16 May 2026

After publishing this essay in an earlier form, I received a criticism that was both friendly and exact. The argument, I was told, was sharp and necessary — but too front-loaded and too generalizing. I had begun too high up in the philosophy. I had spoken too broadly about “Sweden,” as if the whole country, the whole intellectual culture, or the whole state had lost its rational capacity.

That criticism was right.

So this revised version begins by correcting itself.

By “Sweden” I do not mean the Swedish people. I do not mean every Swedish institution. I do not mean the whole Swedish intellectual tradition, which contains powerful resources for objectivity, public reason, secular criticism, democratic reform, and fallibilist humility. I mean something narrower: a recurring pattern in Swedish public handling of Islam, Islamism, Muslim representation, religious organizations, minority politics, immigration, and public funding.

The problem is not that Sweden lacks intelligence. The problem is more specific and more difficult: in this field, Swedish institutions and intellectuals often try to do the right thing, but use the wrong test.

They want to protect Muslims from suspicion, hatred, racism, discrimination, and collective blame. That is necessary. They want to distinguish Islam from Islamism. That is necessary. They want to respect religious freedom. That is necessary. They want the state to avoid judging theological truth. That is necessary.

But in doing all this, they often lose the harder distinction on which an Open Society depends:

protecting persons is not the same as immunizing symbols, doctrines, organizations, or norm systems from criticism.

That is the problem this essay examines.

It does so through three cases: Malmö city’s election flyer, Simon Sorgenfrei’s argument about mosques, halal and the headscarf, and Frederic Brusi’s memorandum on the Muslim Brotherhood in Sweden. The cases are different. One is visual and municipal. One is academic and conceptual. One is bureaucratic and legal. But together they reveal the same pattern: Sweden often protects persons in intention while shielding symbols, organizations, and power structures in practice.

That is not rational compassion.

It is intelligent avoidance.

The distinction Sweden must recover: person, doctrine, organization

The first principle must be absolute: the Muslim is the person.

The doctrine is not the person.

The symbol is not the person.

The organization is not the person.

The gatekeeper is not the person.

A democratic society must protect Muslims from hatred, discrimination, suspicion, harassment, and violence. No Muslim family should be treated as a problem merely because it is Muslim. No woman with a headscarf should be made the target of political rage. No mosque should be attacked as if the building itself were an enemy. No citizen should be forced to carry the burden of a doctrine, organization, or political theology merely because of name, background, ancestry, or religious identity.

But protecting persons is not the same as protecting every symbol, institution, or norm system from criticism.

This is where Swedish debate often collapses.

One side treats Muslim visibility as a threat in itself. That is crude and unjust. The other side, reacting against that crudity, often treats Muslim symbols and institutions as if they were analytically innocent simply because they are not automatically extremist. That is also a mistake.

The Open Society must do something harder:

protect the person and examine the system.

The question is not whether Muslims should be safe. They must be.

The question is whether Swedish institutions can examine the social functions of religious symbols, organizations, gender norms, and political-theological networks without being accused of attacking Muslims as persons.

If Sweden cannot do that, it has not protected pluralism.

It has disabled criticism in the name of pluralism.

The first error: Malmö and the aesthetic narrowing of inclusion

Malmö city’s election flyer for the 2026 election is, on the surface, exactly the kind of thing a democratic municipality should produce. It informs voters that early voting begins on 26 August and continues until election day on 13 September. It reminds them to bring identification. It explains that vote receivers in white vests will be present. It emphasizes that the election is secret and that every vote counts. The flyer appears in several languages: Swedish, English, Arabic, Dari, Persian, and Somali.

As a democratic act of accessibility, this is good and necessary. Citizens must understand how to vote. Bureaucracy should not become a barrier to participation.

But the flyer does not communicate only through text.

It communicates through image.

The same illustration appears across the language versions: a vote receiver in a white vest meets a woman with a headscarf or veil, while a hand places a ballot into the ballot box. The scene is simple. But it is not neutral. It says something about who the municipality imagines as needing invitation into democratic participation.

