Cleft sentences, construction grammar and grammaticalization

This thesis examines the structure and function o f the English //-cleft configuration within the fram ework o f construction grammar. M y analysis begins with the claim that //-clefts are a subtype o f specificational copular sentence. After identifying problems with previous accounts, I outline my own, original analysis o f specificational NP be NPsentences. I argue that specificational meaning involves an asymmetric predication relation and is dependent upon the inherent semantics o f definite noun phrases (rather than syntactic movement). I treat nominal predication set theoretically, as a semantic relation between mem bers and sets. I claim that specificational meaning is brought about by a reinterpretation o f the class-membership relation involving definite NP predicates, whereby the referent is identified as the unique member o f a restricted and existentially presupposed set.As a m em ber o f the family o f specificational copular sentences, the //-cleft inherits properties from the more basic construction. From this, it follows that //-clefts should also involve a nominal predication relation, containing a definite NP predicate. This leads me to argue in favour o f a non-derivational extraposition-from-NP analysis o f //-clefts, in which the pronoun it and the cleft clause (analysed here as a restrictive relative) function together as a discontinuous definite description. M y analysis improves on similar accounts o f this type in two ways. First, since my analysis explains the role that definite descriptions play in the creation of specificational meaning, I am able to explain, rather than simply identify, the numerous similarities between it-clefts and definite noun phrases. Second, my analysis o f specificational sentences as involving a nominal predication relation allows for a straightforward account o f the relationship between specificational and predicational it-clefts.The thesis also examines the historical development of the //-cleft construction. I show that (a) much of the it-cleft’s structure is reminiscent of an earlier stage of the language and (b) the construction has become increasingly schematic and productive over time, sanctioning instances which override inheritance from the more basic specificational schema. In this way, the historical evidence provides an explanation for the it-cleft’s idiosyncratic properties. Together, my synchronic and diachronic analyses add up to a maximally explanatory account of the it-cleft construction