sciencebriefs
all subjects →
11:03in productionCh. 1 · What it claims/ 11:03 · ceiling 15 min
Computing & AI

Church–Turing thesis

It isn’t true — it’s assumed. And everything in computing rests on that assumption.

The Church–Turing thesis is not a theorem but a foundational assumption — one that equates human-effective calculation with machine computation. It holds because no counterexample exists, not because it is proven. Its endurance is its strength — and its limitation.

Chapters & takeaways5
  1. 0:53
    What it claims

    The thesis equates 'effective calculability' with Turing computability — not as proof, but as definition.

  2. 2:36
    How it was done

    Church used lambda calculus; Turing built a tape-based machine — both proved the Entscheidungsproblem unsolvable.

  3. 4:06
    What holds up

    Turing machines, lambda calculus, and general recursion are provably equivalent — and the universal machine works.

  4. 5:36
    What does not

    It cannot be proven — because 'effective method' has no formal definition.

  5. 6:50
    Why it matters beyond the lab

    Turing’s model replaced Gödel’s arithmetic with intuitive mechanics — making computability tangible.

Worth your time?

Yes. Study the whole thing.

4.5/ 5
What works
  • defines the scope of algorithmic computation
  • unifies three independent formalisms
  • grounds all digital computation
What does not
  • proves anything
  • applies to physical systems
  • covers resource-bounded computation
  • defines intelligence
Study it if
  • computer scientists
  • logicians
  • AI researchers
Skip it if
  • engineers building faster chips
  • neuroscientists studying cognition
  • philosophers debating consciousness
The written brief2 min read

What the work claims

The Church–Turing thesis claims that every function calculable by an effective method is computable by a Turing machine — and vice versa. It identifies ‘effective calculability’ with three formally equivalent classes: Turing computable, λ-computable, and general recursive. It asserts that no computational procedure qualifies as an algorithm unless representable as a Turing machine.

How it was done

Church formalised effectively calculable functions as general recursive or λ-definable functions in 1935–36. Turing independently created the Turing machine model in 1936 — a theoretical device manipulating symbols on a tape — and used it to reformulate Gödel’s 1931 results. Church proved the Entscheidungsproblem undecidable using lambda calculus in his 1936 paper. Turing proved its unsolvability by showing the halting problem for Turing machines is undecidable, then introduced the universal Turing machine. Both showed their models were equivalent in computational power.

What holds up

The formal equivalence of Turing computability, λ-computability, and general recursiveness holds. Church and Turing both proved the Entscheidungsproblem unsolvable — Church via lambda calculus, Turing via the halting problem. The universal Turing machine’s capacity to simulate any algorithmic computation holds. The thesis remains unrefuted: no effective method has been shown to compute a function outside these classes.

What does not

It does not prove anything. It cannot be formally proven because ‘effective calculability’ is informally defined. It does not establish physical limits on computation. It does not cover non-algorithmic processes, real-time systems, or bounded-resource computation. It says nothing about efficiency, speed, memory, or implementation.

Why it matters beyond the lab

It anchors all digital computation. Every programming language, compiler, and CPU architecture assumes this thesis. It separates solvable from unsolvable problems in logic, mathematics, and computer science. It underpins modern cryptography, complexity theory, and the theoretical limits of AI — not as speculation, but as the accepted frame for asking ‘what can be computed?’

Is it worth your time

Yes. It defines the boundary of what can be computed — not as engineering limits, but as logical ones. It remains foundational for computing theory, programming language design, and AI theory. Its informal status means it cannot be bypassed by new hardware or algorithms; it must be confronted anew with each claim of ‘beyond-Turing’ computation.

Same field · Computing & AI4 of 32
Up next in Science

Dartmouth workshop

1956 · 9:52

The Dartmouth workshop didn’t build AI — it invented the idea that AI could be built.

9:52