Two Centuries of Controversy

Bayes never published his rule himself; a friend found it among his papers after he died in 1761. From 1774, Laplace independently rediscovered it, stated it in full generality, and put it to practical use. For nearly 200 years it was fiercely debated — many statisticians distrusted it because it relies on a starting guess, the prior — yet practitioners still used it to crack wartime codes and weigh scarce evidence. Cheap computing and big datasets later made those calculations routine, and the idea now sits at the heart of modern AI.

Bayesian reasoning explains why a positive test for a rare disease is often still a false positive. It is why scientific studies need replication. Every claim that should be believed with a percentage — not an absolute yes or no — is a Bayesian claim.