A brief history of algebraic logic from neat embeddings to rainbow constructions

TL;DR

This paper surveys the development of algebraic logic from classical results on neat embeddings to modern rainbow constructions, highlighting connections with graph theory, model theory, and finite combinatorics, and introducing new approaches to longstanding problems.

Contribution

It provides a comprehensive survey of algebraic logic, including new methods using rainbow constructions to address classes of cylindric-like algebras with neat embedding properties.

Findings

Generalizes results on non-atom canonicity and non-finite axiomatizability

Extends results to relativized clique guarded representations

Connects algebraic logic with graph theory and combinatorics

Abstract

We take a long magical tour in algebraic logic, starting from classical results on neat embeddings due to Henkin, Monk and Tarski, all the way to recent results in algebraic logic using so--called rainbow constructions invented by Hirsch and Hodkinson. Highlighting the connections with graph theory, model theory, and finite combinatorics, this article aspires to present topics of broad interest in a way that is hopefully accessible to a large audience. The paper has a survey character but it contains new approaches to old ones. We aspire to make our survey fairly comprehensive, at least in so far as Tarskian algebraic logic, specifically, the theory of cylindric algebras, is concerned. Other topics, such as abstract algebraic logic, modal logic and the so--called (central) finitizability problem in algebraic logic will be dealt with; the last in some detail. Rainbow constructions are used to solve problems adressing classes of cylindric--like algebras consisting of algebras having a neat embedding property. The hitherto obtained results generalize seminal results of Hirsch and Hodkinson on non--atom canonicity, non--first order definabiity and non--finite axiomatizability, proved for classes of representable cylindric algebras of finite dimension$>2$. We show that such results remain valid for cylindric algebras possesing relativized {\it clique guarded} representations that are {\it only locally} well behaved. The paper is written in a way that makes it accessible to non--specialists curious about the state of the art in Tarskian algebraic logic. Reaching the boundaries of current research, the paper also aspires to be informative to the practitioner, and even more, stimulates her/him to carry on further research in main stream algebraic logic.