PERSON
Emil Post
The one-armed logician who, working three guarded hours a day in a Harlem apartment, glimpsed the foundations of computing a decade before Turing—proved that any system of rules generates problems no rule can solve, and left us the clearest possible map of the wall every AI will ever run against.
Emil Post is the ghost that haunts the history of computing—the man who saw it first and could not make the world listen in time. In the academic year 1920 to 1921, holding a fellowship at Princeton, a twenty-three-year-old Post tried to take the three volumes of Whitehead and Russell’s
Principia Mathematica and prove mechanically that it could settle every question put to it. He failed at that goal in the most productive way anyone has ever failed: by trying to reduce all of mathematics to simple string-rewriting rules, he found instead that any such reduction generates problems no rule can decide. He had, in embryo, the incompleteness theorem that would make Gödel famous in 1931 and the theory of computation that would make
Turing famous in 1936. He published almost none of it until it was far too late. His production systems—the formalism he invented