The generous reading is obvious. Malmö wants to say that those who do not belong to the old Swedish majority culture, those with visible minority markers, those who may be new to Sweden, are also part of democracy.

That intention is good.

But representation is not exhausted by intention. Images work through repetition, recognition, and symbolic choice. A public institution can make an error precisely while trying to do good.

The deeper question is this:

when Malmö imagines the Muslim or minority-marked woman, why does it imagine her as veiled?

The problem is not the presence of the veiled woman. She is a citizen, voter, neighbor, worker, mother, student, and democratic subject. She must be visible. She must be protected. She must not be mocked, threatened, or excluded.

The problem is the solitude of the symbol.

If the veiled woman becomes the standard visual shorthand for Muslim female inclusion, then other women disappear: the secular Muslim woman, the ex-Muslim, the liberal believer, the woman who has removed the veil, the woman who fled moral policing, the woman for whom the veil is not dignity but memory of coercion.

For such a woman, Malmö’s image may not feel like inclusion. It may feel like a second symbolic erasure. First, the religious or political system made the veil her assigned identity. Then the Swedish municipality, trying to include her, once again makes the veil the recognized image of her minority existence.

This is not the same as coercion.

But it is symbolic re-inscription.

The error is subtle because it is produced by good intentions. Malmö does not attack the secular Muslim woman. It simply fails to imagine her. It does not impose the veil. It turns the veil into the visual shortcut through which minority womanhood becomes recognizable.

That is why the aesthetic error is so important: the Open Society may exclude not by hostility, but by narrowing the image through which it recognizes the people it wants to include.

Here we see the difference between administrative intelligence and rational correction. Malmö’s intelligence is visible: translation, accessibility, democratic outreach, procedural clarity. But a more rational institution would ask the uncomfortable question:

Who disappears from this image?

Which internal minority within the minority is erased? Which symbol is being normalized as the face of democratic inclusion? Which women are represented, and which women are made harder to see?

In the aesthetic error, kindness becomes a picture, and the picture becomes a cage.

The second error: Sorgenfrei and the insufficiency of labels

Simon Sorgenfrei is right about one important thing: mosques, halal, Ramadan, calls to prayer, and the headscarf are not in themselves Islamism.

That distinction must be made. A society that cannot distinguish between Muslim persons, Muslim piety, Islamic tradition, and Islamist political mobilization becomes unjust and intellectually dishonest.

But the distinction is not enough.

The conceptual error is not that Sorgenfrei makes a false distinction. It is that he treats a true distinction as if it were sufficient.

His argument is animated by a legitimate fear: that the concept of Islamism may expand so broadly that it becomes a weapon against Muslims as such. If every Muslim expression is called Islamism, then the concept loses precision and ordinary religious life becomes suspect.

That danger is real.

But there is an opposite danger: that because something is not Islamism, it is treated as socially uncomplicated.

This is where Swedish debate often becomes evasive. It stops at the label and avoids the function.

Labels are necessary because they prevent crude generalization. But labels become anti-rational when they terminate inquiry before the function has been examined. A symbol does not become harmless merely because it is older than Islamism. An institution does not become liberal merely because it is not violent. A practice does not become private merely because it is religious.

It is not enough to say:

“The headscarf is not Islamism.”

The rational question is:

What social function does the headscarf perform in this context?

For one woman, it may be voluntary religious conviction. For another, family expectation. For another, a sign of modesty. For another, a strategy to avoid suspicion. For another, a political emblem. For another, a memory of coercion.

The same symbol can mean freedom for one woman and unfreedom for another.

A mature society must be able to handle both.

It is not enough to say:

“Mosques are not Islamism.”

The rational question is:

What ideas, authorities, gender norms, loyalties, and boundaries are reproduced in this institution?

What is taught about apostasy, blasphemy, Jews, Christians, atheists, women’s autonomy, homosexuality, Swedish law, religious authority, and the right to leave?

It is not enough to say:

“Halal is not Islamism.”

The rational question is:

When is religious practice private conscience, and when does it become collective boundary-making?

This distinction between label and function is decisive.

The conceptual error consists in mistaking classification for understanding.

The Swedish debate often asks:

Is this Islam?

Is this Islamism?

Is this extremism?

Is this piety?

Is this discrimination?

But the Open Society must also ask:

Does this practice create freedom or obedience?

Does it open doubt or close doubt?

Can the individual say no?

Can women deviate?

Can children grow beyond inherited identity?

Can ex-Muslims speak openly?

Can reformers question prophet, scripture, law, tradition, and authority?

Can the Muslim use their own Backspace Key?

This is where the concept of Islamism must be handled with precision. Islamism is not only violence. Political theology can be law-abiding and still anti-liberal. A movement can reject violence and still work for gender separation, religious social control, group discipline, suspicion toward apostates, distance from majority society, and stronger religious boundaries.

The test is not only:

Is it violent?

The test is:

Is it corrigible?

Can the system tolerate a woman who refuses religious marking? Can it tolerate a child who leaves the faith? Can it tolerate satire? Can it tolerate historical criticism of scripture? Can it tolerate the claim that Islam is a humanly transmitted tradition rather than an immune authority?

If the answer is no, then the Open Society has a problem even without violence.

Sorgenfrei’s necessary distinction therefore needs a second distinction. Yes, mosques, halal and the headscarf are not automatically Islamism. But they are not analytically neutral either. They must be examined functionally.

Otherwise nuance becomes evasion.

The third error: Brusi and the confusion of compliance with openness

Frederic Brusi’s memorandum on the Muslim Brotherhood in Sweden is the deepest of the three cases because it moves from symbol and concept to state judgment.

The issue is not simply one document, one author, or one administrative conclusion. The memorandum reveals a broader problem: the modern bureaucratic state’s difficulty in distinguishing visible legal compliance from real democratic openness.

A democratic state assessing whether a religious organization should receive public support stands at a dangerous intersection. On one side is religious freedom. On the other is the protection of the cognitive conditions that make pluralism possible: criticism, exit, conscience, open inquiry, and the individual’s right to revise inherited authority.

The administrative error is this:

the state asks whether the visible legal threshold has been crossed, but fails to ask whether the social conditions of freedom are being weakened beneath that threshold.

The Brusi memorandum appears intelligent in the bureaucratic sense. It is cautious. It respects legal thresholds. It avoids crude accusation. It asks whether the available Swedish material proves that organizations such as Islamiska Förbundet i Sverige, FIFS, or related actors openly violate democratic conditions.

Do they incite violence?

Do they urge members to break Swedish law?

Do they operate an explicit competing legal system?

Do they openly attack democratic order?

Within that narrow framework, Brusi finds the available evidence insufficient.

In a strictly legal-administrative sense, this may be understandable.

But rationality requires a deeper question:

The absence of visible illegality is not the same as the presence of democratic openness.

A structure can be formally law-abiding and still cultivate an internal culture in which criticism, apostasy, women’s autonomy, secularization, satire, theological revision, or independent conscience becomes socially impossible.

It can speak the language of democracy outwardly and function disciplinarily inwardly.

It can participate in civil society while building a parallel moral ecosystem where individuals are protected from the Open Society’s influence rather than prepared to live freely within it.

This is the inductivist illusion: the state sees no smoking gun and concludes there is no fire. It sees no explosion and concludes the building is safe.

But the building may be silent because the alarms have been disconnected.

A critical-rationalist state would not ask only:

Has the organization broken the law?

It would also ask:

Can inherited authority be criticized?

Can members leave without social death?

Can women reject the group’s gender norms without punishment?

Can young Muslims become secular, liberal, reformist, or ex-Muslim without being treated as traitors?

Can the organization revise its doctrines when they collide with individual freedom?

This is the Backspace Key test for public funding.

A democratic state should not fund only what remains legally quiet. It should fund what leaves the human being freer to criticize, dissent, exit, and revise inherited authority.

