There is a point at which a community can become so accomplished at explaining the hostility directed against it that it gradually loses the ability to distinguish hostility from criticism, criticism from consequence, and consequence from evidence. The danger does not begin with lies. It begins with something more respectable: concern, vigilance, memory, solidarity, the desire to protect. Yet when these are joined to an interpretive system that already knows in advance what every event means, protection can turn into a machine for preserving an image of the world rather than recognizing the world itself.

Oscar Wilde gave Dorian Gray an impossible privilege. His face remained untouched while the consequences of his life accumulated somewhere else, on a portrait concealed from ordinary sight. Politics can produce a similar separation between self-image and consequence. A state, a movement, a media ecosystem, even an entire community may continue to see itself as defensive, responsible, endangered and morally serious while the effects of its own conduct accumulate elsewhere: in the lives of other people, in diplomatic isolation, in public distrust, in radicalization, in language, and eventually in the very hostility that the system then cites as proof that its original picture was correct.

Consider a real attack on a synagogue in the United States. The event deserves factual description, investigation, security analysis and moral condemnation. We should want to know who attacked, why, with what ideological background, what institutional failures made the attack possible, and what might reduce the danger of repetition. Yet watch what can happen after the facts have been established. The event enters an interpretive machine whose categories existed before the event itself.

One attack becomes evidence of a climate; the climate becomes evidence of a wave; the wave becomes evidence of a civilizational hostility already believed to be everywhere. Newspapers cite organizations, organizations cite commentators, politicians respond to the organizations, and newspapers then report the political response as further evidence of the seriousness of the original phenomenon. Soon dozens or hundreds of texts appear to confirm one another, although many may be reflections of the same interpretive structure.

This is where the hall of mirrors begins. The number of reactions is mistaken for the number of independent observations, and repetition acquires the authority of verification. An event may be real, the fear may be justified, and the danger may be serious, while the interpretive amplification surrounding it still becomes epistemically defective. One fact reflected a hundred times remains one fact; a hundred reactions to it do not automatically become a hundred separate pieces of evidence about the world.

The problem becomes more serious when the same structure is applied beyond direct antisemitic violence. Criticism of Israeli policy becomes evidence of hostility to Israel, hostility to Israel becomes evidence of antisemitism, and antisemitism then confirms the claim that earlier criticism was never really about policy at all. Protest confirms the diagnosis, diplomatic pressure confirms the diagnosis, public anger confirms the diagnosis, and growing isolation confirms the diagnosis. Eventually the system becomes capable of explaining every possible response without allowing any response to modify the model that produced the explanation.

At this point we should ask a question far more uncomfortable than whether Israel has enemies. Of course it has enemies, and Jewish communities also face genuine antisemitic threats. The harder question is this: what would have to happen for us to admit that our picture of the world is wrong? If every criticism proves hostility, every isolation proves persecution, every protest proves hatred, and every failure proves the need for more of the same, then there is no longer any possible experience capable of correcting the system.

What remains is not realism but an unfalsifiable narrative. A world that can only confirm us is not a world we are observing; it is a world we have already interpreted before it arrives.

Until this point, however, we may still be making an assumption that is too generous. We assume that the system wants correction. We assume that organizations, political movements, advocacy networks and media institutions ultimately want the problem they describe to become smaller, better understood or eventually resolved.

But what if that is not always structurally true?

This is not an accusation that organizations secretly desire antisemitism, conflict or danger. Such a claim would be both crude and impossible to establish in most cases. The problem is subtler. A threat can be real and still become institutionally useful. Fear can be justified and still sustain mobilization. A community can genuinely require protection while building institutions whose authority, funding, political relevance or identity increasingly depend upon the permanence of danger.

An organization can sincerely want security while operationally needing continued mobilization. A media ecosystem can sincerely want understanding while depending economically and politically on attention, urgency and moral clarity. An advocacy network can sincerely want to protect Israel while gradually becoming dependent on an interpretive structure in which its own mediation remains indispensable.

This creates a distinction we rarely make clearly enough. There is the declared goal, there is the operational goal, and there is the function through which an institution reproduces itself. They may coincide, but there is no reason to assume that they always do.

A system can therefore be epistemically wrong and functionally successful at the same time.

That possibility changes almost everything. If an institution merely misunderstands reality, better evidence may correct it. But if correcting the misunderstanding would also weaken its coalition, its fundraising narrative, its political access, its authority, its audience or its reason for remaining indispensable, then correction becomes more than intellectually uncomfortable. It becomes institutionally expensive.

At that point, the critic does not merely challenge a belief. The critic can threaten an arrangement.

