On Wh-Movement and On Wh-Movement
Noam Chomsky, in his seminal paper ‘‘On Wh-Movement,’’ mentions in passing an unacceptable structure involving an embedded wh-phrase moved beyond a fronted topic. The datum in question is (1a), taken from Chomsky 1977, 105, (132b); (1b) highlights the (updated) analytical details relevant here.

Taking it out of its historical context and disregarding notational differences, I want to investigate structures that superficially look like grammatical counterparts of (1b): a fronted wh-phrase moved over a preposed topic. The structures this chapter revolves around are questions that in volve two wh-phrases in between which a topicalized constituent may or may not be sandwiched in German, the primary language of concern; comparative data will be presented in support of the unfolding proposal. The upshot of this study is that looks are deceiving: what may look like a multiple wh-question involving a fronted topic is in fact multiple topicalization and does not serve as a proper information question.
Empirically, the discussion starts o¤ with the main fact that might be drawn from the paradigm in (2) and related ones, discussed in detail by Beck (1996a): in certain positions, certain elements seem to ‘‘intervene’’ in a certain way and thus destroy an otherwise well-formed dependency.

In essence, I will revisit and extend a partial set of data that served as motivation for the so-called intervention effect, sometimes even dubbed ‘‘Beck effect’’ (Hagstrom 1998), a term popularized by Pesetsky (2000), who investigates a similarly small empirical subset in his study on phrasal movement. The source of apparent ‘‘intervention’’ in the structures dis cussed here is Beck’s intriguing article, the base of my discussion. I will argue against the existence of intervention as understood there and account for the facts observed on the basis of topicalizability. The discussion combines results from and follows up on previous work of mine on multiple wh-questions in German (Grohmann 1998; 2000a, chap. 5; 2003a).
Once the notion of topicalizability is on the table and its relevance to the present subject matter clarified, I will turn to a number of consequences:
the role of topicalizability in the formation of multiple wh-questions in German;
conclusions to be drawn from the notion of topicalizability for the structure of and derivational processes involved in the German clause;
crosslinguistic implications of topicalizability for the syntax and semantics (as well as discourse properties/pragmatics) of (multiple) wh-questions in grammar.
In this sense, the chapter addresses one of the larger issues that Chomsky (1977) presented, namely, how (the rule of) wh-movement ties in with other operations in the grammar, possibly of the same or a related type (what later became known as A ¯-movement), an issue that has subsequently been the focus of many studies on the interaction of such operations as well as intervention in syntax (Cinque 1990; Rizzi 19902). It will be mainly concerned with the formation of multiple wh-questions (in German) and the connection with topicalization at large (across languages).