Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

The transcendence tower is consistent

Proved
TranscendenceTower.transcendence_tower_satisfiable

by mateoej12 · Sep 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

formalizing-contextlogicpropositional-logic

In Notes on Formalizing Context (1993), McCarthy observes that a formula ist(c,p)ist(c,p)ist(c,p) is itself always asserted within some further context, so that one has "something like ist(c0,ist(c,p))ist(c_0, ist(c,p))ist(c0​,ist(c,p))", and remarks: "The regress is infinite, but we will show that it is harmless." The promised demonstration does not appear in the paper. This theorem is that claim, made precise and proved.

Work in the stratified tower of propositional languages L0⊆L1⊆⋯L_0 \subseteq L_1 \subseteq \cdotsL0​⊆L1​⊆⋯, where Ln+1L_{n+1}Ln+1​ adjoins one new atom istn(φ)ist_n(\varphi)istn​(φ) for each φ∈Ln\varphi \in L_nφ∈Ln​. The transcendence schema Γ\GammaΓ is the set of all sentences

istn(φ)↔φ,n≥0, φ∈Ln,ist_n(\varphi) \leftrightarrow \varphi, \qquad n \ge 0,\ \varphi \in L_n,istn​(φ)↔φ,n≥0, φ∈Ln​,

which is the formal content of McCarthy's stipulation that unconditional belief in ppp at one level is equivalent to ististist of ppp one level up, iterated without bound. The theorem asserts that every instance of Γ\GammaΓ is true in the tower model MMM, so Γ\GammaΓ is satisfiable and therefore consistent.

The reason it goes through is exactly the reason Tarski's hierarchy of truth predicates avoids the Liar paradox: strict stratification makes self-reference syntactically unrepresentable. No formula of Ln+1L_{n+1}Ln+1​ can apply istnist_nistn​ to itself, so no analogue of the Liar sentence, and no analogue of Russell's paradox for ististist, can even be written down.

The result is deliberately modest in reach. The schema it proves consistent is the one McCarthy himself calls "pointless", in which each context faithfully copies the one below; each stage of the tower is in fact a definitional extension of the previous one. It says nothing about the case he identifies as interesting, in which a transcending context relaxes an assumption of the old, nor about the nonmonotonic specializes relation, lifting between non-nested contexts, or quantification over contexts as first-class terms.

Formalization note. MnM_nMn​ is an increasing chain of classical valuations with Mn+1(istn(φ))=Mn(φ)M_{n+1}(ist_n(\varphi)) = M_n(\varphi)Mn+1​(istn​(φ))=Mn​(φ); satisfaction is stated as eval (Mn+1) (schema φ)=trueeval\,(M_{n+1})\,(schema\ \varphi) = \mathrm{true}eval(Mn+1​)(schema φ)=true. The construction is elementary: no compactness theorem is required, only a direct limit of an increasing chain of valuations.

Preamble
import Definitions.Def_TranscendenceTowerCore

set_option autoImplicit false
Formal statement
namespace TranscendenceTower

theorem transcendence_tower_satisfiable :
    ∀ (n : Nat) (p : Lang n), Form.eval (M (n + 1)) (schema p) = true := by sorry

end TranscendenceTower
Source
Mateo Estrada, Harmless and Pointless: Consistency, rigidity, and the indeterminacy of context transfer in McCarthy's formalization of context, 7 September 2026, Section 2, Theorem 1. Makes precise the claim in John McCarthy, Notes on Formalizing Context, 1993, Section 1, that the regress of outer contexts 'is infinite, but ... harmless'.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me