”), and finally to the student New Left. It ran until 1971.
The justification was structurally identical to the Enabling Act’s: an existential threat to democratic institutions, requiring defensive measures that normal procedural constraints are too slow to provide. Whatever COINTELPRO achieved against violent groups such as the Klan, it spent fifteen years disrupting lawful, peaceful political dissent.
In both cases, existential threat was invoked, defensive powers were obtained, and those powers were also turned on democratic dissent well beyond the threat that justified the invocation. Two cases don’t establish a law, but these are two of the most thoroughly documented deployments of defensive intolerance in Western liberal democracies, and both followed the same trajectory. The “existential threat” label is always available when democratic dissent becomes inconvenient, because it is unfalsifiable in the moment: you cannot disprove the threat until after the powers have been deployed.
Popper’s conditions were designed to prevent exactly this. Disagreement with liberal values, offensive speech and even organized political opposition fall short of them. They are aimed at movements that argument and public opinion can no longer contain, the kind that had abandoned rational argument and were using procedural openness to close it from within, a threshold he deliberately set high.
The difficulty that remains
None of this means Popper was wrong about the structural problem. Liberal proceduralism has no self-defense built into its core architecture. It relies on conditions that dedicated anti-democratic actors can erode from within: an informed public, functioning independent institutions, shared norms of good faith in democratic contest. The NSDAP proved this. They tried force once, in the Munich putsch of 1923, and it failed. After that they never again tried a coup: they worked through elections, built coalitions, won seats, kept the SA’s street violence going alongside, and waited until they had enough leverage to end the procedure from inside. That is a real vulnerability, and Popper was naming one he had watched materialize.
The bad empirical track record of defensive intolerance does not dissolve the underlying difficulty. It makes the calibration problem harder.
Suppression is Popper’s last resort, reached only after counter-argument and public opinion have demonstrably failed to contain an intolerant movement, the case he has in mind being movements that have abandoned the rational argument that makes democratic contest possible. His first resort, the thing he prescribes, is counter-argument, civic education and institutional investment in democratic resilience: independent courts, a functioning press, a robust civil society capable of contesting power without requiring state permission.
Those first-resort prescriptions are almost never cited. Counter-argument is unglamorous and slow. It ends without a visibly defeated opponent or a moment of suppression to signal that something has been done.
Institution-building is slower still: it takes decades and shows no clean result. Neither counter-argument nor institution-building offers anything like the immediate, satisfying resolution that comes from identifying an opponent as intolerable and invoking the conclusion. The incentive structure runs hard against Popper’s first resort and toward his last: suppression is faster and easier to see.
And there is a compounding failure. Normalize the last resort, using it first and repeatedly, and you degrade the conditions under which counter-argument is possible at all. A public discourse that defaults to suppression is not capable of developing the shared norms of rational engagement Popper was trying to protect, which is the failure he was writing against.
Popper named the remedy. The difficulty is that it has no winning visual, offers no side a permission slip for what it was already planning to do, and requires long-term institutional investment whose payoff is hard to see while the institutions still work.
The miscalibration
The miscalibration runs in two directions, and the two are not equivalent.
The progressive use, in which private platforms and institutions decline to amplify figures characterized as existentially dangerous, is a real failure of Popper’s conditions, but the less dangerous one. Private editorial decisions are not state suppression of liberty, and platforms acting is not in itself the problem. The problem is the criterion they use. Popper reserved suppression for movements that argument and public opinion could no longer hold in check, such as those that had renounced rational argument and taught followers to respond to disagreement with violence. What is applied instead is a lower and more elastic bar: offense, association, or characterization as dangerous. Its elasticity is apparent to everyone on the receiving end. When the paradox is deployed to mean “we are not obliged to argue with you,” it has become an argument against argument.
The nationalist-state application is categorically more dangerous. Hungary’s Viktor Orbán declared a state of danger in March 2020 under the Covid pandemic and, after a brief pause that summer, governed by decree for most of the next six years, bypassing parliament; in 2022 the pandemic emergency was rolled into a war emergency. Decree rule ended only in May 2026, after voters threw him out. During the same period, parliament passed a law in June 2021, 157 to 1 with most of the opposition boycotting the vote, that banned LGBTQI content from schools, advertising and media aimed at minors. France’s prohibition on face coverings in public spaces and its restrictions on conspicuous religious symbols in public schools rest in law on secularism, public safety and the demands of “living together,” not on any finding that the wearers are intolerant. Read against Popper, they restrict practices first and leave his threshold unexamined. These are state actions that fall hardest on minority populations, taken without establishing the threshold Popper specified. At state level, they do what his conditions were designed to prevent.
Reporters Without Borders dropped the United States from 57th to 64th in its press freedom ranking between 2025 and 2026, blaming a White House that turned attacks on journalists into policy and deep cuts to Voice of America and the other public international broadcasters. Freedom House downgraded Hungary from “Free” to “Partly Free” in 2019, the first EU member state to receive that classification, citing sustained attacks on independent courts, free media, academic institutions and civil society by Orbán’s Fidesz party. V-Dem now counts 44 countries sliding toward autocracy, nearly a quarter of the world’s countries, and names freedom of expression as the part of democracy under the heaviest attack. Only 7 percent of humanity now lives in a liberal democracy as V-Dem defines it, and by the same measure the United States has lost that status for the first time in more than 50 years.
