The project aims at modelling the factor "frequency" in language change if the change is seen as a change in the cognitive representation of a linguistic form.
The current restructuring of the English modal system has long been noted as an ongoing language change process. Previous work (especially Krug 2000) has shown that the reduction of quasi-modals such as BE going to, HAVE got to, etc. may lead to the emergence of a new class of modals comprising gonna, gotta, wanna, etc. However, the fundamental question at which point and how a new lexical item is created, has hitherto remained unaddressed.
It is this question that my project seeks to illuminate on the basis of three research questions: Is there a reanalysis and how is it actualized? What are the effects of the frequencies of the reduced forms as compared to the (periphrastic) source forms? What factors determine the choice of a full or reduced form, and do these factors change over time?
It can be shown that 'gonna'/'gotta'/'wanna' do move towards a conceptual independence from their source forms in language use. This I call the 'emancipation effect'. By this emancipation, they lose the characteristics of reduction, and also separate themselves from the more general pattern of 'to'-contraction (which also produces forms like "tryna", "needa"). The most frequent one, 'gonna', is the most advanced on this path.