Abstract

It appears that both English and French contain a mixture of ergative and transitive verbal constructions. How useful is reference to a dual-structure model of syntax in determining the bias of a particular language? Certain verbal constructions in English and French could be generated from either transitive or ergative deep structures; we must therefore turn to other criteria - syntactic cohesion and predominant constituent structures in 'surface structure' - for a non-arbitrary - and adequate - decision