Variation between bare and to -infinitive complements with Middle English causative verbs

The present study investigates the factors that underlie the choice between bare and to -infinitive complements with Middle English causative verbs. Scholars have put forward several hypotheses over the years, which range from metrical considerations to directness of the causative situation and the semantic features of the infinitive verb. However, it is still unclear what is the driving factor behind the choice of the infinitival form and, in addition, none of these accounts has provided a quantitative analysis of the data. By means of a conditional inference tree and random forests, this article sets out to explore what are the factors that ultimately determine the choice of the infinitival form and how they interact with one another. The analysis relies on a dataset of over 2,500 causative constructions collected from numerous Middle English corpora and include several causative verbs, namely cause, do, gar, hatan, let and make . The results indicate that the selection of the infinitival form is not straightforward but often depends on several interacting factors that vary across the different causative verbs. Theoretically, the paper suggests that bare and to -infinitive complements should be considered as alternations of a more schematic construction underspecified for the marking of the infinitive.