The paper investigates a particular logical system called the "minimal well-determined logic," which works with just two basic operations: conjunction (the "and" connective) and implication (the "if...then" connective). What makes this logic unusual is that its proof system deliberately leaves out a rule called modus ponens, which is the standard rule allowing you to conclude B from "if A then B" together with A. The authors construct a formal calculus for this system and set out to understand exactly which statements count as universally true (tautologies) within it.
The central contribution is a new kind of semantics, meaning a mathematical framework that gives precise meaning to the logical formulas. The authors interpret the connectives using algebraic structures called lower semilattices with a greatest element, which are essentially sets where any two elements have a well-defined "greatest lower bound." The key innovation is that implication is not interpreted as a total function (one that always produces a value) but as a partial function, defined only in certain cases based on the ordering structure of the semilattice. This is a meaningful departure from classical approaches, where logical connectives are always fully defined. The authors prove soundness (every provable formula is true under the semantics) and completeness (every formula true under the semantics is provable), establishing that the proof system and the semantics perfectly match each other.
As a practical payoff, the paper shows that deciding whether a given formula is a tautology of this logic can be done efficiently, specifically in polynomial time, and the authors provide a concrete algorithm for doing so. This is notable because many non-classical logics have decision problems that are computationally hard or even undecidable. The semantic framework developed here also opens doors for further study, including whether the logic can always be characterized by finite models and how the minimal system relates to its various extensions.