Alan Turing
The Question That Built the Modern World
Before anyone had built a computer, Alan Turing had already worked out the theoretical limits of what every computer that would ever exist could possibly do.
The Idea
In 1936, Turing published a paper solving a problem in mathematical logic that most people had never heard of — and in doing so, accidentally invented the blueprint for every programmable machine on the planet. The problem was Hilbert's Entscheidungsproblem: could there be a mechanical procedure, a pure algorithm, that could determine whether any given mathematical statement was true or false? Turing's answer was no. But the way he proved it was the revelation. To reason about what computation could and couldn't do, he first had to define what computation was. He described an imaginary machine — now called a Turing machine — that reads symbols from a tape, follows rules, and writes new symbols. Simple to describe, and yet theoretically capable of performing any calculation that any computer ever built can perform. The key insight hiding inside this abstraction is universality: one machine, given the right instructions, can simulate any other machine. This is why your phone can run a spreadsheet, a game, and a voice assistant — it is, at its core, a universal Turing machine. What looks like a hardware story is actually a philosophy-of-mind story. Turing wasn't asking 'how do we build faster adding machines?' He was asking something stranger and deeper: what does it mean to follow a rule? What is the difference between a mind and a mechanism? He never fully separated those two questions.
In the World
The most vivid demonstration of Turing's ideas in action isn't Silicon Valley — it's Bletchley Park, 1941. The German military was encrypting its communications with the Enigma machine, and the volume of intercepted messages was overwhelming human codebreakers. Turing, working with a small team in Hut 8, built the Bombe: an electromechanical device that could systematically test millions of possible Enigma settings, searching for the configuration that would turn nonsense back into German. It wasn't a Turing machine in the strict theoretical sense, but it was a direct child of the same thinking — the idea that reasoning could be mechanised, that pattern-finding could be offloaded from a human mind to a process running on hardware. Historians estimate that Turing's work at Bletchley shortened the war in Europe by somewhere between two and four years. Churchill reportedly called the codebreakers his 'geese that laid the golden eggs and never cackled.' What makes this more than a war story is what Turing himself understood: the Bombe was not magic. It worked because he had thought carefully, years before, about what machines could and couldn't do in principle. The theory came first. The machine followed. That sequence — abstract reasoning preceding practical engineering — is exactly how the most important computing breakthroughs tend to happen, and it's a pattern Turing essentially established as the discipline's founding move.
Why It Matters
There's a version of Turing's story that reduces him to a tragic biography — the genius who was prosecuted by the state he had helped save. That version is true, and it matters. But it can accidentally obscure the stranger, more useful thing: that his ideas are still the water we swim in. Every time you hear someone ask whether an AI is 'really' thinking, or whether a language model 'understands' anything, they are circling questions Turing posed in 1950 in his paper on the imitation game. He didn't answer them — he deliberately left them open, as provocations. The practical upshot for how you might think differently: the next time someone describes a piece of software as doing something almost human, it's worth pausing on that word 'almost.' Turing's framework suggests the gap between 'following rules very quickly' and 'thinking' is not obviously a matter of speed or scale. It might be a matter of kind. Whether you find that troubling or exhilarating probably tells you something about how you relate to the machines accumulating around you.
A Question to Ponder
If a machine can perfectly simulate any thought process you describe to it, at what point — if any — does the simulation become the real thing?
Get a new one of these every morning.
Start learning with Thinkable