The Book of Why: The New Science of Cause and Effect

This is the same law that makes insurance companies so profitable, despite the uncertainties in human affairs.

How can machines (and people) represent causal knowledge in a way that would enable them to access the necessary information swiftly, answer questions correctly,

“Strictly speaking, for ‘due to’ read ‘associated with.’ ” This set the pattern for generations of scientists after him. They would think “due to” and say “associated with.”

of the data increases, leaving a single objective conclusion in the end.

The recognition that causation is not reducible to probabilities has been very hard-won,

selectively break the rules of logic.

Counterfactual learners, on the top rung, can imagine worlds that do not exist and infer reasons for observed phenomena.

Computers are not good at breaking rules,

This is perhaps the most important role of Bayes’s rule in statistics: we can estimate the conditional probability directly in one direction, for which our judgment is more reliable, and use mathematics to derive the conditional probability in the other direction, for which our judgment is rather hazy.