Fully Evaluated Left-Sequential Logics

📅 2024-03-21
🏛️ arXiv.org
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This paper addresses the lack of a formal, axiomatic foundation for “fully evaluated left-ordered logics” (FELs)—a family of binary-valued logics modeling short-circuit evaluation order—by systematically constructing and fully axiomatizing four variants: free, memory, conditional, and static FELs, along with three-valued extensions (incorporating an undefined value *U*) for the first three. Method: It introduces a unified evaluation-tree semantics and develops independent, complete axiom systems that rigorously characterize the evaluation behavior of left-ordered expressions. Contributions/Results: (1) It establishes a logical strength hierarchy among the four FELs and their precise semantic correspondences; (2) it proves that the three-valued extensions of the free, memory, and conditional FELs are all logically equivalent to Bochvar’s strict three-valued logic; (3) it provides sound and complete axiomatizations for all closed left-ordered propositional formulas, delivering minimal independent axiom sets—including *U*—for each three-valued system. These results fill a foundational theoretical gap in the formal treatment of left-ordered logics.

Technology Category

Application Category

📝 Abstract
We consider a family of two-valued"fully evaluated left-sequential logics"(FELs), of which Free FEL (defined by Staudt in 2012) is most distinguishing (weakest) and immune to atomic side effects. Next is Memorising FEL, in which evaluations of subexpressions are memorised. The following stronger logic is Conditional FEL (inspired by Guzm'an and Squier's Conditional logic, 1990). The strongest FEL is static FEL, a sequential version of propositional logic. We use evaluation trees as a simple, intuitive semantics and provide complete axiomatisations for closed terms (left-sequential propositional expressions). For each FEL except Static FEL, we also define its three-valued version, with a constant U for"undefinedness"and again provide complete, independent axiomatisations, each one containing two additional axioms for U on top of the axiomatisations of the two-valued case. In this setting, the strongest FEL is equivalent to Bochvar's strict logic.
Problem

Research questions and friction points this paper is trying to address.

Define and analyze two-valued left-sequential logics
Develop complete axiomatisations for closed terms
Introduce three-valued versions with undefinedness constant
Innovation

Methods, ideas, or system contributions that make the work stand out.

Two-valued fully evaluated logics
Evaluation trees for semantics
Three-valued versions with undefinedness
🔎 Similar Papers
No similar papers found.
Alban Ponse
Alban Ponse
Guest Researcher, Informatics Institute, University of Amsterdam
Theoretical Computer ScienceProcess AlgebraProgram and Thread AlgebraShort-Circuit Logic
D
Daan J. C. Staudt
Informatics Institute, University of Amsterdam, Science Park 900, 1098 XH, Amsterdam, The Netherlands