In 1988, the Soviet Union looked permanent. Western analysts debated whether Gorbachev's reforms would liberalize the system or merely adapt it. The Berlin Wall stood. The satellites held. The regime had ruled for seventy years and showed no signs of imminent collapse.

Three years later, it was gone.

What happened wasn't primarily military or economic (though both mattered). Something stranger occurred: a system that had demanded public loyalty for generations discovered that almost no one actually believed. The expressed preferences of 280 million people turned out to be almost entirely false. When the costs of honesty dropped below the costs of performance, the whole edifice dissolved in months.

The analysts had been reading the public signal. The public signal was a collective fiction.

Timur Kuran calls this preference falsification: the act of misrepresenting your genuine beliefs under social pressure. It's not quite lying (you may not even know you're doing it). It's not quite conformity (you haven't actually changed your mind). It's something more subtle: a gap between public expression and private belief that becomes so normal you stop noticing it.

The phenomenon is universal. You've done it. Everyone has. You laughed at a joke you didn't find funny because the room was laughing. You nodded along with an opinion you privately doubted because disagreement felt costly. You expressed enthusiasm for a project you thought was misguided because your career depended on enthusiasm.

These small performances seem harmless. But they compound. Each falsified preference becomes a data point that others read as genuine. They adjust their own expressions accordingly. The public signal drifts further from private reality, and everyone responds to a consensus that doesn't exist.

This is the hall of mirrors: a social environment where everyone is performing for everyone else, and no one knows what anyone actually believes.

The consequences are profound. Reform becomes impossible because the problems are inadmissible.

Consider an institution where everyone privately knows something is broken. The strategy isn't working. The leader is ineffective. The culture is toxic. But each person looks around, sees others performing contentment, and concludes that they must be the outlier. So they perform contentment too. The next person sees this performance, draws the same conclusion, adjusts their own expression.

The dysfunction becomes undiscussable because no one knows it's widely recognized. Everyone is waiting for someone else to say what everyone thinks. No one does.

This is why bad regimes persist far longer than their actual support would suggest. The Soviet Union didn't have genuine mass support by 1985. It had mass compliance and mass falsification. The distinction is invisible until it suddenly isn't.

And this is why change, when it comes, arrives not gradually but as a cascade.

A preference cascade begins when something shifts the cost-benefit calculation for honesty. A prominent defection. A crack in enforcement. A moment when someone says the unsayable and doesn't get destroyed.

Others who share the private belief but have been hiding it now see that expression is possible. They speak. Their speaking emboldens others. The cascade accelerates. What looked like stable consensus evaporates because it was never consensus at all.

The Arab Spring. The MeToo movement. The sudden collapse of COVID restrictions after two years of apparent public support. In each case, the expressed preferences were hiding a different reality. When the dam broke, the flood revealed what had been there all along.

The pattern also works in darker directions. Norms against cruelty or bigotry can collapse just as quickly if the enforcement mechanism weakens. People who performed tolerance discover (or reveal) that the tolerance was performance. The preference cascade runs toward worse rather than better.

The terrifying implication: you cannot know, from public expression, what a society actually believes. The signal is unreliable. The consensus might be genuine. It might be collective theater. You can't tell the difference until it breaks.

Now we get to the part that's harder to face: you might be lying to yourself.

Kuran's analysis focuses on the gap between public and private belief. But there's another gap, less examined: between what you think you believe and what you actually believe.

Perform a position long enough and it colonizes your cognition. The social rewards for holding certain views (approval, belonging, professional safety) create genuine positive feelings when you express those views. The performance starts to feel like conviction. You lose access to whatever doubt or dissent might have existed underneath.

This is not hypocrisy in the simple sense. The hypocrite knows they're pretending. This is something more complete: a state where the pretense has become invisible even to the pretender.

Jung saw this. The persona (the social mask we wear) can fuse with the ego until we no longer know where the mask ends and the self begins. We become our performance. The authentic response, if it still exists, is buried too deep to reach.

There's a reason The Mask resonated. Stanley Ipkiss puts on an artifact that supposedly reveals his "true self," and what emerges is a manic, id-driven creature he barely recognizes. The film plays it for laughs, but the horror is real: the mask becomes the face. Wear it long enough and you can no longer take it off.

Fromm saw it too. His "automaton conformity" describes people who adopt the beliefs and desires of their social environment so thoroughly that they experience these adopted patterns as their own. The escape from freedom is complete: you no longer have to choose what to believe because you've outsourced the choosing.

I suspect this is pervasive on both sides of the present divide.

On the left: the professional-class progressive who has expressed the correct views on race, gender, climate, and capitalism for so long that they couldn't tell you which views are genuine conviction and which are credential maintenance. The DEI statement, the land acknowledgment, the pronoun declaration: these began as performances for some and have become, through repetition, indistinguishable from belief. To question any of it would be to question themselves. So they don't question.

When Elon Musk acquired Twitter, a preference cascade became visible. People began saying things publicly that had apparently been inadmissible. The vehemence of the reaction from progressives revealed something: if these were fringe views, they'd be ignorable. The fury suggested recognition that the expressed consensus was thinner than it had appeared.

On the right: the Republican official who has genuflected toward Trump for years, defending the indefensible, rationalizing the irrational. Did it begin as pure calculation? Perhaps. But calculation performed long enough becomes conviction. The official may no longer be able to access whatever reservations once existed. To admit doubt now would be to admit that years of public statements were false. The psyche protects itself from that recognition.

Liz Cheney said publicly what many Republicans say privately. The interesting question is how many have falsified their private preferences too: convinced themselves that they believe what they've been performing, because the alternative is too destabilizing.

This creates a particular kind of trap. If you don't know what you actually believe (because social pressure and repetition have obscured it), you can't participate authentically in discourse. You can only perform. And your performance becomes another data point that others read and respond to, compounding the falsification.

The hall of mirrors gets worse: not only do you not know what others believe; you don't know what you believe. Everyone is performing for everyone else, including themselves.

This might explain why political discourse feels so hollow: the sensation that people are reciting positions rather than reasoning toward them, that the arguments are post-hoc justifications for tribal commitments, that changing anyone's mind (including your own) has become nearly impossible.

The positions aren't really positions. They're identity markers, performed and internalized until the performance is all that remains.

I don't have a clean solution. Preference falsification is a rational response to social costs, and those costs are real. Honesty can destroy careers, relationships, reputations. The incentives are what they are.

But I think there's value in simply naming the phenomenon. Knowing that public opinion is unreliable. Knowing that consensus might be theater. Knowing that your own beliefs might be more performed than you'd like to admit.

And perhaps: being suspicious of your strongest convictions. The beliefs you hold most loudly, most publicly, most tribally. Are those the ones you've examined most carefully? Or are they the ones most shaped by social reward and repetition?

The genuine article tends to be quieter. Real conviction doesn't need constant performance, bumper stickers, signs in your window or yard. It can tolerate doubt, ambiguity, revision. It doesn't depend on the room's approval.

When you find yourself certain, righteous, eager to signal: that's worth examining. You might be expressing authentic belief. You might be performing for an audience (including yourself). You might not know the difference anymore.

That uncertainty is uncomfortable. But it might be the beginning of honesty.