The aim of the present paper is to develop an analysis of concealed questions (=CQ) in English, concentrating specifically on their semantic interpretation. Assuming that the most typical CQs are ones that involve an ‘NP+“complement” relative clause (=RC)’ structure such as the italicized part in (i) (the bracketed part is the RC in question), we first discuss some aspects of complement RCs, including those aspects related to their relatives, antecedents, and gaps inside the RCs:
(i) I don't know the kind of man he was ten years agol. In doing so, we specifically argue that the familiar ‘variable’ interpretation available to the gaps inside ordinary restrictive RCs cannot be made available to the cases of the complement RCs involved in CQs ; we argue, instead, that in CQs the gaps in question correspond to N' that denotes ‘generic concept.’ A crucial idea in the present study is the ‘generic concept and its instantiation.’ This idea, together with the ‘definiteness and the attendant existential presupposition, ’ plays a central role in determining the details of the semantic interpretation of complement RCs, and for that matter, that of CQs containing such RCs.
In the final part of the paper, we propose a grammatical process called ‘reanalysis, ’ whereby a relative clause structure (when the noun phrase containing the RC occurs with a predicate such as tell and show) is reanalyzed as an interrogative clause structure at LF ; by introducing the variable ‘X, ’ we claim that it will be possible for the noun phrase containing an RC to be reinterpreted ‘similarly’ to an indirect wh question. We also claim that the existence of the ‘non-information-seeking’ reading of CQs follows from our analysis of RCs.
The Linguistic Society of Japan