engleski

Prepositional antonymy in Croatian: a corpus approach

Prepositions as a word class pose various questions as to the relation between lexical and functional language units and their place in the lexicon (Jolly 1991, Šarić and Reindl 2001). Though often referred to as function words, prepositions show a) systematic semantic relations, ie. near-synonymy, polysemy, antonymy and b) a wide variety of lexical and functional (grammatical) uses, indicating a complex interplay of systematic features and contextual modifications which participate in the formation of their meaning. Semantic relations such as antonymy are mostly discussed in terms of adjectives, nouns and verbs, leaving out a detailed description of antonymy effects in other word classes such as prepositions (e.g. Lyons 1977, Cruse 1986, Jones et al. 2012). By adopting the methodology of antonymy research developed for identifying and extracting antonyms from corpora, we examine the co-occurrence of prepositional antonyms in the Croatian National Corpus. We take up the cognitive linguistic position of examining antonymy as a prototype based category based on both conceptual opposition and contextual modifications (Paradis et al. 2009), and we observe its workings on the novel prepositional dataset. Based on the primary domains and conceptual structures that motivate prepositional opposition formation, we divide the antonyms into spatial (directional and locational), temporal and non- dimensional types. For each of the antonym types there are different contextual modifications and conceptual structures that shape these antonymy relations, indicating a complex interplay between language system and language use.

prepositions; lexical relations; antonymy; opposition; Croatian

nije evidentirano