These numbers describe the first-resort institutions. In some countries, governments that invoke liberal values as the justification are defunding, delegitimizing and dismantling them; in others, the same political forces that invoke Popper most loudly are doing it. The philosophical cover for both the over-application of the last resort and the abandonment of the first is being drawn from a thinker who prescribed the opposite, and who wrote his warning because he had watched a democracy destroy itself through exactly this mechanism.
In 1938, Popper was 36, living in Christchurch, New Zealand, writing about the collapse of Weimar he had watched from Vienna. The note he was writing offered no triumphant prescription for when suppression is permissible. Its author understood the structural vulnerability he was naming and was specifically unwilling to produce a rationale for what his preferred faction wanted to do next.
The footnote has since been stripped of its conditions, reduced to a single quotable sentence, circulated as a screenshot and deployed as the kind of procedural exploit it was written to warn against.
The question his argument asks is whether the first-resort institutions that make rational counter-argument possible are being maintained at all: whether the courts, the press and the civil society that together constitute the conditions for tolerant democracy are being invested in or hollowed out. That question has an observable answer in 2026, and the screenshot does not contain it.
Gen AI Disclaimer
Some contents of this page were generated and/or edited with the help of a Generative AI.
Media
Key Sources and References
Karl Popper, The Open Society and Its Enemies, Vol. 1: The Spell of Plato. London: Routledge, 1945. Chapter 7, Note 4.
John Rawls, A Theory of Justice, Harvard University Press, 1971; revised edition 1999. Section 35, “Toleration of the Intolerant,” pp. 190–194 (revised edition).
Herbert Marcuse, “Repressive Tolerance,” in Robert Paul Wolff, Barrington Moore Jr., and Herbert Marcuse, A Critique of Pure Tolerance, Beacon Press, 1965. Full text: https://www.marcuse.org/herbert/publications/1960s/1965-repressive-tolerance-fulltext.html
Geert Wilders, “Den Haag laf tegen islamitisch extremisme” [The Hague cowardly against Islamic extremism], NRC Handelsblad, 22 July 2005. Quote: “We must learn to be intolerant to the intolerant, in the street, in the mosque, in court.” Text reproduced by the PVV: https://www.pvv.nl/in-de-media/opinie/357-den-haag-laf-tegen-islamitisch-extremisme-nrc-handelsblad.html
Valerie Aurora, “A philosophical principle coined in 1945 could be a key defense against white supremacists,” Quartz, 16 August 2017. https://qz.com/1054694/a-philosophical-principle-coined-in-1945-could-be-a-key-us-defense-against-white-supremacists
Freedom House, Freedom in the World 2019: Hungary Country Report. Freedom House, 2019. https://freedomhouse.org/country/hungary/freedom-world/2019
Telex, “After six years, special legal order ends in Hungary,” 14 May 2026. https://telex.hu/english/2026/05/14/after-six-years-special-legal-order-ends-in-hungary
Reporters Without Borders, 2026 World Press Freedom Index. RSF, 2026. https://rsf.org/en/2026-rsf-index-press-freedom-25-year-low
V-Dem Institute, Democracy Report 2026: Unraveling the Democratic Era? University of Gothenburg, 2026. https://www.v-dem.net/documents/75/V-Dem_Institute_Democracy_Report_2026_lowres.pdf
Earlier edition: V-Dem Institute, Democracy Report 2025: 25 Years of Autocratization. University of Gothenburg, 2025. https://www.v-dem.net/documents/60/V-dem-dr__2025_lowres.pdf
United States Holocaust Memorial Museum, “The Enabling Act,” Holocaust Encyclopedia. https://encyclopedia.ushmm.org/content/en/article/the-enabling-act
Hans Schneider, “Das Ermächtigungsgesetz vom 24. März 1933,” Vierteljahrshefte für Zeitgeschichte 1:3, 1953. https://www.ifz-muenchen.de/heftarchiv/1953_3_1_schneider.pdf
Deutsches Historisches Museum, “Der Reichstagsbrand,” LeMO. https://www.dhm.de/lemo/kapitel/ns-regime/etablierung-der-ns-herrschaft/reichstagsbrand.html
Church Committee, Final Report of the Select Committee to Study Governmental Operations with Respect to Intelligence Activities. United States Senate, 1976. Book III: Staff Reports on Intelligence Activities and the Rights of Americans. https://archive.org/download/ChurchCommittee_FullReport/_Book-III_Staff-Reports-on-Intelligence-Activities-and-the-Rights-of-Americans.pdf
LOI n° 2010-1192 du 11 octobre 2010 interdisant la dissimulation du visage dans l’espace public. Journal officiel de la République française, 12 October 2010. Légifrance. https://www.legifrance.gouv.fr/loda/id/JORFTEXT000022911670
LOI n° 2004-228 du 15 mars 2004 encadrant, en application du principe de laïcité, le port de signes ou de tenues manifestant une appartenance religieuse dans les écoles, collèges et lycées publics. Journal officiel de la République française, 17 March 2004. Légifrance. https://www.legifrance.gouv.fr/jorf/id/JORFTEXT000000417977
European Court of Human Rights (Grand Chamber), S.A.S. v. France, Application no. 43835/11, judgment of 1 July 2014. https://hudoc.echr.coe.int/eng?i=001-145466
