A way to interpret Łukasiewicz logic and basic logic (Q1005968)

A long-standing problem in the study of fuzzy logics is to understand and justify the motivation behind them. Two traditional approaches have tried to tackle this problem. A semantical approach aims to develop a framework for dealing with statements which are not necessarily fully true or fully false and modifies classical propositional logic accordingly. The idea is to associate with propositions many truth values, rather than restricting the values to 0 and 1: all the elements from the real unit interval are permitted. Several connectives can be used to combine many-valued propositions generalizing the classical case. However, the choice of which connectives to use seems rather arbitrary and to be made ad hoc. A syntactical approach sees fuzzy logics as members of the family of substructural logics, since they too lack certain structural rules. Still, even in this case, it seems hard to comprehend what fuzzy logics exactly reason about. The author proposes a new approach to the problem introducing an alternative semantics. The focus is on two prominent fuzzy logics: Łukasiewicz logic and Hájek's Basic Logic BL, i.e., the logic of continuous t-norms and their residual implications. The author relies on Boolean algebras and Heyting algebras and defines certain equivalence relations (called a-equivalence relations) on them. Such relations define equivalence classes that are interpreted as expressing ambiguity of crisp properties. The definition of certain operations over the set of equivalence classes allows him to define the structure of an MV-algebra (in the case of Boolean algebras) and the structure of a BL-algebra (in the case of a Heyting algebra). All totally ordered MV-algebras and all totally ordered Heyting algebras can be shown to be constructed in this way. Łukasiewicz logic and BL formulas are evaluated in the constructed MV-algebras and BL-algebras (respectively), making the interpretation of the new semantics explicit. Several examples of such constructions are given, while the specific proofs of the characterization can be found in [\textit{T. Vetterlein}, ``Residuated lattices arising from equivalence relations on Boolean and Brouwerian algebras'', Math. Log. Q. 54, No. 4, 350--367 (2008; Zbl 1182.03110)].