Search for a command to run...
Abstract This book presents a profound critique of syntactic theory and syntactic argumentation. Recent syntactic theories are essentially formal models for the representation of grammatical knowledge. These theories posit complex syntactic structures in the analysis of sentences, consisting of atomic primitive syntactic categories and relations. The result of this approach to syntax has been an endless cycle of new and revised theories of syntactic representation. The book argues that these types of syntactic theories are incompatible with the grammatical variation found within and across languages. The extent of grammatical variation demonstrates that no scheme of atomic primitive syntactic categories and relations can form the basis of an empirically adequate syntactic theory. This book defends three theses: (i) constructions are the primitive units of syntactic representation, and grammatical categories are derivative; (ii) the only syntactic structures are the relations between a construction and the elements that make it up (that is, there is no need to posit syntactic relations); and (iii) constructions are language-specific. Constructions are complex units pairing form and meaning. Grammatical categories within and across languages are mapped onto a universal conceptual space, following the semantic map model in typology. The structure of conceptual space constrains how meaning is encoded in linguistic form, and reflects the structure of the human mind.