Edited by Svenja Kranich and Tine Breban
[Studies in Language Companion Series 218] 2021
► pp. 261–290
Numerous corpus studies have firmly established that in recent English, most core modals (e.g. must) have been declining in frequency, while the semi-modals (e.g. have to) have been on the rise (cf. e.g. Krug 2000; Mair & Leech 2006; Leech & Smith 2006). The present paper follows the constructional approach to modal meaning taken by Cappelle and Depraetere (2016), who show that modals tend to occur in certain contexts with particular meaning. An investigation of a sample of may and must in COHA (1960s & 2000s) shows that the decline of the modals (and the foreshadowed loss of some) may be witnessed particularly in the demise of certain constructions. A close-up investigation shows the construction we + may + verb of speaking/reasoning is on its way to be lost. Relevant factors (genre, function, culture) are discussed and suggestions about the general implications of the loss of this particular construction will be presented.