The state must be neutral toward theological truth. It should not decide whether Islam, Christianity, Judaism, atheism, or any other worldview is metaphysically true.

But neutrality toward theology must not become blindness toward power.

A state need not judge whether a doctrine is true in order to judge whether an organization strengthens or weakens the conditions of freedom.

The relevant question is not:

Is this theology true?

The relevant question is:

Does the institution built around it protect the individual’s right to doubt, object, reinterpret, leave, and live as a citizen rather than as the managed representative of a collective religious identity?

There is another danger: liberal vocabulary can be mistaken for liberal structure.

A document may use words such as democracy, pluralism, integration, participation, secularism, and diversity. But words do not prove internal freedom. East Germany called itself the German Democratic Republic. The word “democratic” did not make it democratic.

A closed system can wear the Open Society’s language as camouflage.

This is why administrative intelligence can become dangerous: it may read the vocabulary of openness while failing to examine the structure of closure.

The question is therefore not whether liberal words appear.

The question is whether internal conditions correspond to them.

Can members criticize founders?

Can inherited doctrines be historicized?

Can women reject conservative gender roles?

Can dissenters leave without moral excommunication?

Can the organization publicly reject anti-liberal elements in the Muslim Brotherhood tradition?

Can Muslim identity be interpreted by individuals rather than administered by organizational gatekeepers?

If not, liberal vocabulary may be aesthetic camouflage.

The deepest category error is “civil society.” A legally registered religious association may have the outer form of civil society while lacking its inner ethic. Civil society in the Open Society is not merely a legal category. It is the space where free persons associate, disagree, reform, leave, and create new forms of belonging.

A closed ideological organization can adopt civil society’s form while weakening civil society’s function.

The question is therefore not only whether an organization is formally part of civil society.

The question is whether it increases or decreases the individual’s capacity for independent judgment.

Does it produce citizens or managed identities?

Does it protect conscience or collective discipline?

Does it serve the person or the organization’s claim over the person?

If the state does not ask these questions, it may subsidize the very structures that make some individuals less free.

In the administrative error, the state does not abolish freedom directly. It recognizes, funds, and legitimizes structures without first asking whether freedom inside them is real.

The disappearing secular Muslim

The most morally urgent consequence of this drift is the disappearance of the secular Muslim, the liberal Muslim, the reformist believer, the ex-Muslim, the woman who rejects conservative gender norms, and the young person who wants to think freely.

This disappearance occurs in all three cases.

The aesthetic error makes the secular or dissenting Muslim hard to see.

The conceptual error makes their criticism harder to formulate.

The administrative error makes the gatekeeper harder to challenge.

What begins as inclusion can therefore end as a new structure of invisibility.

In the Malmö flyer, the Muslim or minority woman appears through the veil. The woman who came to Sweden to escape compulsory religious visibility disappears.

In Sorgenfrei’s argument, the concern to avoid collective suspicion of Muslims risks making Islamic symbols and institutions analytically too safe. The ex-Muslim, the unveiled woman, and the internal dissident become secondary to the fear of stigmatizing religious expression.

In the Brusi memorandum, organized religious actors risk being treated as civil society representatives of Muslims. But when the state legitimizes the most organized religious gatekeepers, the internal critic loses leverage. She can now be told not only that she betrays the group, but that Swedish institutions themselves recognize the group’s representatives as democratic.

This is the cruel irony of bureaucratic inclusion.

The state thinks it is protecting a minority. In practice, it may strengthen organized authorities over that minority.

This is what I call suicidal empathy: the tendency of Open Society institutions to sympathize with group identity in a closed system at the expense of individual persons trapped within it.

Empathy for persons is indispensable. But empathy becomes suicidal when it grants immunity to the structures that discipline persons.

Liberal democracy’s task is not to protect the strongest gatekeepers within minority communities.

Its task is to protect the individual — especially when that individual stands against both majority prejudice and minority coercion.

How the three errors reinforce one another

The three errors are not isolated. They reinforce one another.

The aesthetic error normalizes the symbol.

