Why Multiple Clefts Are Disallowed

This paper presents an analysis of clefts. In Korean, wh-expressions cannot be clefted along with another constituent, and true multiple clefts do not seem to exist. We discuss and explain two apparent exceptions to this generalization: (i) way ‘why’, which does appear in such clefts, and (ii) multiple clefts read with an echo intonation (which will be called “echo clefts” throughout this paper). We show that the seemingly exceptional behavior of way ‘why’ in this instance can be attributed to the same reason as its exceptional behavior observed elsewhere (Huang 1982, inter alia): way ‘why’ is a clausal modifier, and as such, modifies the clause in whose specifier it is merged. The examples where way ‘why’ appears side by side with a clefted phrase are thus examples where only one constituent has been clefted. The exceptional behavior of echo clefts also falls out from the oft-observed peculiarity of echo questions. The lack of locality effects, which sets apart echo questions from run-of-the-mill questions, is what makes these constructions possible and suggests that a different mechanism is involved in the interpretation of echo questions. This paper is organized as follows. Section 2 introduces the data in question along with Hiraiwa and Ishihara’s (2002) proposal for Japanese clefts, the latter of which will provide a syntactic backdrop for our analysis. Section 3 is devoted to the discussion of way ‘why’ in clefts and elsewhere, and section 4, to echo clefts. Section 5 discusses implications of our analysis and issues for further research.