This helps explain why criticism so often becomes information about the critic before it becomes information about reality. Instead of asking whether an argument is true, we ask who made it. Instead of examining the warning, we classify the person who issued it. “Anti-Israel,” “hostile,” “extremist,” “delegitimizing,” “foreign-influenced” — such categories may sometimes describe real phenomena, but when they function as preliminary filters through which inconvenient evidence must pass, they stop protecting knowledge and begin protecting an institutional equilibrium.

Inconvenient facts no longer need to be refuted. They merely need to arrive from the wrong mouth.

This is not uniquely Jewish or Israeli. Political parties do it, corporations do it, military institutions do it, ideological movements do it, and entire states have done it. The particular danger in a highly developed pro-Israel advocacy and media ecosystem is not that it is uniquely manipulative, but that political effectiveness can make the epistemic problem harder to see.

A support system can win while becoming progressively worse at learning. It can raise money, defeat candidates, secure legislation, dominate the vocabulary of debate and marginalize opponents while weakening the mechanisms through which it would discover that circumstances have changed, that its inherited model of Israel’s interests has become inadequate, or that its own conduct is producing part of the reaction it interprets as external hostility.

Political success is therefore not evidence of epistemic success. Sometimes the mechanism producing victory also suppresses the signals that would reveal strategic error.

The question then becomes more disturbing. What happens to an advocacy organization when the threat it was built to confront diminishes? What happens to a political movement when its enemy becomes less useful? What happens to a media ecosystem when proportion is less rewarding than alarm? What happens to institutions whose authority partly depends upon remaining necessary?

Again, none of this requires cynical individuals sitting in a room deciding to preserve a conflict. Systems can acquire dependencies that nobody consciously designed. A political ecosystem may sincerely seek protection while reproducing the very conditions under which permanent protection remains necessary.

Perhaps the most dangerous political systems are therefore not those that fail to solve their problems. They are those that gradually learn how to live from them.

This makes the ethical question unavoidable. What does it mean to support Israel if support gradually becomes an obligation to explain away conduct we would condemn in anyone else? There is a profound difference between defending Israel and defending every inherited definition of Israel’s interests, just as there is a difference between protecting Jews from hatred and protecting organizations speaking in their name from criticism.

To defend someone does not mean defending everything they do. A serious friend can say: I will defend your right to exist, your right to security and your right not to be hated, but I will not defend cruelty merely because it is yours. If support removes responsibility, it ceases to be solidarity and becomes the manufacture of innocence.

There is something deeply infantilizing in the demand that loyalty should exempt a state from the standards applied to others. A moral subject is not someone whose actions must always be excused; it is someone capable of being held responsible. If we insist that Israel must remain permanently innocent regardless of what it does, we do not treat Israel as a serious political and moral subject. We turn it into an exception protected from adulthood.

This is where Dorian Gray returns with greater force. The portrait is no longer simply a false self-image. It may become part of an institutional arrangement that needs the image to remain convincing. If public trust declines, the world is hostile. If allies distance themselves, they are weak. If universities protest, they are radicalized. If journalists investigate, they are biased. If critics multiply, antisemitism is spreading. The portrait survives because the world has been assigned the task of carrying everything the portrait refuses to show.

Wilde’s Dorian Gray at least had one terrible advantage. Somewhere, hidden away, there remained a portrait capable of registering what the face would not reveal. Our danger is greater when the portrait itself becomes part of the media system, when it writes the editorials, selects the categories, explains the reactions, mobilizes the audience and assures us every morning that nothing essential has changed.

The hardest question is therefore not whether the portrait lies. It is whether we have built institutions that need the portrait to remain convincing.

Once a system depends upon the image it produces, recognition becomes dangerous not only psychologically but institutionally. Reality is no longer merely difficult to see; it becomes inconvenient to acknowledge.

The opposite of illusion is not optimism. It is recognition. Without recognition, security becomes choreography, strategy becomes repetition, solidarity becomes exemption, and moral certainty becomes a technique for avoiding what is directly in front of us.

Dorian Gray was not destroyed because the portrait was ugly. He was destroyed because truth had become unbearable to the life built around hiding it.

Yochanan Schimmelpfennig

P.S.

As this essay was being completed, Itamar Ben-Gvir publicly advocated killing “30 to 40” people in Gaza every night, explicitly extending the category beyond those posing an immediate threat and describing some people there as unworthy of life. Associated Press also noted how little attention the statement received in Israel’s domestic media.

I do not add this as another outrage competing for attention. I add it because it raises the question at the center of this essay with almost unbearable clarity: what happens when a system becomes extraordinarily sensitive to the language, intentions and networks of others, while losing the same sensitivity toward itself?

Perhaps the problem is no longer that the portrait is flattering.

Perhaps the problem is that everyone has learned where not to look.