Topic, Focus, and Syntactic Representa- tions

University of Edinburgh and University of Pennsylvania1. IntroductionCopular sentences like (1) may appear straightforward; they are certainlycommon.(1) The person I most wanted to meet was Tom Lehrer.Particularlysince the work of Higgins 1979, however, it has become clear thatsentences like (1), together with pseudoclefts like (2), to which they are inti-mately related, pose challenging questions to our understanding of linguisticrepresentations.(2) What I was hoping for was his autograph.In this paper we will indicate briefly what is at issue, discuss in passingsome recent contributions to understanding the problem, and then concentrateon one aspect that we believe is essential to understanding what is happening:the information structure of such sentences. This aspect of specificationalsentences was argued to be important in Heycock and Kroch 1999, but onlypartially integrated into the analysis presented there; in this paper we attemptto go further in developing an account in which the role of information struc-ture is central.2. Setting out the problem: specificational sentencesIt is commonly assumed that at least in some sentences the copula be isnot an argument-taking verb, but rather an identity operator on predicates (butsee Rothstein 2001 for a different view). On this view the “basic” predicatein the main clause in (3a) is the noun phrase a nice woman, just as it is in thesmall clause in (3b):(3) a. Joan is a nice woman.b. I consider [