Questions and Implications Pied-Piping, Feature Movement, and Wh-Subjects
A number of implications and new questions for further study arise if this analysis is on the right track. First, does the pattern predicted for wh-subjects in English occur in any language where phonologically null C and Spec,IP are linearly adjacent, or is there observable variation with respect to this pattern? For example, if a language allows for ‘‘relative pronoun deletion’’ of subject wh-elements in relative clauses and related constructions, then arguably the wh-subject has raised to Spec,CP in overt syntax. The same language should also allow subject topicalization. In any case, the analysis presented here provides for clear empirical tests of its validity.
Second, note again that this analysis is silent about why there is no inversion in root subject wh-questions. One view consistent with the present approach would be that inversion in root wh-questions is triggered by the presence of a wh-phrase in Spec,CP. This view takes inversion to be an ancillary property of root wh-questions, perhaps applying in the post-Spell-Out derivation to PF (see Chomsky 2000, 149n68 and Boeckx and Stjepanovic´ 2001 for a suggestion along these lines for head movement).
Third, the question also arises of how the adjacency requirement is sat isfied in phrasal pied-piping cases, illustrated in (1), with [a picture of whose mother] in Spec,CP.
(1)
One possibility, rooted in classical proposals for phrasal/clausal pied piping, is that the wh-feature of whose percolates to the head of the upper NP prior to the application of Move F, so that the category associated with the wh-feature is the entire phrase a picture of whose mother. A related question arises concerning what rules out left branch extractions like *Whosei did you meet ti mother? Such facts suggest that the object affected by Pied-Pipe must be a maximal phrase and not a head, indicating a dichotomy between head movement and XP-movement that has also been noticed in recent work (in particular, Lasnik (2001) points out that the failure to apply head movement may be remedied by deletion, but such is not the case for XP-movement, suggesting that head movement is a PF phenomenon that interacts with other PF rules).
Finally, ‘‘partial’’ wh-movement constructions such as those found in German may lend further support for the two-chain analysis. In these constructions, the matrix Spec,CP is occupied by an invariant wh-scope marker, with the actual wh-phrase occupying the Spec of the embedded clause. This is shown in (2) (from McDaniel 1989, 569).
(2)
Cheng (2000) has analyzed such constructions as involving feature movement to the matrix CP and partial pied-piping of the wh-phrase to the embedded CP. The wh-feature in the matrix clause is spelled out as was (an invariant scope marker, homophonous with the wh-word was ‘what’), and the wh-phrase may actually take matrix scope (via the raised wh-feature), even though the category itself is not raised into the matrix clause. Under Cheng’s analysis, the wh-feature first moves into the embedded CP, and this triggers pied-piping of the category. The wh feature then moves on to the matrix CP, where it is spelled out. This sub sequent feature movement leaves the wh-phrase behind in the embedded CP. Given such an analysis, the existence of ‘‘partial’’ wh-movement, like the behavior of English wh-subjects, suggests that feature movement may indeed apply and that it may apply independently of category pied piping in overt syntax.
