Abstract
The aim of the paper is to present new findings about semantic and grammatical functions of two specific types of attributive constructions in Czech: pronominal attributive possession (1a) and adjectival attributive possession (1b). Both constructions contrast with a construction formed by a genitive noun phrase (1c), the most general adnominal construction type, see O'Connor et al (2013). (1) CZECH a. můj křeček my hamster 'my hamster' b. ségřin křeček sister:POSS_ADJ hamster 'sister's hamster' c. křeček mý ségry hamster my:GEN sister:GEN 'a/the hamster of my sister' In my corpus study, I adopt a usage-based approach to language (e.g.
Bybee 2007, Hawkins 2004). I describe possessive constructions in Czech with regard to hypotheses based on statistical tendencies in language use.
Specifically, analyzing the data from the Czech corpus of spoken language ORAL2013, I test the hypotheses concerning different frequency distributions of pronominal/adjectival possessive constructions depending on different classes of referents and their position on the animacy (nominal) hierarchy. The main results of the research revealed the relationship between frequency of use and the animacy hierarchy.
I will show which semantic classes of possessive adjectives may be considered the core of the category, how to interpret frequency differences between the possessive adjectives derived from masculines and those derived from feminines and I will discuss possible functional explanations of the whole category of attributive possession in Czech.