Bad News about the Faire-Construction in French

Abstract

This paper discusses a select few of the issues related to French causative constructions with faire. Our ambition here is not to propose a new treatment of the phenomenon, but to call attention to "unsifted data" that do not support the "demotion" hypothesis suggested by Carrie (1975, 1976, 1981). We will present evidence that Comrie's general analysis -- established mainly on the basis of canonical constructions such as sentences (1, 2) below-cannot stand unchanged, and needs further refinement, if it is to account adequately for all possible instances of the constructions in question. 2 (1) a. Valéry mange. - 'Valéry eats' b. Francois fait manger Valéry. - 'François makes Valéry eat' (2) a. Valéry mange un escargot. - 'Valéry eats a snail' b. François fait manger un escargot á Valéry. - 'François makes Valéry eat a snail' The (b)-sentences above are instances of the faire-construction, and the (a)-sentences are their non- causative counterparts. These latter may contain various types of predicates: one -place, two-place, and three-place predicates. The causative construction contains an entity faire, which we shall refer to as a "causative marker ", and an infinitival verbal form, which can be followed by postverbal complements. The introduction of a new element--i.e., the CAUSER (François, in (1) and (2)) --is the source of the "unorthodox" position occupied by the CAUSEE (Valéry, in (1) and (2)), which is "pushed" into a "secondary" position after the infinitive verb manger.