[Paper Review] A sound and complete axiomatization for Dynamic Topological Logic

David Fernández–Duque|arXiv (Cornell University)|Jan 25, 2012

Logic, Reasoning, and Knowledge12 references3 citations

TL;DR

This paper presents a sound and complete axiomatization for Dynamic Topological Logic (DTL) by extending its language with a polyadic topological modality based on the tangled closure operator. The key contribution is proving completeness using a novel proof system that leverages simulations and quasimodels, with the polyadic modality essential for expressing simulability syntactically in topological models.

ABSTRACT

Dynamic Topological Logic (DTL) is a multimodal system for reasoning about dynamical systems. It is defined semantically and, as such, most of the work done in the field has been model-theoretic. In particular, the problem of finding a complete axiomatization for the full language of DTL over the class of all dynamical systems has proven to be quite elusive. Here we propose to enrich the language to include a polyadic topological modality, originally introduced by Dawar and Otto in a different context. We then provide a sound axiomatization for DTL over this extended language, and prove that it is complete. The polyadic modality is used in an essential way in our proof.

Motivation & Objective

• To address the longstanding challenge of finding a complete axiomatization for the full language of Dynamic Topological Logic (DTL) over all dynamical systems.
• To overcome the limitations of the monadic topological modality by introducing a polyadic variant based on the tangled closure operator.
• To enable syntactic expressiveness of simulations—previously only definable in topological semantics—by enriching the language with the polyadic modality.
• To establish completeness of the extended logic, denoted $\mathcal{DTL}^\ast$, through a proof system combining modal and temporal axioms with new polyadic continuity axioms.
• To lay the foundation for future syntactic treatments of structured dynamical systems, such as minimal systems and metric spaces, by extending proof-theoretic methods to richer classes.

Proposed method

• Extends the language of DTL with a polyadic topological modality $\Diamond\Gamma$, where $\Gamma$ is a finite set of formulas, interpreted via the tangled closure operator.
• Introduces a new polyadic continuity axiom: $\Diamond\Gamma \to \Diamond\Gamma$, which captures the behavior of the tangled modality under continuous functions.
• Adapts the quasimodel framework from prior work to accommodate the polyadic modality, ensuring serial and $\omega$-sensible transition relations.
• Constructs canonical quasimodels $\mathfrak{Q}(\Phi)$ from consistent sets of formulas, proving they are valid quasimodels via regularity and openness conditions.
• Uses limit models derived from quasimodels to show that any consistent formula is satisfiable in a topological model, establishing completeness.
• Employs the formula $\mathrm{Sim}(\mathfrak{w})$—definable only in the extended language—to syntactically capture the notion of simulation, which is crucial for the completeness argument.

Experimental results

Research questions

• RQ1 Can a complete axiomatization be provided for Dynamic Topological Logic over the full class of dynamical systems when the language is extended with a polyadic topological modality?
• RQ2 Is the polyadic tangled modality essential for expressing simulability in topological models, and can it be captured syntactically in the extended language?
• RQ3 Does the addition of the polyadic modality allow for a proof-theoretic treatment of DTL that matches the model-theoretic results previously obtained?
• RQ4 Can the completeness proof for the extended logic $\mathcal{DTL}^\ast$ be established using quasimodels and limit models, as in earlier work on $\mathcal{DTL}$?
• RQ5 Can the results be extended to richer classes of dynamical systems, such as minimal systems, by adding axioms for universal modality and density properties?

Key findings

• The paper provides a sound and complete axiomatization for $\mathcal{DTL}^\ast$, the extension of DTL with the polyadic tangled modality, proving that $\models\varphi$ implies $\vdash\varphi$.
• The polyadic modality is essential for expressing the simulation relation $\mathrm{Sim}(\mathfrak{w})$ syntactically in topological models, a property not expressible in the monadic language.
• Canonical quasimodels $\mathfrak{Q}(\varphi)$ are shown to be valid quasimodels, with $|\mathfrak{Q}(\varphi)|$ open and $\mapsto_{\mathfrak{Q}(\varphi)}$ serial and $\omega$-sensible, ensuring regularity.
• The completeness proof relies on constructing a limit model $\lim\mathfrak{Q}(\varphi)$ from a canonical quasimodel, in which any consistent formula $\varphi$ is satisfiable.
• The proof system successfully integrates axioms for the monadic $\mathsf{S4}$, the temporal ${[f]}$ modality, and new polyadic axioms for $\Diamond\Gamma$, forming a coherent and complete system.
• The results suggest that proof-theoretic methods can now be applied to structured classes of dynamical systems, such as minimal systems, by extending the logic with axioms like $\exists\Box p \to \forall{\langle f\rangle}p$.