ABSTRACT
The basic idea of syntactic fictionalism would require expansion to allow for a metalanguage that might well be pragmatic (and semantic) as well as syntactic. The point is to construct a syntactical metalanguage (SL) as an E-substitute for the non-E portion of science. The non-E portion itself is taken as an uninterpreted object language (OL) with fixed vocabulary of symbols, a specifiable set of symbol strings constituting formulae, a subset of these constituting primitive formulae, and certain sequences of formulae constituting valid transformations. An instructive example is afforded by the work of Nelson Goodman and Willard Van Orman Quine on the interpretation of mathematics, in their study, "Steps Toward a Constructive Nominalism". They are concerned, not with the problem of theoretical terms in natural science, but rather with the possibility of a nominalistic interpretation of mathematics.
