Retour vers liste

Détail de la contribution



Abstract/Résumé: The paper targets the asymmetries between internal and external prefixes in Serbo-Croatian (S-C), reaching a uniform analysis in terms of aspectual relativization. Previous analyses of Slavic verb-prefixes form two groups: 1) those generating their two classes in two distinct functional projections, stipulating their asymmetries in the specification of the two projections (Svenonius2004, DiSciullo&Slabakova2005) and 2) those generating them in the same functional (or lexical) projection, and deriving their interpretive and (morpho-)syntactic differences from the type of complement they take (Arsenijevic2007, Žaucer2010). The former analyses fail to explain why the same set of elements (prefixed counterparts of prepositions) appear in both positions and why the quantitative semantic effect of an external prefix is closely related to the meaning of the corresponding preposition (iz ‘from’ for exhaustivity, na ‘on’ for accumulation etc.), among other shortcomings. The latter derive the wrong ordering of prefixes and fail to account for the overt realization of the result preposition (Žaucer2010 for a detailed comparison). The paper argues that Slavic verb-prefixes are best analyzed as instances of concord between the verb and the result predicate, triggered by a telic aspectual operator (along the lines of Zeijsltra2004, or Haegeman&Lohndal2010). The VP starts out as a small clause (parallel to Koopman2005’s account of DP), where the verb first moves for interpretability to V (which can also be treated as the lowest Asp position, parallel to Koopman’s D as the lowest C). Subsequently, VP moves to a higher aspectually active position to facilitate agreement triggered by the operator. This step can be repeated, leading to a stacking of verb prefixes. This analysis avoids the shortcomings of previous analyses (it derives the right ordering, explains the same prefixes in both roles and the double lexicalization of the result preposition etc.), while still capturing the asymmetries (by the lexical nature of the ‘first relativization’), and gives a deeper insight into the nature of aspectual relations (e.g. how the result and process subevent are specified to consist one macro-event). At the theoretical side, it establishes a parallel between the nominal and verbal projections in terms of the capacity of relativization, extends Kayne2009’s argument that lexical nouns do not take complements to the domain of lexical verbs, and recognizes relativization within the verbal domain, where no such pattern has been recognized so far.