Failure used to be quarantined. In medieval European guilds, a craftsman's botched work was hidden, destroyed, or blamed on apprentices — because mistakes were shameful, not informative. The 14th-century surgeon Guy de Chauliac was one of the first in his field to insist that surgical errors be recorded and studied systematically, arguing that a failure observed carefully is worth more than a hundred successes assumed. What he stumbled onto is something modern thinkers rarely credit to medieval medicine: that the act of examining what went wrong forces you to make your own reasoning explicit — to say, out loud or on paper, what you expected to happen and why. That gap between expectation and outcome is not just data. It is the precise location of your hidden assumption. Most creative and analytical work today does the opposite — we iterate forward from what half-worked, keeping the assumption intact, which is why many solutions slowly improve without ever actually solving the original problem. The discipline de Chauliac was gesturing at is this: before you revise the solution, write down the belief the failed attempt was built on. That belief is your real subject matter.

What did you last decide to revise or iterate without first writing down why the previous version failed?

Drawing from History of Medicine / Empirical Method — Guy de Chauliac (14th-century surgeon, author of Chirurgia Magna)