Mind, Matter, and Meaning

Alan Turing

Alan Turing (1912–1954) was a British mathematician and logician, the founder of theoretical computer science and one of the founders of artificial intelligence. In a single 1936 paper he gave the first precise mathematical account of computation; during the Second World War he led the codebreaking work at Bletchley Park that broke the German Enigma cipher; and in 1950 he reframed the question of machine intelligence in terms that have governed the debate ever since. Prosecuted in 1952 for homosexuality, then illegal in Britain, and stripped of his security clearance, he died in 1954 of cyanide poisoning, generally judged a suicide; he received a royal pardon in 2013. His ideas sit at the headwaters of nearly every question this site raises about machines and mind.

Major Works

  • “On Computable Numbers, with an Application to the Entscheidungsproblem” (Proceedings of the London Mathematical Society, 1936) — the Turing machine, computability, and the universal machine
  • “Computing Machinery and Intelligence” (Mind, 1950) — the imitation game (the “Turing test”)

Key Contributions

The Turing machine and computability. Turing analyzed computation into its barest elements — a tape, a finite set of states, rules for reading and writing symbols — and used the resulting abstract machine to make “effectively calculable” mathematically precise. The analysis underwrites the Church–Turing thesis: the functions computable by an idealized mechanical procedure are exactly the Turing-computable ones. It is a claim about which functions can be mechanically computed — not, on its face, a claim about minds. See Computation and the Church-Turing Thesis.

The universal machine. Turing showed that a single machine, given a description of any other, can simulate it. The universal Turing machine is the theoretical ancestor of the stored-program computer — one device that runs any program — and the formal root of the idea that the same computation can be realized on many different physical substrates.

The imitation game (the Turing test). Facing the question “Can machines think?”, Turing judged it “too meaningless to deserve discussion” and proposed replacing it with an operational substitute: a human judge, communicating by text alone with a hidden human and a hidden machine, tries to tell which is which. A machine that reliably passes — that cannot be distinguished from the human by conversation — should, Turing argued, be credited with thinking. The move is deliberately behavioral: it sets aside what is going on inside in favor of what the system can do. Crucially, Turing explicitly bracketed consciousness: replying to “the argument from consciousness,” he declined to settle whether the machine has inner experience, treating that as a separate question from whether it thinks. He also anticipated and answered a battery of objections — the theological objection, the mathematical objection from Gödel, and “Lady Lovelace’s objection” that a machine can only do what it is told.

Relation to the Later Debate

Turing’s framing set the terms for everything that followed. Functionalism and the computational theory of mind inherit his universal-machine picture of substrate-independent organization. Searle’s Chinese Room is, in his own words, a kind of dual of the imitation game — a case designed to show that passing a behavioral language test (manipulating symbols by rule) is not sufficient for understanding. Contemporary work on large language models has revived the test directly: systems now routinely produce human-indistinguishable text, which sharpens rather than settles the question Turing set aside.

My View

Turing is indispensable, and the book engages him at two distinct seams — taking care not to blame him for conflations made in his name.

First, the Church–Turing thesis is not computationalism. Turing told us which functions a machine can compute; he did not thereby tell us that minds are such computations, or that running the right computation suffices for a mind. Treating his thesis as if it answered the metaphysical question is one of the cleaner category errors in the field, and the book separates the two explicitly (see Computation and the Church-Turing Thesis and Simulation and Realization). Turing the logician proved something true and precise; the overreach belongs to his readers.

Second, the Turing test sets exactly the bar the book rejects. The imitation game measures whether a system sounds human, and the whole argument of the book’s final part is that sounding human is the wrong test — what matters is whether a system stands in the world-involving relations meaning and consciousness actually require. I give Turing his due here: in 1950, with behaviorism ascendant and no science of the inner life in sight, an operational test was a reasonable and clarifying move, and he was scrupulous enough to bracket consciousness rather than pretend the test settled it. The error is in his successors who forget that bracketing and wield a passed test as proof of an inner life. The test’s own author would not have made that move. And the reflex the test formalizes — read a mind off fluent output — is precisely the ELIZA effect the book opens with: the projection machine running at full throttle. Turing named the modern question; the book’s answer is that he asked it in the one register guaranteed to miss what we most want to know.