The conceptual error protects the symbol from functional analysis.

The administrative error then legitimizes the organizations and structures that operate behind the symbol, provided they remain legally cautious.

The result is not an open conspiracy. It is something more Swedish, more bureaucratic, and more difficult to criticize: a chain of intelligent evasions.

The municipality says: we only wanted to include.

The academic says: we only wanted to distinguish Islam from Islamism.

The bureaucrat says: we only followed the evidence threshold.

Each statement may be true.

That is precisely why the problem is dangerous.

Anti-rational drift often does not begin with bad intentions. It begins when good intentions are protected from correction.

This is the sequence:

the image narrows reality

the concept prevents the narrowing from being criticized;

the bureaucracy then treats the narrowed reality as administratively legitimate.

That is how an Open Society can remain intelligent while losing rationality.

What causes this Swedish mistake?

The mistake does not come from stupidity. It comes from several overlapping habits.

The first is fear of collective suspicion. Sweden has good reasons to avoid anti-Muslim hysteria. European history contains too many examples of minorities being turned into collective threats. But the fear of collective suspicion can itself become paralyzing. It can make legitimate criticism of doctrines, symbols, and organizations feel morally dangerous.

The second is an overconfidence in categories. Swedish public language often relies on reassuring distinctions: Islam is not Islamism, religion is not extremism, representation is inclusion, civil society is democratic, neutrality is fairness. These distinctions are often necessary. But they become dangerous when they stop inquiry instead of beginning it.

The third is bureaucratic visibility. The Swedish state sees what its instruments are designed to see: statutes, documents, formal representation, legal thresholds, funding criteria, organizational structures, compliance language. It is less good at seeing informal power: shame, family pressure, fear of apostasy, gender discipline, gatekeeper authority, reputational punishment, internal silence.

The fourth is humanitarian self-image. Sweden likes to understand itself as humane, inclusive, anti-racist, and rational. Those are real virtues. But virtues become fragile when they are protected from criticism. A society that wants to see itself as kind may become reluctant to ask whether its kindness is strengthening unkind structures.

The fifth is the loss of its own secular-critical canon.

Sweden has not lacked intellectual resources for this problem. There is a Swedish tradition of objectivity, rational public conversation, provisional concepts, anti-dogmatism, and democratic examination of values. One can think of Hedenius, Hägerström, Gunnar Myrdal, Evert Vedung, and the broader tradition of rational political debate. Alongside Popper and the Open Society, these thinkers remind us that public concepts are tools, not sacred identities; that moral and political claims must be exposed to criticism; and that democratic society depends on the capacity to test its own assumptions.

This tradition is, in Swedish terms, a Backspace Key.

It says: our categories are provisional. Our values must be discussable. Our public language must be corrigible. No concept should become immune because it serves a good intention.

That tradition needs to be reactivated.

Intelligence is not rationality

We can now return to the philosophical distinction — not as the starting point, but as the conclusion earned by the cases.

Sweden is not an unintelligent society.

That is precisely why its present drift is serious.

A crude critic would say that Sweden has become naïve, stupid, weak, or blind. That language misses the deeper problem. Sweden remains one of the most administratively intelligent societies in the world. It can translate public information into multiple languages. It can produce careful legal memoranda. It can design inclusive public campaigns. It can fund civil society, classify organizations, invoke democratic values, and debate religion through academic nuance.

It has institutional intelligence, bureaucratic intelligence, humanitarian intelligence, legal intelligence, and communicative intelligence.

But intelligence is not rationality.

Intelligence can administer an inherited frame. Rationality asks whether the frame is false.

Intelligence can optimize a procedure. Rationality asks whether the procedure should exist.

Intelligence can include. Rationality asks whether inclusion is reproducing a new exclusion.

Intelligence can distinguish Islam from Islamism. Rationality asks whether that distinction, though necessary, is sufficient.

Intelligence can write a careful memorandum saying that direct evidence is insufficient. Rationality asks whether the evidentiary model is too narrow to detect the problem.

