When Everyone Knows That Everyone Knows

Pinker examines common knowledge, which game theorists define as a situation where everyone knows something, everyone knows that everyone knows it, and so on without limit. He argues that this recursive condition, rather than simple shared awareness, explains a large set of otherwise puzzling human behaviour.

The analytical core is that coordination problems require common knowledge to solve. Two people can only act together on an expectation if each is confident the other holds it and knows they hold it. Pinker uses this to explain why public and private knowledge behave differently: a fact that everyone privately knows can leave a situation unchanged, while the same fact said aloud in front of everyone transforms it. This produces his account of revolutions, where a regime survives while discontent is private and collapses when a public event makes the discontent common knowledge.

He applies the framework across several domains. Indirect speech, innuendo, veiled bribes and sexual overtures let people communicate while preventing common knowledge, which preserves the relationship if the offer is declined. Taboos and unspeakable topics function by keeping something out of common knowledge even when everyone believes it. Diplomatic ambiguity, apologies, virtue signalling, ceremonies, national symbols and public rituals all work by generating or blocking the recursive condition. He also connects it to money, authority and property, which depend on shared belief that everyone believes.

The framework is drawn from established work in game theory, particularly on coordination games, and from Pinker's earlier writing on the logic of indirect speech. The main objection is scope: the concept explains many things at the cost of being difficult to falsify in any particular case, and the psychological evidence that people actually reason recursively rather than using simpler heuristics is thinner than the theory requires.