Why You Agree With People You Know Are Wrong

Erich Fromm wrote that conformity is the loss of the self. The person stops being himself, takes on the personality the culture hands him, becomes exactly what everyone expects, and the fear of standing alone goes away. The price, Fromm said, is high. It is the self.

The most famous experiment ever run on that question found something else. Not the loss of the self. The self, fully present, watching itself agree.

What did Asch’s conformity experiment actually find?

Asch found that roughly a third of answers on the loaded trials went to the wrong majority, and about three quarters of subjects conformed at least once. But a quarter never conformed at all, only five per cent conformed every time, and the average subject was independent on most trials.

The setup at Swarthmore in the early 1950s was deliberately simple. A card with one line, a second card with three, and the question of which of the three matched. In the control condition, alone in the room, subjects were wrong less than one per cent of the time. The task was not hard. That was the point. Asch built it so that any wrong answer could only be group pressure, never honest confusion.

Seven confederates sat at the table with one real subject, and on twelve of the eighteen trials they all called a shorter line the match, out loud, in order, before the real subject spoke.

Did Asch’s subjects lose their own judgement?

Mostly, no. Asch interviewed the people who conformed and sorted them into three groups. A small number had actually seen the lines the way the majority called them. Some had come to doubt their own judgement. Most had done neither: they saw the correct line, knew it was correct, and said the wrong one anyway. Asch called this distortion of action, as distinct from distortion of perception or of judgement.

This is the finding that gets lost in the retelling, and there is a paper about exactly that loss. In 1990, Ronald Friend, Yvonne Rafferty and Dana Bramel went through ninety-nine social psychology textbooks published between 1953 and 1984 and tracked how Asch’s results were reported over time. They found a steady drift: authors increasingly emphasised conformity and increasingly downplayed independence. Asch had titled the monograph Studies of Independence and Conformity. Three decades of textbooks turned it into the conformity study.

Two later variations make the mechanism plain. When the subject wrote their answer privately instead of announcing it, conformity fell to around twelve per cent. When a single other person in the room broke ranks with the majority, it dropped close to zero. Deutsch and Gerard named the distinction in 1955: normative influence buys public agreement while leaving private belief untouched, informational influence actually changes what you think. Because the line task was unambiguous, there was almost nothing for informational influence to work on. What Asch measured was people managing an audience.

Is conformity a fixed part of human nature?

No. It varies by culture and by era, and the size of the variation is large enough that the famous 1950s figure should not be treated as a constant.

In 1996 Rod Bond and Peter Smith pulled together 133 Asch-style line studies from 17 countries. Two findings stand out. Conformity in American studies had been declining steadily since the 1950s. And conformity was higher in collectivist countries than individualist ones, across three separate measures of that dimension. Some replications found barely any conformity at all: Perrin and Spencer’s British study is the most-cited example.

Fromm was writing in 1941, watching entire nations organise themselves around a leader. The automaton he described was real and was doing terrible things. But he wrote it as a permanent psychology, a thing that happens to man. The data underneath turns out to describe something conditional: a thing that happens to some people, in some rooms, at a strength that can be turned up or down, and to a quarter of them not at all.

Why is agreeing while knowing better worse than simply conforming?

Because the person who conforms while keeping their private judgement does not get the relief Fromm described. They belong, and they know what it cost, and they carry the knowing.

Fromm’s automaton, if it worked as described, would at least be at peace. It has paid for its belonging with the whole self and no longer has to feel the transaction. There is no I left to notice the gap between what it said and what it saw. That is the version of conformity that most people picture when they picture losing themselves, and it is the version that would hurt least.

Distortion of action is the other thing. Every time you say the shorter line is the match, some part of you files it. And the file grows. This is the mechanism Fromm half saw and named wrong. He thought the danger was disappearing into the crowd. What the transcripts show is not disappearing at all. It is staying entirely yourself while your voice says otherwise, day after day, in ordinary rooms where the pressure is milder than Asch’s and the stakes feel too low to be worth the alone.

Nobody folds on the big things first. They fold on the line length. The thing that does not matter, so why make it strange. This is also why the sensation of being a different person in different rooms is so hard to shake, and why the feeling of always performing does not go away when you get better at the performance.

What does this have to do with Jung’s persona?

Jung’s persona is the face assembled to meet the group’s expectation, and he was careful that having one is not a fault. Nobody can meet the world with the raw whole of themselves. The problem he named was mistaking the persona for the self, letting the mask thicken until you could no longer tell from the inside where the performance ended.

What Asch caught is the moment before that mistake. The seam is still visible, and the person can still feel the difference between the line they see and the line they name. That mask has not fused. It is being held up by hand, at a cost they can still register. That register is worth something, and it is also the thing most likely to wear out.

This is the same structure that shows up when people leave a religion or a tight community and find the shape of the group still sitting inside them, and it sits close to what Sartre was pointing at with bad faith, though bad faith describes the lie told inward and this describes the one told outward while the inside stays clear.

What Asch never found

The question Fromm’s sentence forecloses, and the data reopens, is what happens to someone who holds the mask up by hand for long enough. Whether the file stays legible. Whether the quarter who never conformed were made of something else, or had simply not yet sat in the room where the price of standing alone was set high enough to move them.

Asch could raise conformity by changing the room. He never found the room that got everyone. He also never found the person the room could not touch. Both of those absences are still sitting in the data, and nobody has closed them since.

For the wider frame this sits inside, see what a fractured self actually means, or the rest of The Masks We Mistake for Ourselves.

FAQ

What did the Asch conformity experiment actually find?

About 36.8% of answers on critical trials matched the wrong majority; 75% of subjects conformed at least once; but 25% never conformed and the average subject was independent more often than not.

Did Asch’s subjects lose their own judgement?

No. Interviews showed most saw the correct line and knew it, then said the wrong answer while knowing it was wrong. They kept the private judgement and changed only the public one.

Is conformity a fixed human trait?

No. Bond & Smith’s 1996 meta-analysis of 133 studies across 17 countries found conformity declined in the US after the 1950s and was higher in collectivist cultures. It varies by era and culture.