On the semantics of Czech free relatives

Title: On the semantics of Czech free relatives
Variant title:
  • O sémantice bezhlavých relativních vět v češtině
Author: Šimík, Radek
Source document: Linguistica Brunensia. 2016, vol. 64, iss. 1, pp. 109-129
  • ISSN
    1803-7410 (print)
    2336-4440 (online)
Type: Article
License: Not specified license

Notice: These citations are automatically created and might not follow citation rules properly.

This paper offers a semantic analysis of Czech free relatives and pays particular attention to so called ever free relatives, i.e. free relatives whose wh-word is modified by the ever morpheme (the postfix -koli(v) in Czech). I argue that Czech free relatives should be uniformly analyzed as definite descriptions, denoting the maximal (plural) entity that satisfies the description provided by the free relative. I show that the analysis can be extended to Czech ever free relatives, despite the fact that they only exhibit the universal-like reading and not the identity reading, the latter of which has always figured prominently as an argument for the definiteness of ever free relatives. I argue that the reason for the lack of the identity reading is the stricter variation requirement imposed the denotation of ever free relatives in Czech (compared to, say, English). In particular, I argue that the denotation of an ever free relative in Czech must covary with a variable bound by a clausemate quantifier. This, in effect, makes Czech ever free relatives instances of "dependent definites" – a correlate of the better known "dependent indefinites".