The reason a norm survives after everyone stops believing it

A practice persists that almost nobody defends in private. Asked individually, participants describe it as outdated. In public they comply, and — more importantly — they correct others who do not comply, which supplies everyone else with fresh evidence that the norm is still believed in.

What everyone sees

Persistence is read as evidence of belief, on the assumption that people uphold what they endorse. That inference requires public behaviour to reveal private conviction. Where the cost of visible deviation is high, public behaviour reveals the cost, and conviction becomes unobservable in exactly the situations where it has already collapsed.

What is actually happening

Bicchieri’s analysis of social norms separates personal beliefs from expectations about others, holding that compliance depends on believing others expect it and will sanction deviation, which can persist even when nobody approves of the content. Centola, Willer and Macy modelled the sharper case: participants who privately reject a norm may enforce it more zealously than believers, in order to demonstrate that their own compliance is sincere, and this false enforcement makes the norm appear more widely supported than it is. The result is a self-sustaining arrangement in which the strongest evidence of support is produced by those who least believe it.

Why it stays hidden

The mechanism hides because everyone has access only to public conduct. Each participant knows their own doubt and observes universal compliance, and concludes that they are unusual. The doubt is common and invisible; the compliance is universal and visible, and the visible half is the only thing anyone can use to estimate what others think.