Grail fragments - Serbo-Croatian

Serbo-Croatian

Rudin (1988) observed that in languages like Czech, Polish and Serbo-Croation all wh-expressions must be fronted. Unlike in Bulgarian, elements can appear between the fronted wh-expressions, such as clitics, but also by adverbs and parenticals. This would indicate that the fronted expressions are separate constituents, according to Rudin (1988).

The order of the fronted wh-phrases is relatively free (as long as clitics appear in second position); for instance the subject and object wh-phrase can shift positions. The following files analyze basic wh-questions in Serbo-Croatian. Concentrating mainly on the sets of structural postulates.

• serbocroawh.pl: fragment with only structural postulates, underlying word-order: SVO [pdf]
• serbocroawh2.pl: fragment with only structural postulates, underlying word-order: SOV [pdf]
• Fragments of multiple wh-fronting with higher order type assignments:
  • serbocroawh3a.pl: higher-order-type assignments as Q-types
    • Derivation of "Ko koga vidi" (who what saw) with "who" outscoping
    • Derivation of "koga ko vidi" (what who saw) with "what" outscoping
  • serbocroawh3b.pl: higher-order-type assignments as implication-types [pdf]
  • serbocroawh3c.pl: higher-order-type assignments à la Moortgat (1998)
• serbocroawh4a.pl:- fragment with general higher order types for all the wh-words [pdf]
  • Three derivations of
    1. "Ko je sta kome dao" (who has what to whom given),
    2. "Sta je ko kome dao" (what has who to whom given),
    3. "kome je sta ko dao" (to whom has what who given)
  • serbocroawh4.pl:- fragment with higher order types for all the wh-words [pdf] This fragment an analyse movement of wh-phrases out of an embedded clause. It still derives some wrong word-orders, though...