Slavic languages are known as multiple wh-fronting languages, where all wh-phrases in a multiple interrogative are fronted to the periphery of the clause. This was investigated by Pesetsky (1987), Rudin (1988), Boškovic (2002) among others. The present study explores the nature of a constraint against multiple fronting of certain wh-elements in Slavic, namely, wh-elements that are left-branches. I propose an analysis where Left-branch Extraction (LBE) is treated as head-movement, and derive the prohibition against multiple LBE from minimality, formulated as in Chomsky (1995).
Consider multiple wh-fronting in Russian (1a), while (1b) is unacceptable. Russian, as the majority of Slavic languages, also allows LBE, (2). However, multiple instances of LBE in the same derivation are impossible, (3). The same prohibition has been observed for Serbo-Croatian by Fernandez-Salgueiro (2005).
Since LBE takes place only out of complex wh-phrases, I examine the syntactic behavior of such complex wh-phrases. Boeckx and Grohmann (2004) argue for a unified analysis of scrambling and movement of complex (d-linked) wh-phrases. They view this process as topicalization (i.e. movement to Topic Phrase). Adopting their analysis, I further determine by inserting intervening material between the wh-phrases, as in (4), that multiple complex wh-phrases move to multiple Topic Phrases. I also propose that LBE is head-adjunction to Top0.
In the derivation with multiple LBE, (5), attracting kakoj by the higher Top0 in Step 3 is impossible over the intervening head kakomu adjoined to the lower Top0. Hence, I conclude that multiple LBE is impossible due to a minimality violation. I will also demonstrate how multiple LBE interacts with sluicing (IP-ellipsis), which does not repair such derivations. This is predicted by my analysis since minimality is a derivational constraint.
(1) a. Komy1 cto2 [Ivan dal t1 t2 ]?
whoDAT whatACC Ivan gave
‘Who did Ivan give what?’
b. *Komu1 [Ivan dal t1 cto]?
(2) Kakuju on kupil [t mašinu]?
what-kind he bought car
‘What kind of a car did he buy?’
(3) *Kakomu1 kakoj2 [Ivan zadal t1 studentu] [t2 vopros]?
whichDAT whichACC Ivan gave studentDAT questionACC
‘Which student did Ivan ask which question?’
(4) Kakomu studentu, po tvojemu mneniju, kakoj vopros Ivan zadal?
in your opinion
(5)
Step1: [TopP [Top kakomu Top0] [ Ivan zadal [t studentu] [kakoj vopros]]]
Step2: [TopP Top0 [TopP [Top kakomu Top0] [ Ivan zadal [t studentu] [kakoj vopros]]]]
Step3: *[TopP [Top kakoj2 Top0] [TopP [Top kakomu1 Top0] [ Ivan zadal [t1 studentu] [t2 vopros]]]]
References
Boeckx, C. and K. Grohmann, 2004. "Sub-Move: Towards a Unified Account of Scrambling and D-Linking." In Peripheries: Syntactic Edges and Their Effects. Dordrecht: Kluwer.
Boškovic, Ž. 2002. On multiple wh-fronting. Linguistic Inquiry 33.
Chomsky, N. 1995. The Minimalist Program. MIT Press.
Fernandez-Salgueiro, G. 2005. ‘To move or not to move’: on the incompatibility of multiple wh- movement with left-branch extraction in Serbo-Croatian. Paper presented at CLS 41.
Pesetsky, D. 1987. Wh-in-situ: Movement and Unselective Binding. In The Representation of
(In)definiteness. MIT Press.
Rudin, C. 1988. On multiple questions and multiple wh-fronting. Natural Language and Linguistic
Theory 6.