The nature of fronted wh-words in Korean type languages has been a topic of great controversy, with the widely held assumption being that they behave in a uniform fashion (Hoji 1985, Saito 1989, Takahashi 1993, 1994, Choe 1994, Cheng 1997, Bošković 1999). I challenge this common view, following the original proposal by Choi (2007b), claiming that the fronted wh-words in Korean are a heterogeneous set in that indefinite wh-words constitute focus movement, whereas the non-indefinite adjunct wh-word way ‘why’ is wh-movement. The heterogeneous nature of the fronted wh-words nicely deals with the contrast in superiority and wh-island effects along with the contrast in the cleft construction.

In this paper, I show that VP-front, some cases of wh-movement, and topicalization all share properties that would be explained by adopting some version of minimal reconstruction proposed in Jang (2000). I show that previos analyses proposed to resolve the issue of binding effects exhibited in VP fronting sentences may not be viable to account for wider range of data. That is Huang (1993) proposes that VP-internal Subject Hypothesis and general binding theory are enough to explain the said phenomena. Takano (1995) proposes that reconstruction, which is independently motivated, is needed to explain the binding phenomena in VP fronting and related constructions. This paper shows that neither analysis, as it is now, can cover relevant data and that some more qualification such as Jang (2000) may be needed to conditions on reconstruction.

Man-Kun Yoon. 1999. Unselective Binding and Multiple Wh-Fronting Constructions. Studies in Modern Grammar 15, 23-41. The first aim of this paper is to argue that any event in our world necessarily co-occurs with four adverbials such as time, place, manner and reason adverbials which can turn into four different wh-adjuncts; when, where, why and how. This paper further argues that this phenomenon is language universal. Second aim is that unselective binding applies only to wh-NPs, not wh-adjuncts, and hence covert raising is necessary for the checking of wh-adjuncts. Although wh-NPs can be interpreted in situ by unselective binding, case checking of wh-NPs requires covert or overt raising of them. Third aim is that nominative wh-NPs can be checked by C。 and accusative wh-NPs by C^(omax). More elaborate purposes are given below in introduction: