arXiv Science⌕ Search

arXiv · 2611.01248

Modular proof analysis and cut elimination for the Reflection Calculi $\mathsf{RC}_1$ and $\mathsf{QRC}_1$

Abstract

The reflection calculus with one modality is denoted by $\mathsf{RC}_1$ and its quantified generalization is known as $\mathsf{QRC}_1$. These decidable calculi are of interest since they generate the strictly positive fragments of the (quantified) provability logic for theories like Peano Arithmetic and Heyting Arithmetic. We introduce analytic cut-free calculi ${\mathrm{GRC1}}$ and ${\mathrm{GQRC1}}$ for $\mathsf{RC}_1$ and $\mathsf{QRC}_1$ respectively. The new calculi derive sequents of the form $A\Longrightarrow B$ with $A,B$ strictly positive formulas. In particular, the new calculi do not use (multi)sets nor nested sequents. For both ${\mathrm{GRC1}}$ and ${\mathrm{GQRC1}}$ it is shown that the cut-rule is admissible. As a consequence, it is proven that the calculi are deductively equivalent to $\mathsf{RC}_1$ and $\mathsf{QRC}_1$ respectively. The cutfree systems are used to prove uniform interpolation for the two reflection calculi. In addition, ${\mathrm{GRC1}}$ is used to give a syntactic proof of Dashkov's Theorem to the extent that $\mathsf{RC}_1$ is the strictly positive fragment of the provability logic ${\mathsf{GL}}$. The paper also introduces non-analytic calculi that are conceptually easier but proof-theoretically behave badly. It is shown that even those systems allow direct cut-elimination: the benefits of the strict positivity outweigh the draw-backs of a non-analytic calculus in these cases.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Joost J. Joosten, Cosimo Perini Brogi. 2026-10-31. Modular proof analysis and cut elimination for the Reflection Calculi $\mathsf{RC}_1$ and $\mathsf{QRC}_1$. https://arxiv.org/abs/2611.01248

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Amenable C-sequences and ineffability in L

We give a relative consistency counterexample to the conjecture of Inamdar and Rinot that every regular uncountable cardinal carrying no amenable C-sequence is ineffable in the constructible universe L. The existence of a $Π^2_1$-indescribable cardinal is equiconsistent with the existence of a $Π^2_1$-indescribable cardinal carrying no amenable C-sequence and not ineffable in L. For every ground-model C-sequence on a weakly compact cardinal $κ$, generalized Hechler forcing adds a club D with $D\capβ\subseteq C_β$ at stationarily many inaccessible indices. We use this capture theorem in an iteration of length $κ^+$ after Johnstone's local lottery preparation.

math.LO↗

Omitting Types for Infinitary Real-Valued Logic: A Topological Perspective

For countable vocabularies, we extend the omitting types theorem for countable fragments of infinitary logic to Keisler's general real-valued structures by proving that the space of Henkin types of each such fragment is Čech-complete. The structures interpret function symbols arbitrarily and predicate symbols by bounded real-valued functions, while each formula carries a bound of its own. This extends the setting of Eagle's theorem from metric structures with a prescribed master metric to general structures whose vocabulary prescribes the sorts and bounds of the symbols. Following Keisler and Miller's treatment of logic without equality, we build models from closed terms. Morley's Baire category method then gives every satisfiable theory on a countable fragment a model with countable sorts omitting any countable family of types whose closed-term instances have empty interior in the space of Henkin types satisfying the theory.

math.LO↗

Friedman--Stanley cofinality without Borel completeness

Friedman and Stanley asked whether an invariant Borel class of countable structures must be Borel complete whenever every countable level of their hierarchy reduces to it. We construct a counterexample in one fixed countable language. The structural counterexample and its principal supporting results have been formalized in Lean~4 and it is available on github.

math.LO↗