The Origin and Development of Quasimodal Have to in English

According to Lightfoot (1979: 112), what is perhaps “most remarkable” about the reanalysis of the premodals is the subsequent development of a set of semantically identical but syntactically full verbs to fill the vacuum created. Lightfoot terms these verbs “quasimodals” and dates their appearance with modal meaning in the 15th century. I will question three of Lightfoot’s claims: the date of the modal uses of the quasimodals, their syntactic status, and their relationship to the reanalysis of the premodals. My paper will examine the semantic and syntactic development of three quasimodals in English: have to and ought to, which are equivalent to the modals must or should, and used to, which, although an habitual marker, is frequently equivalent to modal would (and earlier should, will, and shall). I will re-examine van der Gaaf’s hypothesis (1931) that have to and ought to develop from meanings of possession to those of duty, obligation, necessity, and that the change from full verb to auxiliary results in a change in syntactic order, from have + object + infinitive to have + infinitive + object (much like the traditional account of the development of the perfect). I will consider functional and semantic aspects of the development of these verbal periphrases in light of Traugott’s work on grammaticalization (1982, 1988, 1989).