Creation and re-creation: The interplay of activity and structure in language

Abstract

The paper investigates the possibility to give a pragmatic, non-formalist account of sentence structure that is able to explain its contribution to sentence meanings. Specifically it raises the question whether the possibility to construct logical languages in calculus form justifies the expectation that for a natural language it should be possible to construct a fixed system of categories and rules that allows to produce all and only the meaningful sentences of that language. Drawing on ideas that Wittgenstein had developed in his criticism of Frege, this question is answered negatively: It is impossible to formulate such a complete list because the relevant semantic categories are not given in advance, in a realm outside language. Instead, they are created by linguistic activity and are projected into ever new fields of application, changing their semantic value in these projective steps. So speakers on the one hand rely on fixed linguistic structures they only ‘re-create’ (traditional grammar); on the other hand they use these structures creatively in ways that cannot be anticipated and spelled out exhaustively. Both aspects together explain Wittgenstein's neglect of linguistic structure.