The internal syntax of adverbial clauses

The starting point of this paper is the observation that English temporal and conditional clauses resist argument fronting. The first part of the paper discusses the empirical and conceptual problems that arise in relation to the truncation analysis proposed in Haegeman (2003a, 2006b,c). It is shown that the movement analysis of adverbial clauses (going back to Geis, 1970, 1975, and adopted by, among others, Larson, 1985, 1987, 1990; Demirdache and Uribe-Etxebarria, 2004:165–176; Bhatt and Pancheva, 2006; Lipták, 2005; Tomaszewicz, in press; Citko, 2000) allows us to analyse this restriction in terms of an intervention effect. The movement account leads to the prediction (i) that not all English adverbial clauses are incompatible with argument fronting, (ii) that Romance adverbial clauses are compatible with clitic left dislocation, and (iii) that temporal adverbial clauses contain a trigger for French Stylistic Inversion.