Relative Clauses in English conversation: Relativizers, frequency, and the notion of construction

This paper is a usage-based study of the grammar of that set of English Relative Clauses with which a relativizer has been described as optional. We argue that the regularities in the use of relativizers in English can be seen as systematically arising from pragmatic-prosodic factors, creating frequency effects, resulting in some cases highly grammaticized formats: the more the Main Clause and the Relative Clause are integrated with each other, that is, approach monoclausal status, the more likely we are to find no relativizer used; conversely, the more separate the two clauses are, the more likely we are to find an overt relativizer. These findings have led us to suggest that the more monoclausal combinations have become unitary storage and processing chunks. We thus see these findings as a contribution not only to our understanding of Relative Clauses, but to our understanding of syntactic organization in general and of the nature of the grammatical practices in which speakers engage in everyday interactions.