This is the distinction Sweden now urgently needs.

An Open Society does not survive because its institutions are clever, kind, procedural, or well-intentioned. It survives because it preserves error-correction. It protects the Backspace Key: the human and institutional capacity to say:

This category no longer describes reality.

This policy produces the opposite of what it intends.

This symbol is not neutral.

This organization is legally compliant but socially closed.

This inherited moral language is hiding a power relation.

The central thesis is therefore simple:

Sweden’s problem in this field is not lack of intelligence. It is intelligence without sufficient error-correction.

The Backspace Key test for Swedish democracy

Sweden needs a new democratic test.

Not an anti-Muslim test.

Not a culture-war test.

Not a test of theological truth.

A freedom test.

Every publicly supported, publicly recognized, or publicly platformed religious or cultural organization should be assessed by questions such as these:

Can members leave without social punishment?

Can women reject religious or family norms without losing protection?

Can children grow into secular, liberal, reformist, or ex-religious lives?

Can inherited doctrine be criticized publicly?

Can blasphemy, apostasy, satire, and historical criticism be tolerated?

Can the organization accept Swedish law not merely tactically but morally when that law protects individual freedom?

Does public support strengthen individuals, or does it strengthen gatekeepers?

Are secular and dissenting members of the minority visible, or only religiously marked representatives?

Does the institution protect the person while allowing the symbol, doctrine, and organization to be criticized?

These are not anti-Muslim questions.

They are pro-person questions.

They ask Sweden not to optimize inherited error: not to make a closed structure more inclusive, more fundable, more consultable, or more administratively respectable before asking whether it leaves the person free.

They are questions designed to correct the aesthetic error, the conceptual error, and the administrative error at once:

Who is represented?

What is being protected?

Does the institution leave the person free?

The correct liberal principle is:

Protect the person. Criticize the symbol. Test the organization. Defend the right to doubt, leave, and begin again.

Sweden must become more rational, not less humane

Sweden’s task is not to become less compassionate.

It is to become more rational.

The answer to anti-rational drift is not suspicion of Muslims, hostility to religious people, bans on ordinary piety, or contempt for minority visibility. That would betray the Open Society from the other direction.

Muslims must be safe. Mosques must not be treated as enemy structures merely because they are mosques. Women with headscarves must not be harassed. Religious freedom must be defended.

The purpose of this distinction is not contempt.

It is rescue.

The secular Muslim, the liberal believer, the veiled woman, the unveiled woman, the ex-Muslim, the reformer, the child, and the dissenter all require the same protection: protection as persons, not as managed representatives of a doctrine or community.

But ideas, symbols, organizations, and norm systems must remain open to criticism.

The Open Society does not survive by kindness alone. It survives by corrigibility. It must be able to correct its own categories when inclusion excludes, when tolerance immunizes control, when neutrality protects power, when representation erases dissenters, and when democratic language conceals anti-liberal structure.

That means correcting the aesthetic error when representation becomes narrowing.

It means correcting the conceptual error when nuance becomes evasion.

It means correcting the administrative error when compliance is mistaken for openness.

Sweden’s problem is not lack of intelligence. Sweden has intelligence in abundance.

The question is whether Sweden still has the rational courage to use its intelligence against its own errors.

That is the test.

Can Sweden protect Muslims as persons while refusing to surrender criticism of Islamic symbols, institutions, organizations, and political theology?

Can Sweden include the veiled woman without erasing the woman who removed the veil?

Can Sweden distinguish Islam from Islamism without pretending that non-Islamist religious norms are always socially harmless?

Can Sweden assess religious organizations not only by visible legality but by their internal relation to doubt, exit, gender equality, and individual freedom?

Can Sweden keep its Backspace Key?

If it cannot, Sweden will remain intelligent — but increasingly anti-rational.

Its images will be kind.

Its concepts will be nuanced.

Its memoranda will be careful.

And still, the Backspace Key will be harder to reach.

An intelligent society that loses error-correction does not become harmless.

It becomes a beautifully administered cage.