Foundation Primitive Recognition Calculus Finite Certificate Transfer Sound Fait

The countable witness theorem

A certificate is a finite piece of data that stands for a larger claim. Think of a receipt: it records that a transaction happened, and it is small enough to hold in your hand. In the Recognition Science framework, certificates attach to recognition events, the discrete record of distinctions a system makes. The declaration soundFaithfulCover_countable_witnesses proves a precise consequence of this setup: if every genuine witness gets a certificate, a certificate is only issued to genuine witnesses, and a certificate identifies exactly one witness, then the total number of witnesses is countable. Countable means the witnesses can be lined up and matched one-to-one with the natural numbers 1, 2, 3, and so on.

The theorem is proved in a machine-checked library of formal theorems, where every step is verified by a computer. It rests on three conditions. The first, completeness, says every genuine witness carries a certificate. The second, soundness, says a certificate is never issued to a non-witness. The third, faithfulness, says a certificate determines the witness it certifies: two different witnesses can never share the same certificate. The theorem shows these three conditions together force the witness set to be countable, because the certificates themselves are countable and the map from witnesses to certificates is injective, meaning no two witnesses collide.

The power of the theorem shows in what it rules out. A completion that certifies everything, the trick that made a weaker notion vacuous, cannot be faithful as soon as the display type has two distinct points. The theorem also delivers a cardinality obstruction: no sound and faithful certificate cover by a countable certificate system exists for a predicate with uncountably many witnesses. The named instance is the real line. The real numbers are uncountable, so no finite certificate system can soundly and faithfully certify them. Finite distinction data cannot capture the continuum in this way.

What the theorem does not claim is just as important. It does not say recognition is impossible for uncountable domains. It says a particular kind of finite certificate, one that is sound and faithful, cannot cover them. Weaker certificate notions remain available, and the theorem says nothing about them. It also does not claim that the real line itself is countable, only that a sound and faithful finite certification of it is impossible. The theorem is a structural limit, not a statement about what exists.

/-- **Honest hinge.** A sound, faithful certificate cover by a countable
certificate system forces the witness set to be countable. -/
theorem soundFaithfulCover_countable_witnesses
    {N D Cert : Type} [Countable Cert] {C : Completion N D Cert} {P : D → Prop}
    (cover : SoundFaithfulCover C P) :
    Countable {d // P d} := by
  obtain ⟨f, hf⟩ := soundFaithfulCover_injects cover
  rw [← Cardinal.mk_le_aleph0_iff]
  exact le_trans (Cardinal.mk_le_of_injective hf) Cardinal.mk_le_aleph0

/-- **Cardinality obstruction.** No sound, faithful certificate cover by a
countable certificate system exists for a predicate with uncountably many
witnesses. This is the honest content the weak layer could not deliver. -/
theorem no_soundFaithfulCover_of_uncountable_witnesses
    {N D Cert : Type} [Countable Cert] {C : Completion N D Cert} {P : D → Prop}
    (hunc : ¬ Countable {d // P d}) (cover : SoundFaithfulCover C P) : False :=
  hunc (soundFaithfulCover_countable_witnesses cover)

/-- **Named instance: the real line.** No sound, faithful certificate cover by a
countable certificate system exists for the real line. Finite distinction data
cannot soundly and faithfully certify the continuum. -/
theorem no_sound_faithful_certification_of_reals
    {N Cert : Type} [Countable Cert] (C : Completion N ℝ Cert)
    (cover : SoundFaithfulCover C (fun _ : ℝ => True)) : False :=
  no_soundFaithfulCover_of_uncountable_witnesses reals_uncountable_witnesses cover

What this page does not claim

The theorem does not claim recognition is impossible for uncountable domains, only that a sound and faithful finite certification is impossible. The theorem does not claim the real line is countable. The theorem does not apply to certificate systems that drop the faithfulness condition.

Verify this page

Every tagged claim above names its theorem. To check one yourself rather than trust this page, elaborate the source module with Lean 4 and audit its axiom basis:

$ lake env lean IndisputableMonolith/Foundation/PrimitiveRecognitionCalculus/FiniteCertificateTransfer.lean
expected axiom basis: [propext, Classical.choice, Quot.sound] (the Lean kernel's standard three; no RS-specific axioms)

A page whose claims cannot be reproduced this way does not ship. In production, every anchor links to the exact declaration in the public source release, and this block carries the build receipt for the page itself.

Derived articles

This page is generated by a question-recursion engine: the questions its answers raise become the next pages. The current agenda, with open targets marked red: