Names: 
  Pawel Rutkowski, Kamil Szczegot
Title of the paper:
  On the syntax of functional elements in Polish: numerals and the word cos
Affiliation:
  University of Warsaw (Rutkowski)
  Cambridge University (Szczegot)


ON THE SYNTAX OF FUNCTIONAL ELEMENTS IN POLISH: NUMERALS AND THE WORD COS


 As has been noted in the literature (cf. Przepiorkowski 2000a), the syntax of the form cos 'something' resembles closely the agreement pattern exhibited by Polish proper numerals (cardinal numerals above cztery 'four'). 

 In the subject (1a) and accusative object positions, the intrinsically accusative (cf. Franks 1995 and Przepiorkowski 1996) proper numerals always assign genitive to the nouns following them (the so-called Genitive of Quantification - GEN(Q)). However, if such a construction is put into one of the oblique cases (1b), the numeral fails to govern the genitive case and the noun takes the oblique case of the whole DP.

(1)

(a) piec  psow/*psy    je
     five:ACC   dogs:GEN/*ACC    eat     
     'five dogs eat'

(b) z   piecioma   psami/*psow     
     with   five:INSTR   dogs:INSTR/*GEN
     'with five dogs'

 The construction of the form cos + genitive Adjectival Phrase (AP:GEN) exhibits the same unusual syntactic behaviour, as shown in (2a) and (2b). This parallelism between numeral phrases and the construction cos+AP:GEN also occurs in other Slavic languages (cf. Grepl and Karlik 1986). 

(2)

(a)  cos    ciekawego/*ciekawe   istnieje
      something:ACC   interesting:GEN/*ACC   exists     
     'something interesting exists'

(b)  z   czyms    ciekawym / *ciekawego     
     with something:INSTR interesting:INSTR/*GEN     
     'with something interesting'

 The aim of this paper is to provide an explanation for the mixed agreement / non-agreement pattern. We suggest that this syntactic phenomenon can be accounted for in terms of D-structure vs. S-structure case assignment distinction. 

 Chomsky (1986) posits two types of case: structural, assigned at S-structure (based on the position of a nominal phrase at S-structure) and inherent (often referred to as lexical), assigned by a lexical element at D-structure. We argue that numerals and the elements such as cos should be analysed as functional, which is mainly because they form a closed class and can be analysed as semantically vacuous (cf. Pollard and Yoo 1998, Przepiorkowski 2000b).

 Since both numerals and cos are functional elements, they are inserted into the syntactic derivation as late as at S-structure (cf. Emonds 1997) - after the lexical case is assigned, but before the structural case is (cf. Rutkowski 2000, Veselovska 1997). Hence, those elements act as case-assigners when the phrase is in a structural cases (i.e. when the noun or adjective has not been assigned any case at D-structure), but not when it is in a lexical case. In the latter situation the noun or adjective has already been assigned case by the time the numeral or cos is inserted and so the numeral or cos cannot govern genitive but only checks its case features and agrees with the noun or adjective.

 The analysis we argue for shows that the syntax of cos should not be considered "numeral" (as it is in, e.g. Grepl and Karlik 1986). The similarity between the behaviour of numerals and the construction cos+AP:GEN is only caused by the fact that both the numeral and cos are functional heads. These are two independent phenomena driven by the same set of principles. 

References

Chomsky, N. 1986. Knowledge of Language: Its Nature, Origin and Use. New York: Praeger

Emonds, J.E. 1997. Grammar & Lexicon: the English Syntacticon. Ms. University of Durham

Franks, S. 1995. Parameters of Slavic Morphosyntax. New York, Oxford: Oxford University Press

Grepl, M. and P. Karlik. 1986. Skladba spisovne cestiny, Prague: Statni pedagogicke nakladatelstvi

Pollard, C. and E.J. Yoo. 1998. A unified theory of scope for quantifiers and wh-phrases. Jornal of Linguistics 34, 415-445.

Przepiorkowski, A. 1996. Case assignment in Polish: Towards an HPSG analysis. In: C. Grover and E. Vallduvi (eds.). Edinburgh Working Papers in Cognitive Science. Vol.12: Studies in HPSG. Edinburgh: University of Edinburgh. 191-228

Przepiorkowski, A. 2000a. Predicative case agreement with Quantifier Phrases in Polish. Paper presented at the UNC Linguistics Colloquium, University of North Carolina, Chapel Hill, April 2000

Przepiorkowski, A. 2000b. ARG-ST on phrases headed by semantically vacuous words: Evidence from Polish. Paper presented at the 7th International Cinference on HPSG, Berkeley, July 2000 

Rutkowski, P. 2000. Numeral Phrases in Polish and Estonian. To appear in: A. Holmer, Svantesson & Viberg (eds.). Proceedings of the 18th Scandinavian Conference of Linguistics. Travaux de l'Institut de Linguistique de Lund. Lund: Lund University

Veselovska, L. 1997. Agreement patterns of group nouns and quantifiers (in Czech). Ms. University of Olomouc