Published online by Cambridge University Press: 01 January 2026
This article uses formal and usage-based data and methods to argue for a hybrid model of English tensed auxiliary contraction combining lexical syntax with a dynamic exemplar lexicon. The hybrid model can explain why the contractions involve lexically specific phonetic fusions that have become morphologized and lexically stored, yet remain syntactically independent, and why the probability of contraction itself is a function of the adjacent cooccurrences of the subject and auxiliary in usage, yet is also subject to the constraints of the grammatical context. Novel evidence includes a corpus study and a formal analysis of a multiword expression of classic usage-based grammar.
I am grateful to several colleagues who helped me advance this study: to Jen Hay for making data for the corpus study available while I had a Strategic Fellowship at the New Zealand Institute for Language, Brain, and Behaviour in 2015; to Arto Anttila for his collaborations with me in corpus phonology and syntax from our joint Stanford course in 2012 and through summer projects in 2015 and 2017; to Adams Bodomo for his invitation and very generous support of my presentation at LFG '18 at the University of Vienna; to Avery Andrews, Joan Bybee, and Janet Pierrehumbert in particular among those who gave me comments on the first draft; and to the editorial team of this journal, including three anonymous referees who gave detailed and helpful comments on subsequent drafts.