Reinvestigating ambiguity and frequency in reanalysis

A two-step methodology for corpus-linguistic analyses based on bridging use exposure

Abstract

The concept of reanalysis is widely recognized as a basic type of language change. However, given that the initial stage of processes of reanalysis is characterized by identical surface structures of the old and the new interpretation, basic methodological challenges arise for the identification of the innovation stage. Assuming that the notion of ambiguity and the relative frequencies of ambiguous and non-ambiguous uses play a crucial role, I propose a two-step methodology for diachronic corpus studies. The first task is to identify unambiguous switch uses, i.e. the first new uses which define the stage where the new structure has already been established. In a next step, a quantitative analysis of conventional uses and potentially ambiguous bridging uses can then be carried out in order to determine the strength of bridging use exposure (BUE) for the preceding period. Two case studies will illustrate the insights to be gained from such an approach which integrates fine-grained qualitative analyses of ambiguous and non-ambiguous uses in speaker-hearer interaction and quantitative investigations of the global frequencies of different types of uses in the speech community.