Universal Proof Theory: Semi-analytic Rules and Craig Interpolation

Abstract

We provide a general and syntactically-defined family of sequent calculi, called semi-analytic, to formalize the informal notion of a "nice" sequent calculus. We show that any sufficiently strong (multimodal) substructural logic with a semi-analytic sequent calculus enjoys the Craig Interpolation Property, CIP. As a positive application, our theorem provides a uniform and modular method to prove the CIP for several multimodal substructural logics, including many fragments and variants of linear logic. More interestingly, on the negative side, it employs the lack of the CIP in almost all substructural, superintuitionistic and modal logics to provide a formal proof for the well-known intuition that almost all logics do not have a "nice" sequent calculus. More precisely, we show that many substructural logics including UL-, MTL, R, Ln (for n ≥ 3), Gn (for n ≥ 4), and almost all extensions of IMTL, L, BL, RMe, IPC, S4, and Grz, (except for at most 1, 1, 3, 8, 7, 37, and 6 of them, respectively) do not have a semi-analytic calculus. Keywords. Craig interpolation, sequent calculi, substructural logics, linear logics, subexponential modalities

0

Turn this paper into a lesson

ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…