Truth as translation. Part B (Q5947378)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1661014
Language Label Description Also known as
default for all languages
No label defined
    English
    Truth as translation. Part B
    scientific article; zbMATH DE number 1661014

      Statements

      Truth as translation. Part B (English)
      0 references
      0 references
      17 April 2002
      0 references
      Whereas the first part of this paper [ibid. 30, No.~4, 281-307 (2001; Zbl 0983.03003)] deals with propositional languages, the second part concentrates on full first-order languages. The author presents his account in terms of possible worlds, but basically he studies a variety of the rule-of-revision theory of truth. Let \(L_T\) be a first-order language with a unary (truth) predicate. The models of \(L_T\) have the form \((M,S)\) where \(M\) interprets the vocabulary of \(L_T\) with the exception of the truth predicate, which gets \(S\) as its extension. \((M,S_2)\) is a \textit{metaworld} for \((M,S_1)\) iff \(S_1\) contains exactly those sentences of \(L_T\) which are true in \((M,S_1)\). Suppose \(M'\) is a model for the language without the truth predicate. Then Leitgeb shows the following. There is an extension \(M\) of \(M'\) and a series \(S_{n-2},S_{-1},S_0,S_1,S_2,\ldots\) of subsets of the domain of \(M\) such that \((M,S_n)\) is metaworld for \((M,S_{n+1})\), i.e., there is a series of metaworlds ordered like \(\mathbb{Z}\). Moreover, the truth predicate satisfies axioms saying that the truth predicate commutes with connectives and quantifiers in all models \((M,S_k)\) with \(k\leq 0\). For the positive numbers, however, only schematic versions of these axioms hold, e.g., \(Tr[\neg \phi]\leftrightarrow \neg Tr[\phi ]\) for any single sentence \(\phi\) in \(L_T\). Reviewer's remarks: Under fairly general conditions \(M\) will be recursively saturated. This follows from a theorem by \textit{A. H. Lachlan} [Can. Math. Bull. 24, 295-297 (1981; Zbl 0471.03055)]. Leitgeb's construction of \(S_{n-2},S_{-1},S_0,S_1,S_2,\ldots\) suffers from the fact that, from a certain number \(n\) on, truth commutes only schematically with the connectives and quantifiers. In certain important cases this weakness can be overcome. By iterated applications of a construction due to \textit{H. Kotlarski, S. Krajewski} and \textit{A. H. Lachlan} [Can. Math. Bull. 24, 283-293 (1981; Zbl 0471.03054)] one can obtain a series \(S_{n-2},S_{-1},S_0,S_1,S_2,\ldots\) such that all \((M,S_k)\) yield the universally quantified versions and not only the schemata. This is possible because starting from a countable recursively saturated model \((M,S_n)\) one can construct a suitable set \(S_{n+1}\) such that \((M,S_{n+1})\) is again recursively saturated.
      0 references
      truth
      0 references
      translation
      0 references
      quantification
      0 references
      possible worlds semantics
      0 references
      semantically closed languages
      0 references
      nonstandard models
      0 references
      first-order languages
      0 references

      Identifiers