The verbal prefix o(b)– in Croatian and Bulgarian: The semantic network and challenges of a corpus–based study

This study compares the verbal prefix o(b)– in two South Slavic languages, Croatian and Bulgarian, from a cognitive linguistic perspective. We focus on the problems arising when constructing the semantic network of this polysemous prefix, particularly on 1) isolating the prefix’s meaning from the meaning of the base verb and 2) identifying core/dominant sub–meanings for all verbs and giving them corresponding semantic labels. Our approach to morphology is based on extensive databases of verbs collected from dictionaries and a few corpora. However, our work with corpora led to a number of challenges. This study thus has two aims: a) presenting challenges encountered in working out semantic networks of prefixes, and b) presenting challenges related to obtaining reliable (quantitative) results from the corpora.