Lindström theorems in graded model theory (Q2220488)

In this paper, the authors try to resume the idea of a graded model theory for mathematical fuzzy logic (MFL), briefly attempted by \textit{P. Hájek} [Fuzzy logic and Lindström's theorem. Technical Report No. 874. Prague: Institute of Computer Science (2002)], by showing that such a theory is possible at least under certain technical conditions. In particular, they show that necessary conditions for proving analogous of Lindström theorems for MFLs are that the algebras of truth-values must be finite and have a truth-constant for each element of the algebra. After an introductory section, in Section 2 the authors review the basic notions of graded model theory, introduce the definition of an abstract graded logic, give examples and present the key model-theoretic properties eventually involved in their main results. In Section 3, they show that the first-order logic based on an arbitrary fixed finite MTL-chain has all the desired properties discussed in the previous section. Then, in Section 4 they establish six Lindström maximality theorems. As pointed out by the authors themselves in the concluding section, the results presented are interesting not only for MFL but also for classical first-order logic.