Modeling vagueness by nonstandardness (Q1290590)

The paper is a philosophical discussion of some possibilities to model vagueness in mathematics. For this goal, the author takes into consideration nonstandardness and demonstrates that some properties of typically vague predicates (such as ``small'') can be modelled in nonstandard models of arithmetics. The standard numbers represent ``horizon'', i.e. the area where we may be sure that the given numbers are ``small'', and the nonstandard ones represent the area where ``small'' numbers disappear. All these questions are well known and studied by several authors also in fuzzy logic and the Alternative Set Theory (AST). They are closely connected with the ancient Sorites paradox not solvable within classical logic. The author introduces the concept of measurable predicate which is characterized by a function \(\mu: X\longrightarrow N\) such that for all \(x, y\in X\) \(\mu(x)\leq \mu(y) \& P(y)\Rightarrow P(x)\) or \(\mu(x)\leq \mu(y) \& P(x)\Rightarrow P(y)\) holds. Let me comment that this is essentially the idea of gradedness hidden in fuzzy logic. Furthermore, he introduces the concept of vague partition, which is obtained using a certain equivalence relation characterized by some distance measure \(d(x, y)\). The discussion is interesting but I miss reference to other works, especially from fuzzy logic and AST, among those also the book written by the reviewer [\textit{V. Novák}, The alternative mathematical model of linguistic semantics and pragmatics (Plenum, New York) (1992; Zbl 0837.03029)], where a great deal of ideas of this paper are discussed from wider point of view including linguistic semantics.