Accomodating functional predicates in the parallel architecture

Abstract

In this article, we hope to offer a possible treatment for functional predicates – with a special emphasis on modals – within Jackendoff’s (2002, 2007) and Culicover and Jackendoff’s (2005) Parallel Architecture. We argue the structural complexity which is characteristic of cartographic approaches like Cinque’s (1999, 2006), however descriptively adequate, is not strictly obligatory in light of the data explored in the literature. An alternative treatment is possible as long as one adopts different fundamental assumptions, such as, for instance, the abandonment of the syntax-semantics interface uniformity principle. We attempt to show the Parallel Architecture contains machinery that is sufficiently robust to offer alternative explanations for the behavior of functional predicates. Most of the richness of structure that is apparently requested for the explanation of the behavior of tense and modal markers can be transferred to the interface with semantics, or to semantics itself, which must become a new focus of systematic inquiry. The general conclusion, which can also be useful in executing the Minimalist Program, is that we do not need hierarchy and complexity in the syntactic component (and in UG) in order to deal with the facts surrounding functional predicates in linguistic theory.